Cancer Incidence and Survival among Children and Adolescents

SEER PEDIATRIC MONOGRAPH

Cancer Incidence and Survival

among Children and Adolescents:

United States SEER Program

1975-1995

This publication was prepared by:

Cancer Statistics Branch

Cancer Surveillance Research Program

Division of Cancer Control and Population Sciences

National Cancer Institute

6130 Executive Blvd.

Executive Plaza North, Room 343J

Bethesda, Maryland 20892-7352

Fax: 301-496-9949

SEER web address: http://www-seer.ims.nci.nih.gov

Suggested citation for the monograph:

Ries LAG, Smith MA, Gurney JG, Linet M, Tamra T, Young JL, Bunin GR (eds).

Cancer Incidence and Survival among Children and Adolescents: United States

SEER Program 1975-1995, National Cancer Institute, SEER Program. NIH

Pub. No. 99-4649. Bethesda, MD, 1999.

Citation for a chapter should also include the chapter authors and chapter title.

This publication and additional data available on the SEER web site:

http://www-seer.ims.nci.nih.gov

All material in this report is in the public domain and may be reproduced or

copied without permission; citation as to source, however, is appreciated.

Additional Editors

Leslie Bernstein, Ph.D.

Department of Preventive Medicine, University of Southern California/Norris Cancer Center

Charles R. Key, M.D., Ph.D.

New Mexico Tumor Registry

Charles F. Lynch, M.D., Ph.D.

State Health Registry of Iowa

Joseph Simone, M.D.

Utah Cancer Registry

Jennifer Stevens, B.S.

Information Management Services, Inc.

Technical Assistance

Timothy B. Clark, B.A.

Information Management Services, Inc.

Sandra F. Kline

Information Management Services, Inc.

Maureen K. Troublefield

Information Management Services, Inc.

Cancer Incidence and Survival among Children and Adolescents:

United States SEER Program 1975-1995

Editors

Lynn A. Gloeckler Ries, M.S.

Division of Cancer Control and Population Sciences, National Cancer Institute

Malcolm A. Smith, M.D., Ph.D.

Division of Cancer Treatment and Diagnosis, National Cancer Institute

James G. Gurney, Ph.D.

Division of Epidemiology/Clinical Research, Department of Pediatrics,

University of Minnesota

Martha Linet, M.D.

Division of Cancer Epidemiology and Genetics, National Cancer Institute

Thea Tamra, M.D.

Visiting Scientist, Division of Cancer Control and Population Sciences, National

Cancer Institute

John L. Young, Jr., Dr. P.H.

Rollins School of Public Health, Emory University

Greta R. Bunin, Ph.D.

Division of Oncology, University of Pennsylvania School of Medicine and The

Children’s Hospital of Philadelphia

National Cancer Institute

SEER Pediatric Monograph

TABLE OF CONTENTS

Page

Foreword ......................................................................................................................... iii

Acknowledgments: Risk Factor Tables ..................................................................... iv

Acknowledgments: SEER ............................................................................................. v

Chapter Contributors................................................................................................... vi

Introduction ................................................................................................................... 1

Lynn A. Gloeckler Ries, Constance L. Percy, Greta R. Bunin

Chapter I Leukemia ..................................................................................................... 17

Malcolm A. Smith, Lynn A.Gloeckler Ries, James G. Gurney,

Julie A. Ross

Chapter II Lymphomas and Reticuloendothelial Neoplasms ............................... 35

Constance L. Percy, Malcolm A. Smith, Martha Linet,

Lynn A. Gloeckler Ries, Debra L. Friedman

Chapter III CNS and Miscellaneous Intracranial and Intraspinal Neoplasms ..................... 51

James G. Gurney, Malcolm A. Smith, Greta R. Bunin

Chapter IV Sympathetic Nervous System Tumors ................................................. 65

Marc T. Goodman, James G. Gurney, Malcolm A. Smith,

Andrew F. Olshan

Chapter V Retinoblastoma .......................................................................................... 73

John L. Young, Jr., Malcolm A. Smith, Steven D. Roffers,

Jonathan M. Liff, Greta R. Bunin

Chapter VI Renal Tumors ............................................................................................ 79

Leslie Bernstein, Martha Linet, Malcolm A. Smith,

Andrew F. Olshan

Chapter VII Hepatic Tumors ...................................................................................... 91

Marc Bulterys, Marc T. Goodman, Malcolm A. Smith,

Jonathan D. Buckley

Chapter VIII Malignant Bone Tumors ...................................................................... 99

James G. Gurney, Andrine R. Swensen, Marc Bulterys

Chapter IX Soft Tissue Sarcomas ............................................................................ 111

James G. Gurney, John L. Young, Jr., Steven D. Roffers,

Malcolm A. Smith, Greta R. Bunin

National Cancer Institute SEER Pediatric Monograph

TABLE OF CONTENTS

Page

Chapter X Germ Cell, Trophoblastic and Other Gonadal Neoplasms .................... 125

Leslie Bernstein, Malcolm A. Smith, Lihua Liu, Dennis Deapen,

Debra L. Friedman

Chapter XI Carcinomas and Other Malignant Epithelial Neoplasms ............... 139

Leslie Bernstein, James G. Gurney

Chapter XII Cancer Among Infants ......................................................................... 149

James G. Gurney, Malcolm A. Smith, Julie A. Ross

Chapter XIII Cancer Among Adolescents 15-19 Years Old .................................. 157

Malcolm A. Smith, James G. Gurney, Lynn A. Gloeckler Ries

Chapter XIV Childhood Cancer Mortality ............................................................. 165

Lynn A. Gloeckler Ries

Chapter XV Other NCI/ NIH Resources .................................................................. 171

International Classification of Childhood Cancer (ICCC) .................................. 175

Index .............................................................................................................................. 179

FOREWORD

iii

National Cancer Institute SEER Pediatric Monograph

Cancer among children is a substantial public concern. Each year in the United States,

approximately 12,400 children and adolescents younger than 20 years of age are diag-

nosed with cancer. Approximately 2,300 children and adolescents die of cancer each

year, which makes cancer the most common cause of disease-related mortality for chil-

dren 1-19 years of age. This monograph assembles under one cover the most detailed

information available on the incidence of childhood cancer in the United States. These

population-based data will be extremely important in furthering our understanding of

the variations in childhood cancer by histologic type and primary site and the variations

in incidence of these cancers over time. The monograph provides information about

childhood cancer incidence and mortality rates that can enhance the level of public

discourse, and it can be used in planning research that will help us to better understand

these cancers and their causes.

Unlike adult cancers that are usually tabulated by primary site, the childhood cancers

are more meaningfully grouped by histologic type and primary site based on the recently

developed International Classification of Childhood Cancer (ICCC). The monograph

details incidence for 1975-1995 and survival by ICCC group and by patient demographic

characteristics. For each of the major ICCC groups, information on known risk factors is

also presented.

The monograph emphasizes not only ICCC group but also age as important factors in

childhood cancer incidence. The cancers discussed include those occurring in children

younger than 15 years of age as well as those occurring in adolescents up to age 19 years.

Some cancers such as neuroblastoma and hepatoblastoma have highest rates among

infants and young children, while others such as Hodgkin’s disease, germ cell tumors

(e.g., testicular cancer) and bone cancers have higher rates among adolescents. It is

important that different distributions of cancer types by age be considered when re-

search programs are developed to improve outcomes for children and adolescents with

cancer.

I would like to thank and congratulate the scientists at the National Cancer Institute

(NCI) and at the various universities and institutions across the United States who

collaborated to make this monograph possible including the Epidemiology and Cancer

Control Strategy Group of the NCI-supported Children’s Cancer Group, which provided

the review of risk factors. I would also like to thank all of the individuals who make the

SEER Program a reality: staff members of the SEER population-based registries, Infor-

mation Management Services, Inc., and NCI. It is through their diligence that these data

have been collected, analyzed, and interpreted. The monograph highlights the impor-

tance of the SEER Program as a national resource. I believe that this document will

prove to be a seminal reference work on childhood cancer for scientists, policy makers

and the public. All of us look forward to the extensive use of this information and the

stimulation of scientific thought that it will engender and ultimately, the reduction of

cancer incidence and mortality in children.

Richard D. Klausner, M.D.

Director

National Cancer Institute

Acknowledgments: Risk Factor Tables

National Cancer Institute SEER Pediatric Monograph

The individuals listed below from the Epidemiology and Cancer Control Strategy Group, of

the NCI-supported Children’s Cancer Group, provided the review of risk factors for selected

cancers. Dr. Greta R. Bunin provided editorial oversight of this effort.

Jonathan D. Buckley, MBBS, Ph.D.

Greta R. Bunin, Ph.D.

Debra L. Friedman, M.D.

Seymour Grufferman, M.D.

Andrew Olshan, Ph.D.

Leslie L. Robison, Ph.D.

Julie Ross, Ph.D.

National Cancer Institute SEER Pediatric Monograph

The editors wish to thank the Principal Investigators and the staffs of the contract organi-

zations who provided the cancer incidence data for this report. These organizations, funded

through National Cancer Institute (NCI) contracts, include:

Contracting Organization Principal Investigator

Northern California Cancer Center Dr. Dee W. West

Connecticut State Department of Dr. Anthony P. Polednak

Health Mr. Daniel Savino

Emory University Dr. John L. Young. Jr.

Dr. J. William Eley

Dr. Jonathan M. Liff

University of Hawaii Dr. Laurence N. Kolonel

Dr. Marc T. Goodman

The Fred Hutchinson Cancer Dr. David B. Thomas

Research Center Dr. Beth Mueller

University of Iowa Dr. Charles F. Lynch

Dr. Charles E. Platz

Wayne State University Dr. Linda Weiss

Dr. G. Marie Swanson

University of New Mexico Dr. Charles R. Key

University of Southern California Dr. Ronald K. Ross

Dr. Dennis Deapen

Dr. Leslie Bernstein

University of Utah Dr. Joseph Simone

The production of this report would not have been possible without the efforts of the NCI

staff who ensure the quality and completeness of the SEER data: Benjamin Hankey, Limin

Clegg, April Fritz, Carol Johnson, Carol Kosary, Barry Miller, Constance Percy, Barbara

Ravas, Lynn Ries, Gopal Singh, Thea Tamra (visiting scientist) and Elliott Ware of the

Cancer Statistics Branch and Brenda Edwards of the Cancer Surveillance Research

Program.

Computer support services were provided by Information Management Services (IMS), Inc.

Acknowledgments: SEER

National Cancer Institute SEER Pediatric Monograph

Leslie Bernstein, Ph.D.

Department of Preventive Medicine, University of Southern California/Norris Cancer Center

Jonathan D. Buckley, MBBS, Ph.D.

Department of Preventive Medicine, University of Southern California (Los Angeles)

Marc Bulterys, M.D., Ph.D.

University of New Mexico, currently at Centers for Disease Control and Prevention

Greta R. Bunin, Ph.D.

Division of Oncology, University of Pennsylvania School of Medicine and The Children’s Hospital of Philadelphia

Dennis Deapen, Dr. P.H.

Department of Preventive Medicine, University of Southern California/Norris Cancer Center

Debra L. Friedman, M.D.

Division of Hematology/Oncology, Children’s Hospital and Regional Medical Center, Seattle, WA

Marc T. Goodman, Ph.D.

Cancer Research Center of Hawaii

James G. Gurney, Ph.D.

Division of Epidemiology/Clinical Research, Department of Pediatrics, University of Minnesota

Jonathan M. Liff, Ph.D.

Rollins School of Public Health, Emory University

Martha Linet, M.D.

Division of Cancer Epidemiology and Genetics, National Cancer Institute

Lihua Liu, Ph.D.

Department of Preventive Medicine, University of Southern California/Norris Cancer Center

Andrew F. Olshan, Ph.D.

Department of Epidemiology, University of North Carolina

Constance L. Percy, M.S.P.H.

Division of Cancer Control and Population Sciences, National Cancer Institute

Lynn A. Gloeckler Ries, M.S.

Division of Cancer Control and Population Sciences, National Cancer Institute

Steven D. Roffers, PA, CTR

Rollins School of Public Health, Emory University

Julie A. Ross, Ph.D.

Department of Pediatrics and Cancer Center, University of Minnesota

Malcolm A. Smith, M.D., Ph.D.

Division of Cancer Treatment and Diagnosis, National Cancer Institute

Andrine R. Swensen, M.S.

Division of Epidemiology, University of Minnesota

John L. Young, Jr., Dr. P.H.

Rollins School of Public Health, Emory University

Chapter Contributors

INTRODUCTION

National Cancer Institute SEER Pediatric Monograph

INTRODUCTION

Nearly 30 percent of the United States

(US) population is younger than 20 years of

age. Although cancer is rare among those

younger than 20 years of age, it is esti-

mated that approximately 12,400 children

younger than 20 years of age were diag-

nosed with cancer in 1998 and 2,500 died of

cancer in 1998 [1]. As a cause of death,

cancer varies in its relative importance

over the age range from newborn to age 19.

Based on data for 1995, in infants younger

than one year of age, there were fewer than

one hundred cancer deaths (representing

only 0.2% of infant deaths), making it a

minor cause of death in comparison to other

events during the perinatal period. For

children between one and nineteen, cancer

ranked fourth as a cause of death behind

unintentional injuries (12,447), homicides

(4,306), and suicides (2,227). The probabil-

ity of developing cancer prior to age 20

varies slightly by sex. A newborn male has

0.32 percent probability of developing

cancer by age 20, (i.e., a 1 in 300 chance).

Similarly a newborn female has a 0.30

percent probability of developing cancer by

age 20, (i.e., a 1 in 333 chance) [2].

Childhood cancer is not one disease

entity, but rather is a spectrum of different

malignancies. Childhood cancers vary by

type of histology, site of disease origin, race,

sex, and age. To explain some of these

variations, this monograph presents de-

tailed cancer incidence and survival data

for 1975-95, based on nearly 30,000 newly

diagnosed cancers arising in children

during this 21-year interval in the United

States (US). Cancer mortality data col-

lected for the entire US are also shown for

the same time period.

Lynn A. Gloeckler Ries, Constance L. Percy, Greta R. Bunin

MATERIALS AND METHODS (for

definitions and additional details, see

the technical appendix at end of chap-

ter):

Sources of data

The population-based data used in this

monograph for incidence and survival are

from the Surveillance, Epidemiology and

End Results (SEER) Program of the Na-

tional Cancer Institute (NCI) [2]. Informa-

tion from five states (Connecticut, Utah,

New Mexico, Iowa, and Hawaii) and five

metropolitan areas (Detroit, Michigan;

Atlanta, Georgia; Seattle-Puget Sound,

Washington; San Francisco-Oakland,

California; and Los Angeles, California)

comprising about 14% of the United States’

population are used in this monograph.

While Los Angeles did not officially become

a SEER area until 1992, the long standing

cancer registry in Los Angeles provided a

special childhood data file for this study

which included population-based cancer

incidence data back to 1975. This mono-

graph includes 29,659 cancers diagnosed

between 1975 and 1995 in persons younger

than 20 years of age who resided in the

SEER areas listed above: 19,845 cases for

those younger than 15 years of age and

9,814 cases for adolescents aged 15-19

years.

The mortality data are for the same

time period but cover all cancer deaths

among children in the total United States.

Data based on underlying cause of death

were provided by the National Center for

Health Statistics (NCHS).

INTRODUCTION

National Cancer Institute

SEER Pediatric Monograph

Table 1: Percent distribution of childhood cancers by ICCC category

and age group, all races, both sexes, SEER, 1975-95

Age

<5 5-9 10-14 15-19 <15 <20

All Sites - Number of cases 9,402 5,024 5,419 9,814 19,845 29,659

%%%%%%

All Sites 100.0 100.0 100.0 100.0 100.0 100.0

36.1 33.4 21.8 12.4 31.5 25.2

Ia - Lymphoid Leukemia 29.2 27.2 14.7 6.5 24.7 18.7

Ia - excl. Acute Lymphoid 0.2 0.3 0.2 0.1 0.2 0.2

Acute Lymphoid 29.0 27.0 14.5 6.4 24.5 18.5

Ib - Acute Leukemia 4.6 4.1 5.4 4.1 4.7 4.5

Ib - excl. Acute Myeloid 1.9 0.9 1.6 0.9 1.5 1.3

Acute Myeloid 2.8 3.2 3.8 3.2 3.2 3.2

Ic - Chronic myeloid leukemia 0.6 0.7 0.9 1.2 0.7 0.9

Id - Other specified leukemias 0.2 0.2 0.1 0.1 0.2 0.2

Ie - Unspecified leukemias 1.4 1.2 0.8 0.5 1.2 1.0

3.9 12.9 20.6 25.1 10.7 15.5

IIa - Hodgkins' disease 0.4 4.5 11.4 17.7 4.4 8.8

IIb - Non-Hodgkins' Lymphoma 2.0 5.2 6.1 6.0 4.0 4.6

IIc - Burkitt's lymphoma 0.8 2.4 1.9 0.6 1.5 1.2

IId - Miscellaneous lymphoreticular

neoplasms

0.4 0.2 0.3 0.2 0.3 0.3

IIe - Unspecified lymphomas 0.3 0.7 0.9 0.7 0.6 0.6

16.6 27.7 19.6 9.5 20.2 16.7

IIIa - Ependymoma 2.6 1.3 1.1 0.5 1.9 1.4

IIIb - Astrocytoma 6.7 14.2 11.8 6.0 10.0 8.7

IIIc - Primitive neuroectodermal tumors 4.3 6.3 3.1 1.0 4.5 3.3

IIId - Other gliomas 2.2 5.0 2.9 1.5 3.1 2.6

IIIe - Miscellaneous intracranial and

intraspinal neoplasms

0.2 0.3 0.3 0.3 0.3 0.3

IIIf - Unspecified intracranial and

intraspinal neoplasms

0.5 0.6 0.4 0.2 0.5 0.4

14.3 2.7 1.2 0.5 7.8 5.4

IVa - Neuroblastoma and

ganglioneuroblastoma

14.0 2.6 0.8 0.3 7.5 5.1

IVb - Other sympathetic nervous system

tumors

0.3 0.1 0.3 0.1 0.3 0.2

6.3 0.5 0.1 0.0 3.1 2.1

9.7 5.4 1.1 0.6 6.3 4.4

VIa - Wilms' tumor, rhabdoid and clear cell

sarcoma

9.7 5.2 0.7 0.2 6.1 4.2

VIb - Renal carcinoma 0.1 0.1 0.4 0.4 0.2 0.2

VIc - Unspecified malignant renal tumors 0.0 0.0 0.0 0.0 0.0 0.0

I(total) - Leukemia

II(total) - Lymphomas and

reticuloendothelial neoplasms

III(total) - CNS and miscellaneous

intracranial and intraspinal

neoplasms

IV(total) - Sympathetic nervous system

V(total) - Retinoblastoma

VI(total) - Renal tumours

INTRODUCTION

National Cancer Institute SEER Pediatric Monograph

Table 1 (cont’d): Percent distribution of childhood cancers by ICCC category

and age group, all races, both sexes, SEER, 1975-95

Age

<5 5-9 10-14 15-19 <15 <20

All Sites - Number of cases 9,402 5,024 5,419 9,814 19,845 29,659

%%%%%%

2.2 0.4 0.6 0.6 1.3 1.1

VIIa - Hepatoblastoma 2.1 0.2 0.1 0.0 1.0 0.7

VIIb - Hepatic carcinoma 0.1 0.3 0.5 0.5 0.3 0.3

VIIc - Unspecified malignant hepatic

tumors

0.0 0.0 0.0 0.0 0.0 0.0

0.6 4.6 11.3 7.7 4.5 5.6

VIIIa - Osteosarcoma 0.2 2.2 6.6 4.4 2.4 3.1

VIIIb - Chondrosarcoma 0.0 0.1 0.6 0.6 0.2 0.3

VIIIc - Ewing's sarcoma 0.3 2.1 3.7 2.3 1.7 1.9

VIIId - Other specified malignant bone

tumors

0.1 0.1 0.3 0.3 0.2 0.2

VIIIe - Unspecified malignant bone tumors 0.0 0.1 0.1 0.1 0.1 0.1

5.6 7.5 9.1 8.0 7.0 7.4

IXa - Rhabdomyosarcoma and embryonal

sarcoma

3.4 4.2 2.8 1.9 3.4 2.9

IXb - Fibrosarcoma, neurofibrosarcoma and

other fibromatous neoplasms

1.0 1.4 3.1 3.1 1.7 2.1

IXc - Kaposi's sarcoma 0.0 0.1 0.0 0.1 0.0 0.1

IXd - Other specifed soft-tissue sarcomas 0.7 1.2 2.2 2.1 1.3 1.5

IXe - Unspecifed soft-tissue sarcomas 0.4 0.7 1.0 0.9 0.6 0.7

3.3 2.0 5.3 13.9 3.5 7.0

Xa - Intracranial and intraspinal germ-cell

tumors

0.2 0.8 1.3 0.9 0.7 0.7

Xb - Other and unspecified non-gonadal

germ-cell tumors

1.7 0.1 0.5 1.4 1.0 1.1

Xc - Gonadal germ-cell tumors 1.4 1.1 3.0 9.4 1.7 4.2

Xd - Gonadal carcinomas 0.0 0.0 0.4 1.9 0.1 0.7

Xe - Other and unspecified malignant

gonadal tumors

0.0 0.1 0.1 0.3 0.1 0.1

0.9 2.5 8.9 20.9 3.5 9.2

XIa - Adrenocortical carcinoma 0.2 0.1 0.1 0.1 0.1 0.1

XIb - Thyroid carcinoma 0.1 1.0 3.5 7.4 1.2 3.3

XIc - Nasopharyngeal carcinoma 0.0 0.1 0.7 0.8 0.2 0.4

XId - Malignant melanoma 0.4 0.7 2.0 6.8 0.9 2.9

XIe - Skin carcinoma 0.0 0.0 0.1 0.1 0.0 0.0

XIf - Other and unspecified carcinomas 0.2 0.7 2.5 5.7 1.0 2.5

0.5 0.3 0.6 0.8 0.5 0.6

XIIa - Other specified malignant tumors 0.1 0.1 0.1 0.3 0.1 0.1

XIIb - Other unspecified malignant tumors 0.4 0.3 0.5 0.5 0.4 0.4

VIII(total) - Malignant bone tumors

IX(total) - Soft-tissue sarcomas

X(total) - Germ-cell, trophoblastic and

other gonadal tumors

XI(total) - Carcinomas and other

malignant epithelial

neoplasms

XII(total) - Other and unspecified

malignant neoplasms

VII(total) - Hepatic tumors

INTRODUCTION

National Cancer Institute

SEER Pediatric Monograph

In order to calculate rates, population

estimates were obtained from the Bureau

of the Census. In 1990 there were

7,179,865 children residing in the SEER

areas younger than 15 years of age and

9,436,324 younger than 20 years of age. In

the 1990 census, there were about 72

million children younger than 20 years of

age in the whole United States. Twenty-

two percent of the US population is younger

than 15 years of age and an additional 7%

are 15-19 years of age. Annual population

estimates include estimates by 5-year age

groups (<5,5-9,10-14,15-19). Enumeration

of the population at risk by single years of

age was available only for the census years

1980 and 1990. The US Bureau of the

Census provides intercensal population

estimates by 5-year age groups, but not by

single years of age. Therefore, the popula-

tion estimates for 1980 were used in rate

calculations for cases diagnosed from 1976-

84 and the 1990 estimates were used for

cases diagnosed from 1986-94. Whenever

rates by single year of age are shown, the

rates are centered around a decennial

census year, namely, 1976-84 and 1986-94

or the two sets of years combined.

Calculation of rates (see technical appendix)

The incidence and mortality rates are

the annual rates per million person years.

For simplicity, these are labeled as rates per

million. Rates representing more than 5-

years of age are age-adjusted to the 1970

US standard million population. Survival

rates are expressed as percents.

Classification of site and histologic type

The SEER program classifies all cases

by cancer site and histologic type using the

International Classification of Diseases for

Oncology, Second Edition (ICD-O-2) [3]. In

contrast to most cancer groupings, which

are usually categorized by the site of the

cancer, the pediatric classification is deter-

mined mostly by histologic type. The SEER

data have been grouped according to the

International Classification of Childhood

Cancers (ICCC) specifications [4] with a

couple of exceptions for brain cancer.

Please refer to Table 1 for the distribution

by ICCC groupings and age group.

Histologic confirmation

In the SEER program most of the

pediatric cancers (95%) are histologically

confirmed. This is important because most

childhood cancer classifications are based

on histologic types: leukemia, lymphoma,

retinoblastoma, neuroblastoma, etc. The

percentage of histologically confirmed cases,

however, does vary by ICCC category

ranging from a low of 90 percent for the

central nervous system (CNS) (ICCC group

III) to a high of 99 percent for leukemia

(ICCC group I).

OVERVIEW OF CHILDHOOD CANCER

PATTERNS

All sites combined

While grouping all cancer sites to-

gether may be helpful to understanding the

overall cancer burden in young Americans,

it masks the contributions of each primary

site/histology. Therefore, most of the em-

phasis of this monograph is on individual

primary site or histologic groupings; a

separate chapter is shown for each of the

ICCC groupings except group XII which

has few cases.

Overall trends

While the incidence rates for some

forms of childhood cancer have increased

since the mid-1970s, death rates have

declined dramatically for most childhood

cancers and survival rates have improved

markedly since the 1970’s. Each year

approximately 150 children out of every

million children younger than 20 years of

age will be diagnosed with cancer. The

INTRODUCTION

National Cancer Institute SEER Pediatric Monograph

overall cancer incidence rate increased from

the mid-1970’s, but rates in the past decade

have been fairly stable (Figure 1). During

the last time period, 1990-95, there is an

indication of a leveling off or slight decline

in the overall incidence rates for each of the

5-year age groups (data not shown). The

overall childhood cancer mortality rates

have consistently declined throughout the

1975-95 time period (Figure 1). Note that

the data are plotted at the mid-year point

throughout this monograph.

Sex

For all sites combined, cancer incidence

was generally higher for males than fe-

males during the 21-year period (Figure 2).

Yet again, an all-sites-combined-rate masks

the sites/histologies for which there is a

female predominance. For some sites/

histologies, there are other factors such as

age where there are differences by sex. For

example, males have somewhat higher

rates of Hodgkin’s disease for children

younger than 15 years of age, but females

have higher rates for adolescents, 15-19

years of age.

Age (5-year age groups)

The average age-specific incidence

rates for each of the four calendar periods

of observation show similar and much

higher cancer rates for the youngest

(younger than 5 years of age) and oldest

(15-19 years of age) age groups than the

two intermediary age groups (Figure 3).

Even though those aged 15-19 years and

those younger than 5 years of age have

similar incidence rates, they have different

mixtures of sites and histologies. The

cancer incidence rates for 5 to 9 year olds

are similar to those seen among 10-14 year

olds.

Age and ICCC group

Fifty-seven percent of the cancers

found among children younger than 20

Figure 2: Trends in age-adjusted* incidence rates

for all childhood cancers by sex, age <20

all races combined, SEER, 1975-95

)

))

1975 1980 1985 1990 1995

Year of diagnosis

100

150

200

Average annual rate per million

Male

Female

)

*Adjusted to the 1970 US standard population

1975 1980 1985 1990 1995

Year of diagnosis

100

120

140

160

180

Average annual rate per million

Incidence

Mortality

*Adjusted to the 1970 US standard population

Figure 1: Trends in age-adjusted* SEER incidence &

U.S. mortality rates for all childhood cancers

age<20, all races, both sexes, 1975-95

INTRODUCTION

National Cancer Institute

SEER Pediatric Monograph

years of age were leukemia, malignant

tumors of the central nervous system (CNS)

or lymphoma. The relative percentage,

however, varied by age group (Table 1).

Leukemia was the most common diagnosis

for those younger than 5, 5-9, and 10-14

years of age but the relative proportion of it

decreased as age increased, from 36 percent

for those younger than 5 years of age to

only 12 percent for adolescents 15-19 years

of age. For 15-19 year olds, lymphomas

were the most common diagnosis, compris-

ing one-fourth of the cases. The second

most common type of cancer was malignant

tumors of the central nervous system for

younger than 5 and 5-9 years of age, and

lymphoma for 10-14 and leukemia for 15-

19 year olds (Table 1).

Figure 4 shows the numbers of cases

used in this study by ICCC group and age.

Leukemia (group I) had the largest number

of cases. Note that these numbers are over

the period 1975 to 1995 for the SEER areas

and do not represent the total number of

childhood cancers in the US in one year.

These numbers indicate the reliability in

the incidence and survival rates, i.e. large

numbers imply stable rates and small

numbers imply unstable rates. Even

though ICCC groups I-III have most of the

cases, there are differences by age group:

group I has more 1-4 year olds, group II has

more 15-19 year olds and group III has

nearly equal numbers for each age group.

There are less than 1,000 cases each in

groups V, VII and XII. Groups VIII-XI tend

to have fewer children younger than 10

years of age compared to 10-19 years of

age.

Incidence by ICCC group

Figure 5 shows the incidence rates per

million children for each of the ICCC

groups. The highest rates are for groups I

(leukemia), II (lymphoma), and III (CNS).

Figure 3: Trends in age-specific incidence rates for

all childhood cancers by age, all races

both sexes, SEER, 1975-95

(

)

1975-79 1980-84 1985-89 1990-95

Year of diagnosis

100

150

200

250

Average annual rate per million

5-9

10-14

15-19

)

(

Figure 4: Number of cases of all childhood cancers

by ICCC and age group, all races

both sexes, SEER, 1975-95

Leukemia - I

Lymphoma - II

Brain/CNS - III

Sympathetic Nerv. - IV

Retinoblastoma - V

Renal - VI

Hepatic - VII

Bone - VIII

Soft tissue - IX

Germ cell - X

Carcinomas - XI

Other - XII

ICCC Group

02468

Number of cases (in thousands)

1-4

5-9

10-14

15-19

INTRODUCTION

National Cancer Institute SEER Pediatric Monograph

While the ICCC major groupings indicate

which broad groups of sites/histologies are

important, the sub-groups under each are

necessary to really delineate which histolo-

gies are driving these rates. More detailed

information on the ICCC groups and sub-

groups are contained in other chapters.

Race/ethnicity

For many adult cancers, blacks have

higher incidence rates than whites. For

children, however, black children had lower

incidence rates in 1990-95 than white

children overall and for many of the specific

sites (Figure 6). The time period, 1990-95,

was used for racial/ethnic comparisons

because it was the only time period except

for the decennial census years (1980 and

1990) for which the Census Bureau pro-

vided population estimates for racial groups

other than white and black. The largest

racial difference was for leukemia (ICCC I)

where the rate for whites (41.6 per million)

was much higher than that for blacks (25.8

per million). Cancer incidence rates for

Hispanic children and Asian/Pacific Is-

lander children were intermediate to those

for whites and blacks. The rates for Asian/

Pacific Islanders were similar to whites for

leukemia but lower than whites for CNS

and lymphomas. The incidence rates for

American Indians were much lower than

any other group.

Single year of age

For all sites combined, incidence

varied by age with the highest rates in

infants. The incidence rates declined as age

increased until age 9 and then the inci-

dence rates increased as age increased after

age 9. The pattern, however, varied widely

by ICCC group and single year of age. For

example, high rates were seen among the

very young for retinoblastoma (ICCC group

V) and among adolescents for lymphoma

Figure 5: Age-adjusted* incidence rates for

childhood cancer by ICCC group, age <20, all races

both sexes, SEER, 1975-95

Leukemia - I

Lymphoma - II

Brain/CNS - III

Sympathetic Nerv. - IV

Retinoblastoma - V

Renal - VI

Hepatic - VII

Bone - VIII

Soft tissue - IX

Germ cell - X

Carcinomas - XI

Other - XII

ICCC group

0 1020304050

Average annual rate per million

*Adjusted to the 1970 US standard population

Figure 6: Age-adjusted* incidence rates for

childhood cancer by ICCC group and race/ethnicity

age <20, both sexes, SEER, 1990-95

Am. Indian = American Indian/Native American; API = Asian/Pacific Islander

Hispanic = Hispanic of any race and overlaps other categories

*Adjusted to the 1970 US standard population

41.6

25.8

24.8

41.2

48.5

24.7

18.7

14.9

19.5

29.1

10.9

19.9

21.8

66.3

55.1

39.9

60.8

55.8

White Black Am. Indian API Hispanic

Race/ethnicity

100

125

150

175

Average annual rate per million

I - Leukemia

II - Lymphoma

III - CNS

Other

161.7

124.6

79.6

136.8

145.6

INTRODUCTION

National Cancer Institute

SEER Pediatric Monograph

Figure 8

(

((

(((((

(

((

0 1 2 3 4 5 6 7 8 9 1011121314151617181920

Age (in years) at diagnosis

Average annual rate per million

Neuroblastoma (IVa)

Retinoblastoma (V)

Wilms' (VIa)

(

Figure 9

(

((

(

01234567891011121314151617181920

Age (in years) at diagnosis

Average annual rate per million

Hepatic (VII)

Bone (VIII)

Soft tissue (IX)

Germ cell (X)

(

Age-specific incidence rates for childhood cancer

by ICCC group, all races, both sexes, SEER 1986-94

Figure 7

(

0 1 2 3 4 5 6 7 8 9 1011121314151617181920

Age (in years) at diagnosis

100

Average annual rate per million

Ac. Lymph. Leuk (Ia)

Ac. Myeloid Leuk (Ib)

Lymphoma (II)

Brain/CNS (III)

(

INTRODUCTION

National Cancer Institute SEER Pediatric Monograph

(ICCC group II) and germ cell (ICCC group

X) for 1986-94 (Figures 7-9). Among those

older than 9 years of age, there were very

low incidence rates for neuroblastoma

(ICCC group IVa), retinoblastoma (ICCC

group V), Wilms’ tumor (ICCC group VIa),

and hepatic tumors (ICCC group VII).

SURVIVAL

The cancer survival rate for children

has greatly improved over time. Even since

the mid-1970s there have been large im-

provements in short term and long term

survival (Figure 10). There were improve-

ments in survival for many forms of child-

hood cancer (Figure 11). The principal

reason for the gain for total childhood

cancer is due to the improvement in the

survival of leukemia, especially acute

lymphocytic leukemia, which includes

about a third of the pediatric cases. This is

due primarily to improvements resulting

from more efficacious chemotherapy agents.

RISK FACTORS

Throughout this monograph, there are

discussions of potential causes and risk

factors for individual childhood cancers.

The discussion below provides background

for considering the strength of the epide-

miological evidence available for each risk

factor. Since the evidence on risk factors

varies, each risk factor table has the factors

characterized by one of the following:

• Known risk factors: Most epidemi-

ologists consider these characteris-

tics or exposures to be ‘causes’ of the

particular cancer. The scientific

evidence meets all or most of the

criteria described earlier. However,

many individuals in the population

may have the characteristic or

1975 1980 1985 1990 1995

Year of diagnosis

100

Percent surviving

Survival rate:

1year from dx

3 yrs from dx

5 yrs from dx

10 yrs from dx

Figure 10: Trends in relative survival rates for all

childhood cancers, age <20, all races, both sexes

SEER (9 areas), 1975-94

1975-78 1979-82 1983-86 1987-90 1991-94

Year of diagnosis

100

Percent surviving

Leukemia

Lymphoma

Brain/CNS

Sympathetic Nerv.

Retinoblastoma

1975-78 1979-82 1983-86 1987-90 1991-94

Year of diagnosis

100

Percent surviving

Renal Hepatic Bone

Soft tissue Germ cell Carcinomas

& ! &

' $ %

Figure 11: Trends in 5-year relative cancer

survival rates by ICCC group, age <20

all races, both sexes, SEER (9 areas), 1975-94

INTRODUCTION

National Cancer Institute

SEER Pediatric Monograph

exposure and not develop cancer

because there are other contributory

factors.

• Suggestive but not conclusive evi-

dence: The scientific evidence link-

ing these characteristics or expo-

sures to the particular cancer meets

some but not all of the criteria

described earlier.

• Conflicting evidence: Some studies

show the putative risk factor to be

associated with higher risk but

others show no increased risk or

lower risk.

• Limited evidence: Very few studies

have investigated the putative risk

factor. The existing studies may

have investigated the exposure in a

superficial manner or methodologic

issues may make the results difficult

to interpret.

Finding causes of any disease is usu-

ally a long, slow process. Epidemiologists

find clues in one study that they follow-up

in later studies. Only some of the clues are

useful. Current studies are designed to

help us learn whether or not previously

identified clues are likely to lead us to the

causes of a particular cancer. No one study

is likely to prove that a particular exposure

definitely causes a particular cancer. No

single study nor even a large number of

epidemiologic studies will enable a parent

to know why his or her child developed

cancer. However, each well designed and

well executed study will bring us closer to

understanding the causes of these cancers

within populations of children.

Multifactorial etiology

We also do not expect that all children

with a particular cancer developed it for the

same reason. In other words, we do not

think that one exposure, behavior or ge-

netic trait explains all or even a majority of

instances of a particular cancer. Rather, we

expect that a number of exposures and

characteristics of children each contribute

to a proportion of instances of a particular

cancer.

No one factor determines whether an

individual will develop cancer, even if a

specific exposure explains a high proportion

of the occurrence of a specific cancer.

Rather, it is the interaction of many factors

that produces cancer. This concept is

referred to as the multiple causation or

multifactorial etiology. The factors involved

may be genetic, constitutional or behavioral

characteristics of the individual or factors

external to the individual. Among the

many types of factors that might play a role

are genetic, immune, dietary, occupational,

hormonal, viral, socioeconomic, lifestyle,

and other characteristics of the individual

and the biologic, social, or physical environ-

ment.

The concept of multiple causation has

direct implications for the interpretation of

research on the causes of cancer. Suppose

that combinations of laboratory and epide-

miologic studies have shown that exposure

to chemical X causes leukemia. We know

that other factors must play a role since not

all children who were exposed to chemical

X developed leukemia. Thus, there must be

other factors that determine which of the

children exposed to chemical X will develop

leukemia.

Associations versus causes

Frequently, newspapers and television

report that some chemical, dietary habit, or

household product is purported to increase

the risk of cancer. These news stories tell

us about associations between an exposure

and a cancer. In other words, more of the

people who developed cancer than those

without cancer had the exposure. However,

an association between an exposure and

INTRODUCTION

National Cancer Institute SEER Pediatric Monograph

cancer does not necessarily mean that the

exposure causes cancer.

As an example, suppose a case-control

study (see Technical Appendix) finds that

more of the mothers of children with acute

lymphoblastic leukemia (ALL) than moth-

ers of controls used medication Y during

pregnancy. It is possible that medication Y

causes ALL, but there are also other expla-

nations. It may be that mothers of children

with ALL were more accurate in their

reporting of medication use than the con-

trol mothers. Since mothers are asked in

these studies to recall their use of medica-

tion and other substances during a preg-

nancy 5 or 10 years in the past, it is certain

that their reporting is not completely

accurate. Mothers of children with cancer

have probably thought about their expo-

sures during the relevant pregnancies more

intently than control mothers in their

search for an explanation of their children’s

illness. Case mothers may remember short

episodes of medication use whereas control

mothers may have forgotten them. Differ-

ences in the level of recall between mothers

of cases and mothers of controls may be

real or may reflect less accurate recall of

either group of mothers. This type of

differential recall may lead to erroneous

results for either group; such differential

recall would lead to inaccurate or biased

results, a problem designated as recall bias.

A recall bias would lead to an association or

disassociation between the medication and

cancer which would not be causal but

spurious or false. Another explanation of

an association between the medication and

cancer is that medication Y is used to treat

a medical condition and that the condition

rather than the medication confers the risk.

Epidemiologists would say that the condi-

tion is a confounder of the observed associa-

tion between the medication and cancer.

How do epidemiologists decide whether

an association between an exposure and a

disease is one of cause and effect? The

methods and processes of epidemiology and

their limitations make it nearly impossible

for a single study to prove that an exposure

causes a disease. There must be a number

of studies that epidemiologists can evaluate

using a set of criteria. The criteria are

described briefly but the order in which

they are described does not signify relative

importance.

1. Other possible explanations of the

observed association must be ruled

out, such as the medical condition

rather than the medication. In

another example, if one investigates

an association between eating hot

dogs and developing a specific

cancer, one must determine whether

high dietary fat intake or infrequent

fruit eating explains the association

and rule out these factors before

concluding hot dog consumption is

related to risk.

2. Epidemiologists consider the

strength of the association, that is,

the relative risk (see Technical

Appendix). An exposure associated

with a ten-fold increase in risk is

more likely to be a true cause than

an exposure associated with a two-

fold increase.

3. The consistency of an association is

considered. An association observed

in many different studies in differ-

ent populations using different

study methods is likely to be true.

4. The observation of a dose-response

relationship between the exposure

and the disease increases confidence

that the exposure is really related to

the disease. In a dose-response

relationship, the risk of disease

increases or decreases as the

amount of the exposure increases or

decreases. For example, the rela-

tionship between cigarette smoking

INTRODUCTION

National Cancer Institute

SEER Pediatric Monograph

and lung cancer shows a dose-

response in that heavy smokers

have a higher risk than light smok-

ers.

5. The association must be temporally

correct meaning that we must be

sure that the exposure actually

preceded development of the disease.

For example, a study might report

that barbiturate use increased the

risk of brain tumors. However,

barbiturates are used to control

seizures, which are often an early

symptom of a brain tumor. There-

fore, it may not be clear if barbitu-

rate use actually preceded the

development of the brain tumor or if

barbiturates were used to treat an

early symptom before the brain

tumor was diagnosed.

6. A biologically plausible association

is more likely to be true than one

without other supporting evidence.

For example, we have more confi-

dence that chemical X causes brain

tumors in humans if it is known to

cause brain tumors in animals.

All or most of these six criteria must be

met before an association between a dis-

ease and an exposure is considered a causal

association.

Structure of monograph

This monograph consists of a chapter

for each of the principal types of pediatric

cancers as designated by the ICCC. The

ICCC designated group is also used as the

chapter number except for group XII which

is less than 6% of the total and is not

shown in a separate chapter. Each of these

chapters discusses incidence, mortality, and

survival rates of the patients, as well as

trends in these measures by demographic

characteristics. Risk factors are also de-

scribed. The estimated number of cases in

the US for 1998 is given in each chapter.

These numbers are based on the American

Cancer Society’s overall cancer estimate of

12,400 [1] cases and on the SEER site

distribution for 1990-95.

In addition, there are separate chap-

ters on children younger than 1 year of age,

adolescents, and incidence vs. mortality

trends. The monograph is also available

from the SEER home page under publica-

tions (http://www-seer.ims.nci.nih.gov). There

is a technical appendix at the end of this

chapter which defines terms used in the

Introduction and in other chapters; it also

provides more details on methods and data

sources.

TECHNICAL APPENDIX

Age-adjusted rate: An age-adjusted rate is a

weighted average of the age-specific cancer inci-

dence (or mortality) rates, where the weights are the

proportions of persons in the corresponding age

groups of a standard population. The potential

confounding effect of age is reduced when comparing

age-adjusted rates computed using the same

standard population. For this report, the 1970

United States standard million is used as the

standard in computing all age-adjusted rates. Since

rates of childhood cancer vary widely by 5-year age

group, age-adjustment was used for any age group

representing more than one 5-year grouping. Age-

adjustment was performed by 5-year age group and

weighted by the 1970 US standard million popula-

tion.

Age-specific rates: Age-specific rates are usually

presented as a rate per million. The numerator of

the rate is the number of cancer cases found in a

particular 5-year age group in a defined population

divided by the number of individuals in the same 5-

year age group in that population. In this publica-

tion, there are some rates by single year of age for

time periods around the Census. Population

estimates by single year of age, race, sex, and

geographic region are not generally available for

intercensal years. The rates by single year of age

are plotted at half years. For example, the rate for

children age 1 year is plotted at 1.5 years since they

are an average 1 1/2 years of age.

Case-control study: A case-control study is an

epidemiologic study in which a group of individuals

with a disease, the cases, are compared to a group of

individuals without the disease, the controls.

INTRODUCTION

National Cancer Institute SEER Pediatric Monograph

Exposures or characteristics that are more common

in the cases than in the controls may be causes of

the disease. Exposures or characteristics that are

equally common in the cases and controls cannot be

causes of the disease. Almost all studies of child-

hood cancer are case-control studies because this

type of study is very useful in studying relatively

uncommon diseases.

Cohort study: A cohort study is an epidemiologic

study in which the incidence of disease is compared

between a group of individuals with an exposure or

characteristic and a group without that exposure or

characteristic. For example, smokers and nonsmok-

ers are followed and the incidence of heart disease is

compared in the two groups. Or, the incidence of

breast cancer is compared in women with and

without a BRCA1 gene mutation. This type of study

is rarely feasible in investigating the etiology of

childhood cancer. Since childhood cancer is rare,

especially if we consider that each cancer should be

studied separately, huge numbers of children (a few

hundred thousand) would have to be followed to

determine which children developed cancer.

EAPC (Estimated Annual Percent Change): The

Estimated Annual Percent Change (EAPC) was

calculated by fitting a regression line to the natural

logarithm of the rates (r) using calendar year as a

regressor variable, i.e. y = mx + b where y = Ln r

and x = calendar year. The EAPC = 100*(e

- 1).

Testing the hypothesis that the Annual Percent

Change is equal to zero is equivalent to testing the

hypothesis that the slope of the line in the above

equation is equal to zero. The latter hypothesis is

tested using the t distribution of m/SE

with the

number of degrees of freedom equal to the number

of calendar years minus two. The standard error of

m, i.e. SE

, is obtained from the fit of the regression

[5]. This calculation assumes that the rates in-

creased/decreased at a constant rate over the entire

calendar year interval. The validity of this assump-

tion has not been assessed. In those few instances

where at least one of the rates was equal to zero, the

linear regression was not calculated. The differ-

ences between incidence and mortality trends for

the time period 1975-79 versus those for the most

recent five-year period are tested for statistical

significance using a t statistic with six degrees of

freedom defined as the difference in the regression

coefficients divided by the standard error of the

difference [5].

Follow-up: SEER areas attempt to follow-up all

cases till death. Although the overall proportion of

cancer patients of all ages who are lost to follow-up

is only about 5%, for pediatric cases (age 0-19) it is

much larger - about 14%. Since survival rates are

relatively high, follow-up can be difficult, especially

as the child gets older. When children leave their

parents’ home, they change addresses and, espe-

cially for females, they may change last names.

ICCC classification: At the time the World Health

Organization’s (WHO) International Agency for

Research on Cancer (IARC) published their first

monograph on Childhood Cancer [6] in 1988, Dr. R.

Marsden published an annex giving a classification

scheme for childhood cancer that consisted of 12

groups based chiefly on histologic type. The classifi-

cation by Marsden has been modified and is now

called the International Classification of Childhood

Cancers [4].

Incidence rate: The cancer incidence rate is the

number of new cancers of a specific site/type

occurring in a specified population during a year,

expressed as the number of cancers per one million

people. It should be noted that the numerator of the

rate can include multiple primary cancers occurring

in one individual. This rate can be computed for

each type of cancer as well as for all cancers com-

bined. Except for five-year age-specific rates, all

incidence rates are age-adjusted to the 1970 US

standard million population. Rates are for invasive

cancer only, unless otherwise specified.

Mortality data: The mortality data are from public

use files provided by the National Center for Health

Statistics (NCHS) and cover all deaths in the

United States. All mortality rates were based on

the underlying cause of death. The rates presented

for 1975-1978 were coded to the International

Classification of Diseases - 8

revision and for 1979

to 1995 to the ICD 9

revision [7]. Unfortunately

mortality of all specific groups of the ICCC pediatric

classification are not available from US mortality

files due to the fact that the codes used for coding

death certificates do not include such morphologic

types as neuroblastomas and retinoblastomas.

Certain groups can be identified as specific entities

on death certificates: Leukemias, Lymphomas,

Bones, Brain and other CNS tumors, and Hodgkin’s

and Non-Hodgkin’s lymphoma. However, such types

of cancer as Retinoblastomas, Germ cell tumors,

Wilms’ tumor, and certain carcinomas can not be

identified on death certificates. Even though

neuroblastomas are not coded separately, they were

coded to different groups in the ICD-8 and ICD-9.

For these analyses to make the data comparable

over time, deaths coded to sympathetic nervous

system in the 8th revision were combined with

adrenal in the 9th revision.

Mortality rate: The cancer mortality rate is the

number of deaths with cancer given as the underly-

ing cause of death occurring in a specified popula-

tion during a year, expressed as the number of

deaths due to cancer per one million people. This

rate can be computed for each type of cancer as well

INTRODUCTION

National Cancer Institute

SEER Pediatric Monograph

as for all cancers combined. Except for age-specific

rates, all mortality rates are age-adjusted to the

1970 US standard million population.

Population data: Population estimates are obtained

each year from the US Bureau of the Census at the

county level by five-year age group (0-4, 5-9,..., 85

and over), sex, and race (including white and black).

SEER areas make county estimates for each state

available on the SEER areas Home Page

(http://www-seer.ims.nci.nih.gov) for race (whites,

blacks, non-white), 5-year age group, sex, and year of

diagnosis (each year 1973 to 1995). Additional

estimates can be obtained from the US Census

Bureau Home Page.

US Bureau of the Census (BOC) population

estimates for Hawaii were altered according to

independent estimates developed from sample

survey data collected by the Health Surveillance

Program (HSP) of the Hawaii Department of

Health. For Hawaii, the all races and black popula-

tions are the same as those sent by the BOC.

Proportions of the population by different racial

groups from the HSP were used to generate esti-

mates for whites, etc. Since the HSP survey was for

all of Hawaii and not by county, population esti-

mates were not broken down by county. The white

population estimates for Hawaii provided by the

BOC are generally larger than those generated by

the HSP. Since whites in Hawaii account for less

than two percent of the total white population

represented by the SEER reporting areas, white

incidence rates for the entire SEER Program are not

noticeably affected. Procedures for calculating rates

by race for Hawaii are currently under review.

Primary site/histology coding: Originally data for

site and histologic type were coded by the Interna-

tional Classification of Diseases for Oncology (ICD-

O) [8], but in 1990, ICD-O was revised and repub-

lished as the International Classification of Diseases

for Oncology, 2nd Edition (ICD-O-2) [7]. SEER

areas began using ICD-O-2 for cases diagnosed in

1992 and machine converted all previous data to

ICD-O-2. Most data for Non-Hodgkin’s Lymphoma

(NHL) can be classified by the Working Formulation

(WF) based on a conversion from ICD-O-2.

Relative risk: Whether or not an exposure increases

the risk of cancer and how much it does is expressed

in a measure called relative risk. The relative risk

is the risk of disease in those with the exposure

divided by the risk of disease in those without the

exposure.

• Relative risk less than 1.0 - the exposure

appears to lower the risk of the disease. For

example, a relative risk of 0.75 for taking

vitamin X supplements indicates that those

who took vitamin X had a risk that was 75%

of that for individuals who did not take

vitamin X. Or, taking vitamin X lowered one’s

risk by 25%.

• Relative risk of 1.0 - the exposure does not

affect the risk of the disease; the risk is the

same in those with the exposure as in those

without the exposure.

• Relative risk greater than 1.0 - the exposure

appears to increase the risk of the disease.

For example, a relative risk of 3 for taking

medication Y indicates that those taking the

medication had a risk that was three times

that of those not taking the medication.

Relative survival rate: The relative survival rate is

calculated using a procedure described by Ederer,

Axtell, and Cutler [9] whereby the observed survival

rate is adjusted for expected mortality. The relative

survival rate represents the likelihood that a

patient will not die from causes associated specifi-

cally with their cancer at some specified time after

diagnosis. It is always larger than the observed

survival rate for the same group of patients.

Risk factor: A risk factor is a characteristic or

exposure that increases the risk of disease. A risk

factor might be exposure to high levels of radon,

having a diet low in vitamin A, having a family

history of colon cancer, or having a high cholesterol

level.

SEER Program: This program started in 1973, as

an outgrowth of the NCI’s Third National Cancer

Survey. NCI contracts out with various medically

oriented non-profit organizations, local city or state

Health Departments or Universities for collection of

these data. Contracts for collecting this data are

with the entire states of Connecticut, Iowa, New

Mexico, Utah, and Hawaii and with the metropoli-

tan areas of Los Angeles, California; Detroit,

Michigan; San Francisco-Oakland and San Jose-

Monterey, California; Seattle-Puget Sound, Wash-

ington; and Atlanta, Georgia. These organizations

collect data on all cancers except basal and squa-

mous cell skin cancers. Although data are collected

on all people having cancer, the material for this

study used children from birth through age 19 years.

Only residents of the areas designated above are

included so that the base populations can be

properly determined.

INTRODUCTION

National Cancer Institute SEER Pediatric Monograph

Reference List

1. Wingo P. American Cancer Society. Personal

communication based on unpublished data

from Cancer facts and figures, 1998. Atlanta,

1998.

2. Ries LAG, Kosary CL, Hankey BF, Miller BA,

Clegg L, Edwards BK (eds). SEER Cancer

Statistics Review 1973-1995, National Cancer

Institute, http://www.seer.ims.nci.nih.gov, 1998.

3. Percy C, Van Holten V, and Muir C, Eds.

International classification of diseases for

oncology, Second Ed.,World Health Organiza-

tion, Geneva, 1990.

4. Kramarova E and Stiller CA, The International

Classification of Childhood Cancer, Int J

Cancer. 68:759-765, 1996.

5. Kleinbaum DG, Kupper LL, Muller KE.

Applied Regression Analysis and Other

Multivariable Methods. North Scituate,

Massachusetts: Duxbury Press, 1988: 266-268.

6. Parkin DM, Stiller CA, Draper GJ, Bieber CA,

Terracini B, and Young JL (eds). International

incidence of childhood cancer, WHO, IARC

Scientific Publication No.87, Lyon, 1988.

7. World Health Organization, International

Classification of Diseases, 1975 Revision, vols.

1 and 2, Geneva, 1977.

8. World Health Organization, International

Classification of Diseasess for Oncology, First

Edition, Geneva, 1976.

9. Ederer F, Axtell LM, Cutler SJ. The relative

survival rate: A statistical methodology. Natl

Cancer Inst Monogr 1961; 6:101-121.

National Cancer Institute

SEER Pediatric Monograph

ICCC ILEUKEMIA

National Cancer Institute

SEER Pediatric Monograph

Malcolm A. Smith, Lynn A. Gloeckler Ries, James G. Gurney, Julie A. Ross

HIGHLIGHTS

Incidence

♦ For the years from 1990-95, the leukemias represented 31% of all cancer cases

occurring among children younger than 15 years of age and 25% of cancer cases

occurring among those younger than 20 years of age. In the US there are approxi-

mately 3,250 children diagnosed each year with leukemia and 2,400 with acute

lymphoblastic leukemia (ALL).

♦ The relative contribution of leukemia to the total childhood cancer burden varies

markedly with age, being 17% in the first year of life, increasing to 46% for 2 and 3

year olds, and then decreasing to only 9% for 19 year olds (Figure I.1).

♦ The two major types of leukemia were ALL comprising nearly three-fourths and

acute non-lymphocytic comprising 19%.

♦ There was a sharp peak in ALL incidence among 2-3 year olds (> 80 per million)

which decreases to a rate of 20 per million for 8-10 year olds. The incidence of ALL

among 2-3 year olds is approximately 4-fold greater than that for infants and is

nearly 10-fold greater than that for 19 year olds (Figure I.2a).

♦ Leukemia rates are substantially higher for white children than for black children,

with rates of 45.6 versus 27.8 per million for the period from 1986-95 for children

0-14 years old (Table I.4). This difference between white and black children is

most apparent when examining rates of leukemia by single year of age (Figure

I.3), with a nearly 3-fold higher incidence at 2-3 years of age for white children

compared to black children.

♦ The incidence of leukemia among children younger than 15 years of age has

shown a moderate increase in the past 20 years (Figure I.4) with the trend prima-

rily reflecting an increase in ALL incidence during this period. The rates of leuke-

mias other than ALL did not appear to increase from 1977 to 1995 (Figure I.5)

Survival

♦ Survival for children with ALL has markedly improved since the early 1970s, and

overall survival for all children with ALL is now approximately 80% (Figure I.8).

A number of improvements in treatment during this period have undoubtedly

contributed to the improved survival.

♦ Survival for children with ALL is very dependent upon age at diagnosis, with the

most favorable outcome observed for children older than 1 year of age and younger

than 10 years of age.

Risk factors

♦ With the exception of prenatal exposure to x-rays and specific genetic syndromes,

little is known about the causes of childhood ALL (Table I.5).

♦ Different risk factors are emerging for childhood AML that distinguish the disease

from ALL, and this may provide avenues for future epidemiological studies (Table

I.6).

ICCC I

National Cancer Institute

SEER Pediatric Monograph

LEUKEMIA

INTRODUCTION

The leukemias of childhood are cancers

of the hematopoietic system, involving in

most cases, malignant transformation of

lymphoid progenitor cells [1] and less

commonly transformation of myeloid pro-

genitor cells [2]. The leukemias account for

the largest number of cases of childhood

cancer and are the primary cause of cancer

related mortality of children in the United

States. Approximately 3,250 children and

adolescents younger than 20 years of age

are diagnosed with leukemia each year in

the US, of which 2,400 are acute lympho-

blastic leukemia. For the years from 1986-

94, the leukemias represented 32% of all

cancer cases occurring among children

younger than 15 years of age and 26% of

cancer cases occurring among those

younger than 20 years of age. However,

the relative contribution of leukemia to the

total childhood cancer burden varied mark-

edly with age, being 17% in the first year of

life, increasing to 46% for 2 and 3 year olds,

and then decreasing to only 9% for 19 year

olds. To further illustrate the contribution

Figure I.1 gives the incidence rates for both

leukemia and total cancer (the sum of

leukemia and non-leukemia) by single year

of age.

This chapter focuses on the following

topics related to the incidence of leukemia

among children in the United States: (1)

the relative frequencies of the leukemia

subtypes that occur among children; (2)

variation in the incidence of the specific

types of leukemia by age; (3) differences in

incidence between males and females; (4)

differences in incidence between white and

black children; and (5) variation in leuke-

mia incidence over time. In terms of sur-

vival for children with leukemia, the chap-

ter focuses on three primary topics: (1)

comparison of survival rates for children

with ALL and AML; (2) the impact of age at

diagnosis on survival; and (3) the remark-

Figure I.1: Total childhood cancer age-specific incidence rates

by leukemia versus non-leukemia, all races, both sexes, SEER, 1986-94

25 25

27 27

25 25

224

131

118

112

70 68

108

118

136

167

178

203 208

< 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

Age (in years) at diagnosis

100

150

200

250

300

Average annual rate per million

Leukemia

Non-Leukemia

Enumeration of the population at risk by single years of age was

available only for the census years 1980 and 1990. The US Bureau

of the Census provides intercensal population estimates by 5-year

age groups, but not by single years of age. Therefore, the

population estimates for 1980 were used in rate calculations for

cases diagnosed from 1976-84 and the 1990 estimates were used for

cases diagnosed from 1986-94.

ICCC ILEUKEMIA

National Cancer Institute SEER Pediatric Monograph

Table I.1: Percent distribution within ICCC subcategories for leukemia and age-adjusted*

incidence rates for specific ICD-O codes, age <20, all races, both sexes, SEER, 1975-95

Diagnostic Group Specific Diagnosis: Rate per million % of Cases

Ia: Lymphoid leukemia 29.2 100.0%

9820: Lymphoid leukemia, NOS 0.2%

9821: Acute lymphoblastic 99.2%

9822: Subacute lymphoid 0.0%

9823: Chronic lymphocytic 0.1%

9824: Aleukemic lymphoid 0.2%

9825: Prolymphocytic leukemia 0.1%

9826: Burkitt's cell leukemia 0.5%

9827: Adult T-cell 0.0%

9850: Lymphosarcoma cell 0.0%

Ib: Acute non-lymphocytic 7.6 100.0%

9840: Erythroleukemia 0.4%

9841: Acute erythremia 0.2%

9861: Acute myeloid leukemia 68.7%

9864: Aleukemic myeloid 0.0%

9866: Acute promyelocytic 7.1%

9867: Acute myelomonocytic 9.3%

9891: Acute monocytic leukemia 9.1%

9894: Aleukemic monocytic 0.0%

9910: Acute megakaryoblastic 5.1%

Ic: Chronic myeloid leukemia 1.3 100.0%

9863: Chronic myeloid leukemia 98.6%

9868: Chronic myelomonocytic 1.4%

Id: Other specified leukemias 0.2 100.0%

9830: Plasma cell leukemia 0.0%

9842: Chronic erythremia 0.0%

9860: Myeloid leukemia, NOS 33.3%

9862: Subacute myeloid 0.0%

9870: Basophilic leukemia 0.0%

9880: Eosinophilic leukemia 0.0%

9890: Monocytic leukemia, NOS 8.3%

9892: Subacute monocytic 0.0%

9893: Chronic monocytic 0.0%

9900: Mast cell leukemia 0.0%

9930: Myeloid sarcoma 58.3%

9931: Acute panmyelosis 0.0%

9932: Acute myelofibrosis 0.0%

9940: Hairy cell leukemia 0.0%

9941: Leukemia 0.0%

Ie: Unspecified leukemias 1.2 100.0%

9800: Leukemia, NOS 20.5%

9801: Acute leukemia, NOS 79.5%

9802: Subacute leukemia, NOS 0.0%

9803: Chronic leukemia, NOS 0.0%

9804: Aleukemia leukemia, NOS 0.0%

*Adjusted to the 1970 US standard population

ICCC I

National Cancer Institute

SEER Pediatric Monograph

LEUKEMIA

able improvements in survival rates for

children with ALL during the past 20

years.

Classification system

Before discussing topics related to

childhood leukemia incidence and outcome,

it is necessary to describe the specific

diagnoses that are included among the

Diagnostic Groups for leukemia of the

International Classification of Childhood

Cancer (ICCC). Table I.1 illustrates that

acute lymphoblastic leukemia (ALL) ac-

counted for approximately 99% of cases

among the lymphoid leukemia (Ia) diagnos-

tic group, so that this ICCC diagnostic

group is essentially synonymous with ALL.

The “acute non-lymphocytic leukemia”

diagnostic category Ib is henceforth re-

ferred to as the acute myeloid leukemia

(AML) category since this is the preferred

terminology [3], and it encompasses the

various subtypes of AML that occur in

children. The other three ICCC diagnostic

categories combined accounted for only 6-

7% of total leukemia cases in children. The

chronic myeloid leukemias diagnostic group

(Ic) included approximately 3% of leukemia

cases occurring in the younger than 20

years of age group during the period from

1990-95, while the “other specified leuke-

mia” diagnostic group (Id) included fewer

than 1% of the leukemia cases. Approxi-

mately 3% of leukemia cases were included

in the unspecified leukemia category (Ie)

for the period from 1990-95.

INCIDENCE

Age-specific incidence

Table I.2 shows the incidence and

relative proportion of specific diagnostic

categories by 5-year age groups. For the

younger than 15 years of age, ALL repre-

sented 78% of leukemia cases, while the

AML subgroup (Ib) represented 16% of

cases. The relative frequency of AML

increased in the second decade of life as

that of ALL decreased. While AML repre-

sented only 13-14% of leukemia cases in

the first 10 years of life, it accounted for

36% of leukemia cases among 15-19 year

olds. The incidence and relative contribu-

tion of the chronic myeloid leukemias also

increased with age, representing about 9%

of cases among 15-19 year olds.

As is apparent from Table I.2, the

incidence of leukemia among children

varied considerably with age. Figure I.2a

illustrates that this variation was the

result of a sharp peak in ALL incidence

among 2-3 year old children (incidence over

80 per million), which returned to a rate of

20 per million for 8-10 year old children.

The incidence of ALL among 2-3 year old

Table I.2: Age-adjusted incidence rates per million for specific leukemia by age groups

all races, both sexes, SEER, 1990-95

Age (in years) at diagnosis <5 5-9 10-14 15-19 <15*

Total leukemia 72.4

(100%)

38.0

(100%)

25.9

(100%)

26.0

(100%)

43.8

(100%)

ALL

58.1

(80%)

30.6

(81%)

17.4

(67%)

13.0

(50%)

34.0

(78%)

AML (Ib)

10.3

(14%)

5.0

(13%)

6.2

(24%)

9.3

(36%)

7.0

(16%)

CML (Ic) 1.1

(2%)

0.7

(2%)

1.1

(4%)

2.2

(9%)

1.0

(2%)

Other specified leukemias

(Id)

0.3

(-)

0.3

(1%)

0.1

(-)

0.1

(-)

0.2

(1%)

Unspecified leukemias (Ie)

2.2

(3%)

1.0

(3%)

0.6

(2%)

1.1

(4%)

1.2

(3%)

* Rates are adjusted to the 1970 US standard population. Numbers in parentheses represent the

percentage of the total cases for the specific age group.

ICCC ILEUKEMIA

National Cancer Institute SEER Pediatric Monograph

children was approximately 4-fold greater

than that for infants and was nearly 10-

fold greater than that for 19 year olds. The

distinctive shape of the age-incidence curve

for ALL can be well-described by a math-

ematical model which assumes that child-

hood ALL results from two events required

for full malignant transformation with the

first of these events occurring in utero [4].

Experimental confirmation has been ob-

tained for the initiation of childhood ALL in

utero [5,6].

The incidence of AML in children also

varied with age (Figure I.2b), but with a

different pattern than that for ALL. AML

rates were highest in the first 2 years of

life, but subsequently decreased with a

nadir at approximately 9 years of age,

followed by slowly increasing rates during

the adolescent years. The incidence of

leukemia cases in the “chronic myeloid

leukemia” category (Ic) likewise showed

substantial variation with age. As shown

in Figure I.2c, there was a peak in inci-

Figure I.2a: ALL (Ia): 1986-94, and AML (Ib): 1976-84

and 1986-94 age-specific incidence rates, all races

both sexes, SEER

(

01234567891011121314151617181920

Age (in years) at diagnosis

100

Average annual rate per million

ALL

AML

(

Figure I.2b: AML (Ib) age-specific incidence rates, all races

both sexes, SEER, 1976-84 and 1986-94 combined

(

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Age (in years) at diagnosis

Average annual rate per million

AML

(

ICCC I

National Cancer Institute

SEER Pediatric Monograph

LEUKEMIA

dence for males in the first year of life, with

subsequent lower rates for both males and

females until the late teen years. The

increase in CML incidence in the later teen

years appeared to represent the early

portion of the increasing incidence curve for

adult-type CML [7]. On the other hand, the

cases coded as “chronic myeloid leukemia”

in the first few years of life almost certainly

reflect cases of juvenile myelomonocytic

leukemia (previously termed juvenile

chronic myeloid leukemia), a diagnosis

associated primarily with young males

[8,9].

Sex-specific incidence

Table I.3 illustrates the incidence of

the various leukemia types separately for

males and females for the years 1990-95.

The incidence of ALL among children

younger than 15 years of age was consis-

tently higher among males (approximately

20%) relative to females. For the 15-19

year olds, however, the male preponderance

was greater, with males having a 2-fold

higher ALL incidence than females. The

incidence of AML was similar for males and

females for all age groups. For the CML

category, there was a 4-fold higher rate for

males than females for the younger than 5-

year age group, a difference that was not

present for older age groups. As noted

above, these cases likely represent juvenile

myelomonocytic leukemia, which has a

known male predominance [9].

Table I.3: Male to female ratios of age-adjusted leukemia incidence rates per million by

type and age group, all races, SEER, 1990-95

<5 Yrs

Rate

M/F Ratio

5-9 Yrs

Rate

M/F Ratio

10-14 Yrs

Rate

M/F Ratio

15-19 Yrs

Rate

M/F Ratio

<15 Yrs

Rate*

M/F Ratio

<20 Yrs

Rate*

M/F Ratio

Total leukemia

M = 78.5

F = 65.9

M/F = 1.2

M = 40.3

F = 35.7

M/F = 1.1

M = 28.4

F = 23.1

M/F = 1.2

M = 28.4

F = 23.4

M/F = 1.2

M = 47.4

F = 40.1

M/F = 1.2

M = 42.7

F = 36.0

M/F = 1.2

ALL M = 63.7

F = 52.3

M/F = 1.2

M = 32.2

F = 29.0

M/F = 1.1

M = 18.8

F = 16.0

M/F = 1.2

M = 17.2

F = 8.6

M/F = 2.0

M = 36.7

F = 31.2

M/F = 1.2

M = 31.9

F = 25.6

M/F = 1.2

AML (Ib) M = 10.0

F = 10.6

M/F = 0.9

M = 5.8

F = 4.3

M/F = 1.3

M = 6.7

F = 5.7

M/F = 1.2

M = 8.3

F = 10.4

M/F = 0.8

M = 7.4

F = 6.7

M/F = 1.1

M = 7.6

F = 7.6

M/F = 1.0

Chronic myeloid

leukemia (Ic)

M = 1.7

F = 0.4

M/F = 4.3

M = 0.4

F = 1.0

M/F = 0.4

M = 1.4

F = 0.9

M/F = 1.6

M = 1.6

F = 2.9

M/F = 0.6

M = 1.1

F = 0.8

M/F = 1.4

M = 1.3

F = 1.3

M/F = 1.0

*Adjusted to the 1970 US standard population

Figure I.2c: Chronic myeloid leukemia age-specific

incidence rates, all races, both sexes

SEER, 1976-84 and 1986-94 combined

)

))

)

))

0 1 2 3 4 5 6 7 8 9 1011 121314151617 181920

Age (in years) at diagnosis

Average annual rate per million

Male

Female

)

ICCC ILEUKEMIA

National Cancer Institute SEER Pediatric Monograph

Figure I.3: Leukemia age-specific incidence rates for white (1986-94)

and black (1976-84 and 1986-94) children, SEER

01234567891011121314151617181920

Age (in years) at diagnosis

100

110

120

130

140

Average annual rate per million

White

Black

Black-white differences in incidence

Leukemia rates were substantially

higher for white children younger than 15

years of age compared to black children,

with rates of 45.6 versus 27.8 per million

for the period from 1986-95 (Table I.4).

This difference between white children and

black children was most apparent when

examining rates of leukemia by single year

of age (Figure I.3), with a nearly 3-fold

higher incidence at 2-3 years of age for

white children compared to black children.

The difference in leukemia incidence was

primarily the result of lower ALL rates

among black children (Table I.4), with ALL

incidence for white children younger than 5

years of age being more than twice that for

black children (63.2 versus 26.9 per mil-

lion). A lower ALL incidence for black

children was observed for each 5-year age

group up to 20 years of age. The incidence

of AML, unlike that for ALL, was similar

for white and black children for all age

groups (Table I.4).

Table I.4: White to black ratios of age-adjusted leukemia incidence rates per million by

type and age group, both sexes, SEER, 1986-95

<5 Yrs

Rate*

W/B Ratio

5-9 Yrs

Rate

W/B Ratio

10-14 Yrs

Rate

W/B Ratio

15-19 Yrs

Rate

W/B Ratio

<15 Yrs

Rate*

W/B Ratio

<20 Yrs

Rate*

W/B Ratio

Total

Leukemia

W = 77.2

B = 38.4

W/B = 2.0

W = 37.5

B = 28.6

W/B = 1.3

W = 27.4

B = 18.4

W/B = 1.5

W = 26.5

B = 15.2

W/B = 1.7

W = 45.6

B = 27.8

W/B = 1.6

W = 40.9

B = 24.7

W/B = 1.7

ALL W = 63.2

B = 26.9

W/B = 2.4

W = 31.8

B = 1.8

W/B = 1.8

W = 19.8

B = 11.6

W/B = 1.7

W = 14.3

B = 6.4

W/B = 2.2

W = 36.8

B = 18.3

W/B = 2.0

W = 31.2

B = 15.4

W/B = 2.0

AML (Ib)

W = 10

B = 7.7

W/B = 1.3

W = 3.8

B = 6.1

W/B = 0.6

W = 5.3

B = 5.1

W/B = 1.0

W = 8.3

B = 7.1

W/B = 1.2

W = 6.2

B = 6.2

W/B = 1.0

W = 6.7

B = 6.4

W/B = 1.1

*Adjusted to the 1970 U.S. standard population

ICCC I

National Cancer Institute

SEER Pediatric Monograph

LEUKEMIA

TRENDS

The incidence of leukemia among

children younger than 15 years of age

increased in the past 20 years, as shown in

Figure I.4a. The estimated annual percent-

age change (EAPC) for total leukemia for

the period from 1977 to 1995 was 0.9% per

year, with the trend primarily reflecting an

increase in ALL incidence during this

period (EAPC for ALL, 0.9%). The rates of

leukemias, other than ALL, did not in-

crease significantly from 1977 to 1995

(Figure I.5), although the small number of

cases diagnosed each year for the less

common leukemia types results in consider-

able scatter in year to year rates, which

makes interpretations of trends difficult.

The higher rate of nonspecific classification

of leukemia cases (ICCC Category Ie) in the

years prior to 1977 (greater than 5 per

million in 1973 and 1974, but 1-2 per

million after 1977), is the reason for re-

stricting examinations of trends over time

for specific leukemia diagnoses to the

period from 1977 to 1995 [10].

While a model based on a constantly

increasing rate can be applied to the ALL

and the leukemia incidence data to esti-

mate an EAPC, visual inspection of the

incidence of ALL and total leukemia from

1977 to 1995 suggests that reality is more

complicated (Figure I.4a). For example,

ALL incidence for the 9 SEER areas and

the Los Angeles area combined peaked in

1989, and rates have been 5-10% below this

peak value in subsequent years. The

situation is further complicated by different

time trend patterns for ALL for the 9 SEER

areas compared with the Los Angeles area.

For the 9 SEER areas, ALL incidence has

been more or less stable since 1984. On the

other hand, ALL incidence for the Los

Angeles area showed more variability,

being higher in the late 1970s and early

Figure I.4b: Trends in ALL and non-ALL

age-adjusted* incidence rates, age <15

all races, both sexes, SEER, 1977-95

1977 1980 1983 1986 1989 1992 1995

Year of diagnosis

Average annual rate per million

ALL (9 SEER)

ALL (LA)

Non-ALL (LA)

Non-ALL (9 SEER)

*Adjusted to the 1970 US standard population

Figure I.4a: Trends in leukemia and ALL

age-adjusted* incidence rates, age <15

all races, both sexes, SEER, 1977-95

)

))

)

))

)

1977 1980 1983 1986 1989 1992 1995

Year of diagnosis

Average annual rate per million

Total Leukemia

ALL

Non-ALL

)

*Adjusted to the 1970 US standard population

ICCC ILEUKEMIA

National Cancer Institute SEER Pediatric Monograph

1980s, then decreasing to a nadir in 1984-

85, and subsequently increasing to rates

higher than those for the 9 SEER areas

(Figure I.4b). For 1990 to 1995, ALL rates

can be calculated for Hispanic and non-

Hispanic children. For non-Hispanic chil-

dren, the ALL rates were similar between

Los Angeles (LA) and the 9 SEER areas.

However, the ALL rates for Hispanic chil-

dren were higher than those for non-His-

panic children. Since over one-half of the

children younger than 15 residing in Los

Angeles were Hispanic, a much higher

proportion than for the 9 SEER areas, the

overall higher ALL rates in Los Angeles can

be explained by the higher proportion of

Hispanic children living in LA. In addition,

between 1990 and 1995, there were in-

creases in the population of Hispanic chil-

dren under 15 and decreases in the popula-

tion of non-Hispanic children under 15 in

LA. This change in population characteris-

tics would tend to produce increased rates

of ALL due to the higher rates of ALL

among Hispanic children. Therefore, the

higher ALL rates in LA can be explained by

the higher proportion of Hispanic children

in LA, and the increase in ALL rates for LA

can be at least partially explained by

increases in the percentage of Hispanic

children in LA. Ongoing monitoring of

childhood leukemia trends, as well as

epidemiologic and basic laboratory studies,

are needed to develop a better understand-

ing of the pathogenesis of childhood leuke-

mia and an enhanced ability to explain

changes in leukemia incidence over time.

Figure I.6 illustrates the incidence of

ALL for white and black children for the

period from 1977 to 1995. While the inci-

dence of ALL for white children increased

at an overall rate of approximately 1% per

year since 1977, there was no apparent

increase in ALL rates for black children

during this same period.

Figure I.5: Trends in non-ALL leukemia age-adjusted* incidence rates

age <15, all races, both sexes, SEER, 1977-95

1977 1982 1987 1992

Year of diagnosis

0.01

0.1

Average annual rate per million (log scale)

IB (ANLL)

IC (CML)

ID (Other)

IE (Unspecified)

*Adjusted to the 1970 US standard population

ICCC I

National Cancer Institute

SEER Pediatric Monograph

LEUKEMIA

Figure I.7 illustrates the variation in

ALL incidence for white children for specific

5-year age groups. Incidence rates for

white children 0-4, 5-9, 10-14, and 15-19

years of age demonstrated modest increases

when the entire time period was considered.

Because the incidence of ALL was greater

for those younger than 5 years of age than

for the older age groups, the increasing

rates for those younger than 5 years of age

accounted for the largest proportion of the

overall increase in ALL rates for white

children.

SURVIVAL

Survival for children with ALL mark-

edly improved from 1975-84 to 1985-94,

with 5-year survival rates for children

younger than 20 years of age increasing

from 61% to 77% (Figure I.8). A number of

improvements in treatment during this

period have undoubtedly contributed to the

improved survival rate, including: a) identi-

fication of increasingly effective methods of

central nervous system prophylaxis [1]; b)

identification of the contribution of treat-

ment intensification to improved outcome

for selected groups of patients [27-29].

Examples of effective methods of treatment

intensification include use of post-induction

consolidation with high-dose methotrexate

[28, 29] and use of post-remission re-induc-

tion/re-consolidation regimens (“delayed

intensification”) [27].

Survival for children with ALL is very

dependent upon age at diagnosis. For the

years 1985-94, 5-year survival rates were

highest for the 1-4 year age group and the

5-9 year age group (85% and 80%, respec-

tively) (Figure I.8). Infants had the poorest

outcome (37% 5-year survival rate), fol-

Figure I.7: Trends in ALL age-specific incidence

rates by year of diagnosis, white children

both sexes, SEER, 1977-95

)

1977 1982 1987 1992

Year of diagnosis

Average annual rate per million

<5 Years 5-9 Years

10-14 Years 15-19 Years

& )

+ %

)

1977 1982 1987 1992

Year of diagnosis

Average annual rate per million

White

Black

)

*Adjusted to the 1970 US standard population

Figure I.6: Trends in ALL age-adjusted*

incidence rates by race, age <15

both sexes, SEER, 1977-95

ICCC ILEUKEMIA

National Cancer Institute SEER Pediatric Monograph

lowed by the 15-19 year age group (51% 5-

year survival rate). The favorable progno-

sis of 1-9 year old children is likely related

to the relatively high proportion of cases in

this age range with favorable biological

subtypes (e.g., cases with hyperdiploid DNA

content or with the TEL-AML1 gene rear-

rangement) [30-32]. The poor prognosis for

infants with ALL reflects the high fre-

quency of cases with rearrangements of the

MLL gene on chromosome band 11q23

[33,34]. The less favorable outcome for

adolescents and young adults is likely due

in part to the increased relative frequency

of higher risk ALL subtypes (e.g., Philadel-

phia chromosome positive ALL and T-cell

ALL).

For the younger than 20 year old

population, 5-year survival rates were

slightly higher for females than for males

(79% versus 75%) (Figure I.8). Five-year

survival rates for black children younger

than 20 years of age with ALL were lower

than for white children (64% versus 78%).

While the poorer outcome for black children

with ALL could represent differences

between black and white children in the

pharmacokinetics or pharmacodynamics of

the drugs used for ALL treatment or differ-

ences in access to health care, the relative

paucity among black children of the most

curable ALL subtypes that occur at higher

incidence among white children younger

than 10 years of age may also contribute to

the poorer outcome observed for black

children.

Survival for children with AML was

substantially lower than that for children

with ALL (Figure I.9). While outcome for

children with AML improved significantly

from 1975-84 to 1985-94, 5-year survival

rates were only 41% for the period 1985-94

for the younger than 20 year old age group.

In contrast to ALL, older children and

adolescents with AML had outcome that

was similar to that observed for the

Figure I.8: ALL 5-year relative survival rates by sex, race, age and

time period, SEER (9 areas), 1975-84 and 1985-94

Total Male Female White Black <1 1-4 <5 5-9 10-14 15-19

100

Percent surviving 5 years

1975-84

1985-94

Sex Race Age

ICCC I

National Cancer Institute

SEER Pediatric Monograph

LEUKEMIA

younger than 5-year age group, with the

younger than 5-year age group having

somewhat lower outcome than for the older

age groups. As was the case for ALL,

outcome for females with AML was some-

what better than outcome for males. In

contrast to the poorer outcome for black

children with ALL, for AML outcome was

similar for white and for black children

younger than 20 years of age.

RISK FACTORS

Tables I.5 and I.6 summarize current

knowledge of the causes of childhood ALL

and AML. With the exception of prenatal

exposure to x-rays and specific genetic

syndromes, little is known about the causes

of and risk factors for childhood ALL [13].

It is important to note that ALL is a hetero-

geneous grouping of biological subtypes of

leukemia, and smaller studies of the past

may have lacked sufficient statistical power

to examine potential risk factors. Thus, one

emerging theme concerning the etiology of

childhood ALL is the need to separately

study different biological groups of ALL.

For example, the cases of ALL that arise in

infants and that have rearrangements of

the MLL gene on chromosome 11 appear to

have different epidemiological associations

than cases that arise in young children that

typically have B-precursor ALL

immunophenotype and hyperdiploid DNA

content [14-16]. Recognition of the need to

study these different ALL subtypes inde-

pendently has been one impetus for larger

studies of the etiology of ALL. In these

larger studies some intriguing associations

have emerged that can be followed up

further in more focused investigations. For

example, high birth weight and maternal

history of fetal loss have been associated

primarily with ALL occurring in children

younger than 2 years of age [17-19].

From a global perspective, childhood

ALL appears to be much more common in

Figure I.9: AML 5-year relative survival rates by sex, race, age and

time period, SEER (9 areas), 1975-84 and 1985-94

Total Male Female White Black <5 5-9 10-14 15-19

100

Percent surviving 5 years

1975-84

1985-94

Sex

Race

Age

ICCC ILEUKEMIA

National Cancer Institute SEER Pediatric Monograph

countries of the developed world than in

those of the developing world, a difference

that has been attributed to different pat-

terns of exposure of children in these popu-

lations to infectious agents [20,21]. A

delayed pattern of infections, as found in

countries such as the United States, is

hypothesized to somehow cause (or allow)

ALL to occasionally develop, possibly

through alterations in immune function.

However, in the absence of an understand-

ing of specific infectious agents or specific

immune perturbations associated with the

pathogenesis of ALL, it is not possible to

apply these theories to explain the trends

in incidence for childhood ALL observed in

the United States since the 1970s.

Table I.5: Current knowledge on causes of acute lymphoblastic leukemia (ALL)

Exposure or Characteristic Comments References

Sex Overall, there is about a 30% higher incidence in males

compared to females.

16,22,36

Age There is a peak in the incidence between the ages of about

2 and 5.

16,22,36

Race There is an approximate 2-fold higher risk in white

children compared to black children.

16,22,36

Higher socioeconomic status Increased risk has been fairly consistently associated with

the most common ALL (diagnosed at ages 2 - 5 years). It

is unknown what aspect of higher SES is relevant but

higher age of exposure to infectious agents has been

hypothesized.

16,22,36,37

Ionizing radiation (in utero) In past studies, there was a consistent, increased risk

(about 1.5 fold) of leukemia associated with prenatal

diagnostic x-ray exposure. However, this is unlikely to be

an important risk factor for childhood leukemia today due

to fewer x-rays, increased shielding, and lower radiation

levels.

16,22,36,38,39

Ionizing radiation postnatal

(therapeutic)

Therapeutic radiation for such conditions as tinea capitis

and thymus enlargement has been associated with an

increased risk.

16,22,36,40

Down syndrome,

neurofibromatosis, Shwachman

syndrome, Bloom syndrome,

ataxia telangiectasia,

Langerhans cell histiocytosis, and

Klinefelter syndrome

Increased occurrence is associated with these genetic

conditions and is particularly apparent in children with

Down syndrome for whom there is a reported 20-fold

increased risk of leukemia.

16,22,26,36

High birth weight (> 4000 grams) Several studies have reported an elevated risk

(approximately 2-fold) in larger babies, particularly for

children diagnosed younger than two years of age.

16,22,36,41

Maternal history of fetal loss

prior to the birth of the index

child

Approximately 2-5 fold increased risk of leukemia has

been noted in a few studies; particularly in children

diagnosed younger than two years of age.

16,18,22,36,42

Maternal age > 35 at pregnancy A slight increased risk has been somewhat inconsistently

associated with older maternal age.

16,22,36

First born or only child Slight increased risk reported but birth order may be

surrogate marker for exposure to infectious agent.

16,22,36

Known risk factors

Factors for which evidence is

suggestive but not conclusive

ICCC I

National Cancer Institute

SEER Pediatric Monograph

LEUKEMIA

Table I.5 (cont’d): Current knowledge on causes of acute lymphoblastic leukemia (ALL)

Smoking prior to and during

pregnancy

Some studies have reported an increased risk associated

with maternal smoking during pregnancy but others have

not. A few recent studies have suggested a modest

increased risk (about 1.5 fold) associated with paternal

smoking prior to pregnancy.

16,22,36,43-48

Parental occupations and

occupational exposures

Isolated reports associated with parental exposure to

motor vehicle exhaust, hydrocarbons, and paints. This is

the subject of several current epidemiologic studies.

16,22,36

Postnatal infections Evidence is very inconsistent. 16,22

Diet A few reports have suggested that meat consumption

(particularly, cured meats) is associated with an increased

risk. Maternal diet and childhood leukemia has not been

explored in any detail and further study is warranted.

49,50

Electromagnetic fields A few studies have reported a slight increased risk for

children living near high voltage power lines; others have

reported no association. A recent large study of U.S.

children with ALL found little or no association between

risk of ALL and electromagnetic field exposure. Other

large epidemiologic studies evaluating this exposure are

ongoing.

16,22,51-54

Vitamin K prophylaxis in

newborns

Although an increased risk of ALL was first reported in

the early 1990's, several large studies since then have

found no association.

16,22,55-60

Maternal alcohol consumption

during pregnancy

Unlikely to be an important risk factor for ALL (but see

AML).

16,22

Postnatal use of chloramphenicol One study reported quite substantial increased risks

(approximately 10-fold) with postnatal use of this broad-

spectrum antibiotic.

Ultrasound 16,22

*Note that the majority of these risk factors have been reviewed recently in references 16,22,36; only

selected references are presented for additional reading.

Factors for which evidence is

inconsistent or limited

Factors unrelated to risk

Different risk factors are emerging for

childhood AML that distinguish the disease

from ALL, and this may provide avenues

for future epidemiological studies. For

example, exposure to specific chemotherapy

agents has been associated with an in-

creased risk of childhood AML, in contrast

to the rarity of treatment-related ALL [22-

24]. With this information, it may be

possible to design epidemiological studies to

examine exposures to environmental

agents that have a biologic nature that is

similar to these chemotherapy agents [14].

Further

associations with factors such as

benzene and pesticides that have emerged

in a few studies suggest that childhood

AML may share risk factors with adult

AML, and this is being investigated in

several large epidemiological studies [25].

Finally, as for ALL, the associations of AML

with genetic syndromes are compelling, as

illustrated by the magnitude of risk of AML

in Down syndrome [26].

SUMMARY

ALL is by far the most common type of

leukemia occurring in children and shows a

distinctive age-distribution pattern, with a

marked incidence peak at 2-3 years of age.

ICCC ILEUKEMIA

National Cancer Institute SEER Pediatric Monograph

Known risk factors

Factors for which evidence is

suggestive but not conclusive

Factors for which evidence is

inconsistent or limited

Table I.6: Current knowledge on causes of acute myeloid leukemia (AML)

Exposure or Characteristic Comments References

Race The highest incidence rates are reported in the

Hispanic children.

16,22,36

Chemotherapeutic agents Increased risk is associated with prior exposure to

alkylating agents or epipodophyllotoxins.

16,22,36,62,63

Ionizing radiation (

in utero

) In past studies, there was a consistent, increased risk

(about 1.5 fold) of leukemia associated with prenatal

diagnostic x-ray exposure. However, this is unlikely

to be an important risk factor for childhood leukemia

today due to fewer x-rays, increased shielding, and

lower radiation levels.

16,22,36

Down syndrome, neurofibromatosis,

Shwachman syndrome, Bloom

syndrome, familial monosomy 7,

Kostmann granulocytopenia, Fanconi

anemia

Increased occurrence associated with these genetic

conditions, particularly with Down syndrome. One

report suggests as high as a 500-fold increased risk of

a specific type of AML in Down syndrome.

16,22,26,36

Maternal alcohol consumption during

pregnancy

Three studies have reported an increased risk

(approximately 1.5-2 fold) in mothers who drank

alcoholic beverages during pregnancy. These

associations have been particularly apparent in

children diagnosed younger than three years of age.

16,22,36,64

Parental and child exposure to

pesticides

Increased risk has been noted in a few studies and in

adult AML data; subject of several current

investigations.

16,22,25,36,45

Parental exposure to benzene Exposure has been associated with an increased risk

in several studies; again follows adult AML data; also

subject of several current investigations.

16,22,36,45

Maternal use of recreational drugs

during pregnancy

One report suggested that maternal marijuana use

during pregnancy was associated with increased risk.

Radon A few correlational studies have suggested an

increased risk of childhood and adult AML in areas

with high radon concentrations; this is a subject of

several current epidemiologic studies of AML.

16,22,36

Postnatal use of chloramphenicol As in ALL, one study found quite substantial

increased risks of AML (approximately 10-fold) with

postnatal use of this broad-spectrum antibiotic.

*Note that the majority of these risk factors have been reviewed recently in references [16,22,36]; only

elected references are presented for additional reading.

ICCC I

National Cancer Institute

SEER Pediatric Monograph

LEUKEMIA

The peak at 2-3 years of age is much less

apparent for black children than for white

children, with this difference accounting for

the substantially lower incidence of ALL

observed for black children. By contrast

with ALL, the age-distribution pattern for

AML shows highest rates in the first two

years of life, with decreasing incidence until

10 years of age followed by increasing rates

thereafter. Again in contrast to ALL, the

incidence of AML in black children and

white children is similar.

The improvement in survival for chil-

dren with ALL over the past 35 years is one

of the great success stories of clinical oncol-

ogy. Survival rates for childhood ALL were

below 5% in the early 1960s, but are now

approaching 80% [35]. Outcome for chil-

dren with AML has also improved, but 5-

year survival rates have increased to only

the 40% range. Black children with ALL

have poorer outcome than do white chil-

dren, but for AML there is similar outcome

for black children and white children. Since

the peak in ALL incidence at 2-3 years of

age is much lower for black children than

for white children and since these ALL

cases in young children are known to have

the most favorable prognosis, the poorer

outcome for black children may reflect in

part a different distribution of biological

subtypes of ALL in black children compared

to white children.

Reference List

1. Margolin J, Poplack D: Acute lymphoblastic

leukemia. In Principles and Practice of Pediat-

ric Oncology (Pizzo P, Poplack D, eds). Philadel-

phia: Lippincott-Raven, 1997, pp 409-462.

2. Golub T, Weinstein H, Grier H: Acute myelog-

enous leukemia. In Principles and Practice of

Pediatric Oncology (Pizzo P, Poplack D, eds).

Philadelphia: Lippincott-Raven, 1997, pp 463-

482.

3. Cheson BD, Cassileth PA, Head DR, et al:

Report of the National Cancer Institute-

sponsored workshop on definitions of diagnosis

and response in acute myeloid leukemia. J

Clin Oncol 8:813-9, 1990.

4. Smith M, Chen T, Simon R: Age-specific

incidence of acute lymphoblastic leukemia in

U.S. children: in utero initiation model. J Natl

Cancer Inst 89:1542-1544, 1997.

5. Gale KB, Ford AM, Repp R, et al: Backtracking

leukemia to birth: identification of clonotypic

gene fusion sequences in neonatal blood spots.

Proc Natl Acad Sci U S A 94:13950-4, 1997.

6. Ford AM, Bennett CA, Price CM, et al: Fetal

origins of the TEL-AML1 fusion gene in

identical twins with leukemia. Proc Natl Acad

Sci U S A 95:4584-8, 1998.

7. Ries L, Kosary C, Hankey B, et al: SEER

Cancer Statistics Review, 1973-1994. Bethesda,

MD: National Cancer Institute, 1997.

8. Hess JL, Zutter MM, Castleberry RP, et al:

Juvenile chronic myelogenous leukemia. Am J

Clin Pathol 105:238-48, 1996.

9. Arico M, Biondi A, Pui CH: Juvenile

myelomonocytic leukemia [see comments].

Blood 90:479-88, 1997.

10. Gurney JG, Davis S, Severson RK, et al: Trends

in cancer incidence among children in the U.S.

Cancer 78:532-41, 1996.

11. Bunin GR, Feuer EJ, Witman PA, et al: Increas-

ing incidence of childhood cancer: report of 20

years experience from the greater Delaware

Valley Pediatric Tumor Registry. Paediatr

Perinat Epidemiol 10:319-38, 1996.

12. Linet MS, Devesa SS: Descriptive epidemiology

of childhood leukaemia. Br J Cancer 63:424-9,

1991.

13. Chow W-H, Linet M, Liff J, et al: Cancers in

children. In Cancer Epidemiology and Preven-

tion (Schottenfeld D, Frameni Jr. J, eds). New

York: Oxford, 1996, pp 1331-69.

14. Ross J, Potter J, Robison L: Infant leukemia,

topoisomerase II inhibitors, and the MLL gene.

J Natl Cancer Inst 86:1678-1680, 1994.

15. Ross JA, Potter JD, Reaman GH, et al: Mater-

nal exposure to potential inhibitors of DNA

topoisomerase II and infant leukemia (United

States): a report from the Children’s Cancer

Group. Cancer Causes Control 7:581-90, 1996.

16. Ross JA, Davies SM, Potter JD, et al: Epidemi-

ology of childhood leukemia, with a focus on

infants. Epidemiol Rev 16:243-72, 1994.

17. Ross JA, Potter JD, Shu XO, et al: Evaluating

the relationships among maternal reproductive

history, birth characteristics, and infant

leukemia: a report from the Children’s Cancer

Group. Ann Epidemiol 7:172-9, 1997.

18. Yeazel MW, Buckley JD, Woods WG, et al:

History of maternal fetal loss and increased

risk of childhood acute leukemia at an early

age. A report from the Childrens Cancer

Group. Cancer 75:1718-27, 1995.

ICCC ILEUKEMIA

National Cancer Institute SEER Pediatric Monograph

19. Yeazel MW, Ross JA, Buckley JD, et al: High

birth weight and risk of specific childhood

cancers: a report from the Children’s Cancer

Group. J Pediatr 131:671-7, 1997.

20. Greaves M: Aetiology of acute leukemia. Lancet

349:344-349, 1997.

21. Smith MA, Simon R, Strickler HD, et al:

Evidence that childhood acute lymphoblastic

leukemia is associated with an infectious agent

linked to hygiene conditions. Cancer Causes

Control 9:285-98, 1998.

22. Robison L, Ross J: Epidemiology of leukaemias

and lymphomas in childhood. In Bailliere’s

Clinical Paediatrics (Chessells J, Hann I, eds).

London: WB Saunders Co., 1995, pp pp. 639-

657.

23. Smith M, Rubinstein L, Ungerleider R:

Therapy-related acute myeloid leukemia

following treatment with epipodophyllotoxins:

Estimating the risks. Med Pediatr Oncol 23:86-

98, 1994.

24. Hunger S, Sklar J, Link M: Acute lymphoblas-

tic leukemia occurring as a second malignant

neoplasm in childhood: Report of three cases

and review of the literature. J Clin Oncol

10:156-163, 1992.

25. Buckley JD, Robison LL, Swotinsky R, et al:

Occupational exposures of parents of children

with acute nonlymphocytic leukemia: a report

from the Childrens Cancer Study Group.

Cancer Res 49:4030-7, 1989.

26. Zipursky A, Brown E, Christensen H, et al:

Leukemia and/or myeloproliferative syndrome

in neonates with Down syndrome. Semin

Perinatol 21:97-101, 1997.

27. Tubergen D, Gilchrist G, O’Brien R, et al:

Improved outcome with delayed intensification

for children with acute lymphoblastic leukemia

and intermediate presenting features: A

Childrens Cancer Group Phase III Trial. J Clin

Oncol 11:527-537, 1993.

28. Land VJ, Shuster JJ, Crist WM, et al: Compari-

son of two schedules of intermediate-dose

methotrexate and cytarabine consolidation

therapy for childhood B-precursor cell acute

lymphoblastic leukemia: a Pediatric Oncology

Group study. J Clin Oncol 12:1939-45, 1994.

29. Mahoney DH, Jr., Shuster J, Nitschke R, et al:

Intermediate-dose intravenous methotrexate

with intravenous mercaptopurine is superior to

repetitive low-dose oral methotrexate with

intravenous mercaptopurine for children with

lower-risk B-lineage acute lymphoblastic

leukemia: a Pediatric Oncology Group phase

III trial. J Clin Oncol 16:246-54, 1998.

30. Trueworthy R, Shuster J, Look T, et al: Ploidy

of lymphoblasts is the strongest predictor of

treatment outcome in B-progenitor cell acute

lymphoblastic leukemia of childhood: A Pediat-

ric Oncology Group Study. J Clin Oncol 10:606-

613, 1992.

31. Rubnitz JE, Shuster JJ, Land VJ, et al: Case-

control study suggests a favorable impact of

TEL rearrangement in patients with B-lineage

acute lymphoblastic leukemia treated with

antimetabolite-based therapy: a Pediatric

Oncology Group study. Blood 89:1143-6, 1997.

32. Rubnitz JE, Downing JR, Pui CH, et al: TEL

gene rearrangement in acute lymphoblastic

leukemia: a new genetic marker with prognos-

tic significance. J Clin Oncol 15:1150-7, 1997.

33. Rubnitz JE, Link MP, Shuster JJ, et al: Fre-

quency and prognostic significance of HRX

rearrangements in infant acute lymphoblastic

leukemia: a Pediatric Oncology Group study.

Blood 84:570-3, 1994.

34. Hilden JM, Frestedt JL, Moore RO, et al:

Molecular analysis of infant acute lymphoblas-

tic leukemia: MLL gene rearrangement and

reverse transcriptase-polymerase chain

reaction for t(4; 11)(q21; q23). Blood 86:3876-82,

1995.

35. Harras A: Cancer: Rates and Risks. In National

Institutes of Health publication No. 96-691.

Bethesda, MD: National Cancer Institute, 1996.

36. Sandler DP, Ross JA: Epidemiology of acute

leukemia in children and adults. Semin Oncol

24:3-16, 1997.

37. McWhirter W: The relationship of incidence of

incidence of childhood lymphoblastic leukaemia

to social class. Br J Cancer 46:640-645, 1982.

38. MacMahon B, Newill V: Birth characteristics of

children dying of malignant neoplasms. J Natl

Cancer Inst 37:687-698, 1962.

39. Ford D, Paterson J, Treuting W: Fetal exposure

to diagnostic x-rays, and leukemia and other

malignant disease of childhood. J Natl Cancer

Inst 22:1093-1104, 1959.

40. Ron E, Modan B: Thyroid and other neoplasms

following childhood scalp irradiation. In

Radiation Carcinogenesis, Epidemiology and

Biological Significance (Boice J, Fraumeni J,

eds). New York, NY: Raven Press, 1984, pp 139.

41. Ross JA, Perentesis JP, Robison LL, et al: Big

babies and infant leukemia: a role for insulin-

like growth factor-1? Cancer Causes Control

7:553-9, 1996.

42. Kaye SA, Robison LL, Smithson WA, et al:

Maternal reproductive history and birth

characteristics in childhood acute lymphoblas-

tic leukemia. Cancer 68:1351-5, 1991.

43. Neutel CI, Buck C: Effect of smoking during

pregnancy on the risk of cancer in children. J

Natl Cancer Inst 47:59-63, 1971.

44. van Steensel-Moll HA, Valkenburg HA,

Vandenbroucke JP, et al: Are maternal fertility

problems related to childhood leukaemia? Int J

Epidemiol 14:555-9, 1985.

ICCC I

National Cancer Institute

SEER Pediatric Monograph

LEUKEMIA

45. Shu XO, Gao YT, Brinton LA, et al: A popula-

tion-based case-control study of childhood

leukemia in Shanghai. Cancer 62:635-44, 1988.

46. Ji BT, Shu XO, Linet MS, et al: Paternal

cigarette smoking and the risk of childhood

cancer among offspring of nonsmoking mothers.

J Natl Cancer Inst 89:238-44, 1997.

47. Sorahan T, Prior P, Lancashire RJ, et al:

Childhood cancer and parental use of tobacco:

deaths from 1971 to 1976. Br J Cancer 76:1525-

31, 1997.

48. Sorahan T, Lancashire RJ, Hulten MA, et al:

Childhood cancer and parental use of tobacco:

deaths from 1953 to 1955. Br J Cancer 75:134-

8, 1997.

49. Peters JM, Preston-Martin S, London SJ, et al:

Processed meats and risk of childhood leuke-

mia (California, USA). Cancer Causes Control

5:195-202, 1994.

50. Sarasua S, Savitz DA: Cured and broiled meat

consumption in relation to childhood cancer:

Denver, Colorado (United States) [see com-

ments]. Cancer Causes Control 5:141-8, 1994.

51. Wertheimer N, Leeper E: Electrical wiring

configurations and childhood cancer. Am J

Epidemiol 109:273-84, 1979.

52. Savitz DA, Wachtel H, Barnes FA, et al: Case-

control study of childhood cancer and exposure

to 60-Hz magnetic fields [see comments]. Am J

Epidemiol 128:21-38, 1988.

53. Verkasalo PK, Pukkala E, Hongisto MY, et al:

Risk of cancer in Finnish children living close

to power lines [see comments]. BMJ 307:895-9,

1993.

54. Linet MS, Hatch EE, Kleinerman RA, et al:

Residential exposure to magnetic fields and

acute lymphoblastic leukemia in children [see

comments]. N Engl J Med 337:1-7, 1997.

55. Golding J, Greenwood R, Birmingham K, et al:

Childhood cancer, intramuscular vitamin K,

and pethidine given during labour [see com-

ments]. BMJ 305:341-6, 1992.

56. Klebanoff MA, Read JS, Mills JL, et al: The

risk of childhood cancer after neonatal expo-

sure to vitamin K [see comments]. N Engl J

Med 329:905-8, 1993.

57. Ekelund H, Finnstrom O, Gunnarskog J, et al:

Administration of vitamin K to newborn

infants and childhood cancer. BMJ 307:89-91,

1993.

58. McKinney PA, Juszczak E, Findlay E, et al:

Case-control study of childhood leukaemia and

cancer in Scotland: findings for neonatal

intramuscular vitamin K [see comments]. BMJ

316:173-7, 1998.

59. Parker L, Cole M, Craft AW, et al: Neonatal

vitamin K administration and childhood cancer

in the north of England: retrospective case-

control study [see comments]. BMJ 316:189-93,

1998.

60. Passmore SJ, Draper G, Brownbill P, et al:

Case-control studies of relation between

childhood cancer and neonatal vitamin K

administration [see comments]. BMJ

316:178-84, 1998.

61. Shu XO, Gao YT, Linet MS, et al: Chloram-

phenicol use and childhood leukaemia in

Shanghai. Lancet 2:934-7, 1987.

62. Pui CH, Behm FG, Raimondi SC, et al:

Secondary acute myeloid leukemia in chil-

dren treated for acute lymphoid leukemia

[see comments]. N Engl J Med 321:136-42,

1989.

63. Smith M, McCaffrey R, Karp J: The secondary

leukemias: Challenges and research direc-

tions. J Natl Cancer Inst 88:407-418, 1996.

64. Shu XO, Ross JA, Pendergrass TW, et al:

Parental alcohol consumption, cigarette

smoking, and risk of infant leukemia: a

Childrens Cancer Group study. J Natl Cancer

Inst 88:24-31, 1996.

65. Robison LL, Buckley JD, Daigle AE, et al:

Maternal drug use and risk of childhood

nonlymphoblastic leukemia among offspring.

An epidemiologic investigation implicating

marijuana (a report from the Childrens

Cancer Study Group). Cancer 63:1904-11,

1989.

ICCC IILYMPHOMAS AND RETICULOENDOTHELIAL NEOPLASMS

National Cancer Institute

SEER Pediatric Monograph

Constance L. Percy, Malcolm A. Smith, Martha Linet, Lynn A. Gloeckler Ries, Debra L. Friedman

HIGHLIGHTS

Incidence

♦ Approximately 15% of childhood malignancies were lymphomas making them the

third most frequent type of cancer in children following leukemia and malignant

brain tumors. The percentage of childhood cancer that were lymphoma varied by

age from only 3% for children younger than 5 years of age to 24% for 15-19 year olds

(Figure II.1).

♦ The two predominant types of lymphomas were Hodgkin’s disease and the non-

Hodgkin’s lymphomas (NHL). For younger children NHL was more frequent than

Hodgkin’s disease, while the reverse was true for adolescents (Figure II.3).

♦ In the US, approximately 1,700 children and adolescents younger than 20 years of

age are diagnosed with lymphomas each year of which approximately 850-900 are

cases of Hodgkin’s disease and 750-800 are cases of NHL.

♦ The most common subtypes of Hodgkin’s disease were nodular sclerosis (70% of

cases); mixed cellularity (16% of cases); lymphocytic predominance (7% of cases);

cases not otherwise specified (NOS) (6% of cases); and lymphocytic depletion subtype

(<2% of cases) among children younger than 20 years of age. The relative frequen-

cies of the mixed cellularity and nodular sclerosis subtypes were age and sex depen-

dent (Figure II.2 and Table II.2).

♦ The incidence of Hodgkin’s disease for children and adolescents younger than 20

years of age decreased slightly between 1975 and 1995, from 14.5 per million (1975-

79) to 12.1 per million (1990-95) (Table II.3).

♦ The non-Hodgkin’s lymphomas of children are a heterogeneous group of tumors,

with Burkitt’s and Burkitt-like tumors predominating among 5-14 year olds, and

with diffuse large cell lymphomas being the most common subtype among 15-19

year olds (Figure II.9).

♦ The incidence of NHL varied much less by age than Hodgkin’s disease (Figure II.3).

NHL incidence increased up until age 4 years where it reached a plateau of approxi-

mately 10 per million (Figure II.4), which was maintained until the second decade of

life when rates increased again.

♦ The incidence of NHL was higher in males than females (Figure II.10) and higher

among whites than blacks (Figure II.12).

♦ The incidence of NHL among children younger than 15 years of age was fairly

constant over the past 21 years, while there appeared to have been a slight increase

in incidence for the 15-19 year old population (Table II.3 and Figure II.13).

Survival

♦ The 5-year survival rate was 91% for Hodgkin’s disease (Figure II.8) for children

and adolescents younger than 20 years of age, compared to 72% for NHL (Figure

II.14).

♦ The 5-year survival rate for those younger than 20 years of age with NHL increased

from 56% in 1975-84 to 72% in 1985-94 (Figure II.14).

Risk factors

♦ For Hodgkin’s disease arising in young adults, genetic susceptibility may be a factor

for some cases, based on the greatly increased risk for Hodgkin’s disease in young

adult monozygotic twins of patients with Hodgkin’s disease compared to the risk of

dizygotic twins of patients with Hodgkin’s disease (Table II.4).

♦ Congenital immunodeficiency syndromes and acquired immunodeficiency syndrome

(AIDS) are associated with an increased risk of NHL (Table II.6).

National Cancer Institute

SEER Pediatric Monograph

ICCC II LYMPHOMAS AND RETICULOENDOTHELIAL NEOPLASMS

INTRODUCTION

The lymphomas, combining Hodgkin’s

disease and the non-Hodgkin’s lymphomas

(NHL), are the third most frequent type of

cancer in children following leukemia and

malignant brain tumors. NHL and

Hodgkin’s disease are both malignancies of

lymphoid cells, and each includes distinc-

tive biological subtypes. In the US, ap-

proximately 1,700 children and adolescents

younger than 20 years of age are diagnosed

with lymphomas each year, of which 850-

900 are Hodgkin’s disease and 750-800 are

non-Hodgkin’s lymphoma. The lymphomas

account for 10% of malignancies among

children younger than 15 years of age and

15% of malignancies among those younger

than 20 years of age. Figure II.1 illustrates

that the contribution of Hodgkin’s disease

and the non-Hodgkin’s lymphomas to the

overall childhood cancer burden is mark-

edly age dependent, increasing from only

3% of cancers among children younger than

5 years of age to 24% of cancers arising

among those 15-19 years of age.

MATERIAL AND METHODS

The International Classification for

Childhood Cancers (ICCC) Group II of

Lymphomas and Reticuloendothelial Neo-

plasms is divided into 5 subgroups [1]:

a. Hodgkin’s disease;

b. Non-Hodgkin’s lymphoma;

c. Burkitt’s lymphoma;

d. Miscellaneous lymphoreticular

neoplasms; and

e. Unspecified lymphomas.

Since only 3% of the registered child-

hood lymphomas and reticuloendothelial

neoplasms in the Surveillance, Epidemiol-

ogy and End Results (SEER) program for

the period 1975-95 were not histologically

confirmed, all cases were included in the

analyses. Table II.1 shows the frequency

distribution of these cases by sex among

children younger than 20 years of age

diagnosed with lymphoma. For the pur-

poses of this chapter, Hodgkin’s disease

(defined by subgroup IIa) and NHL (defined

by subgroups IIb, IIc, and IIe) are consid-

ered separately because of their distinctive

nature. The miscellaneous lymphoreticular

neoplasms (IId) are excluded from further

analysis as there were few cases (only 79

cases) in this subgroup from 1975-95.

Table II.1: Number of cases of lymphoma by type

and sex, age <20, all races, SEER, 1975-95

Total Males

(%)

Females

(%)

Total 4595 2734

(59%)

1861

(41%)

Hodgkin’s (IIa) 2613 1353

(52%)

1260

(48%)

NHL (IIb,c,e) 1903 1334

(70%)

569

(30%)

Miscellaneous (IId) 79 47

(59%)

(41%)

Figure II.1: Hodgkin's disease and NHL as a percent of

total childhood cancer, by age, all races, both sexes

SEER, 1990-95

<5 5-9 10-14 15-19 <15 <20

Age (in years) at diagnosis

Relative percent

Hodgkin's

NHL

National Cancer Institute SEER Pediatric Monograph

ICCC IILYMPHOMAS AND RETICULOENDOTHELIAL NEOPLASMS

Hodgkin’s disease is an unusual malig-

nancy of lymphoid cells whose distinctive

hallmark is the Reed-Sternberg cell. In

most cases Reed-Sternberg cells appear to

be clonal in nature and from the B lympho-

cyte lineage, as evidenced by immunoglobu-

lin gene rearrangements in these cells [2-5].

Reed-Sternberg cells account for only a

small percentage of the tumor mass [6],

with most of the tumor composed of a

reactive infiltrate of lymphocytes, plasma

cells, and eosinophils. The Reed-Sternberg

cells of some patients with Hodgkin’s

disease contain copies of the Epstein-Barr

virus (EBV) genome [7,8]. Detection of

EBV in Reed-Sternberg cells is more com-

mon among cases diagnosed in young

children and for cases of the mixed cellular-

ity subtype [7,8] (see Histologic subtype

and Risk Factors sections below).

Histologic subtype

Childhood Hodgkin’s disease, similar to

that arising in adults, is usually classified

according to the Rye classification scheme,

which includes four histologic subtypes:

lymphocytic predominance, mixed cellular-

ity, lymphocytic depletion, and nodular

sclerosis [9]. The lymphocytic predomi-

nance subtype is now recognized as having

a distinctive biological and clinical behavior

from the other subtypes of Hodgkin’s dis-

ease and is relatively uncommon among

children [10,11]. Among children younger

than 20 years of age diagnosed with

Hodgkin’s disease during 1975-95, the

nodular sclerosis subtype was by far the

most common and accounted for 70% of all

cases. Mixed cellularity was the second

most common subtype (16% of cases),

followed by lymphocytic predominance (7%

of cases), and cases not otherwise specified

(NOS), (6% of cases). The lymphocytic

depletion subtype was distinctly uncommon

(<2% of cases) among Hodgkin’s disease

cases diagnosed in those younger than 20

years of age.

The relative frequencies of the mixed

cellularity and nodular sclerosis subtypes

were age and sex dependent. Mixed cellu-

larity subtype was more common among

children younger than 10 years of age (32%

of all Hodgkin’s cases) than among those

diagnosed at 10-14 or 15-19 years of age

(15% and 13% of Hodgkin’s cases, respec-

tively) (Table II.2 and Figure II.2). Mixed

cellularity was also more common among

males than females in each age group.

Table II.2: Percent distribution of Hodgkin’s by subtype, age and

sex all races, SEER, 1975-95

Age (in years) at

diagnosis

<10 10-14 15-19

Both M F Both M F Both M F

Mixed Cellularity

32% 34% 25% 15% 21% 9% 13% 17% 10%

Nodular Sclerosis

46% 44% 51% 69% 59% 79% 74% 67% 81%

Lymphocytic

Predominance

14% 15% 12% 7% 10% 3% 5% 7% 2%

Figure II.2: Hodgkin's disease age-adjusted* incidence

rates by Rye classification, sex and age

all races, SEER (10 areas), 1977-95

0.8

1.9

4.5

1.3

2.7

5.4

0.4

3.5

1.3

8.9

1.9

22.2

0.8

9.9

0.6

1.9

3.8

0.9

2.5

4.6

0.3

1.2

<10 10-14 15-19 <10 10-14 15-19 <10 10-14 15-19

Average annual rate per million

Mixed cellularity

Nodular sclerosis

Other

Both Sexes

Male

Female

*Adjusted to the 1970 US standard population

HODGKIN’S DISEASE

National Cancer Institute

SEER Pediatric Monograph

ICCC II LYMPHOMAS AND RETICULOENDOTHELIAL NEOPLASMS

Nodular sclerosis subtype accounted for the

vast majority (81%) of Hodgkin’s disease

among 15-19 year old females, and a some-

what lower percentage (approximately

67%) of Hodgkin’s disease among 15-19

year old males (Table II.2).

Age-specific incidence

The overall annual incidence rate for

Hodgkin’s disease among children younger

than 20 years of age was 12.1 per million

for the years 1990-95. For Hodgkin’s

disease, the contribution to overall child-

hood cancer incidence was less than 1%

among children younger than 5 years of

age, but increased to 16% for 15-19 year

olds (Figure II.1). The incidence of

Hodgkin’s disease was markedly age depen-

dent (Figures II.3 and II.4), with incidence

increasing from less than one per million

for children in the first 3 years of life, to

Figure II.3: Hodgkin's disease and NHL age-specific and age-adjusted*

incidence rates by age, all races, both sexes

SEER, 1990-95

6.4

8.2

10.7

16.3

8.6

10.5

0.9

3.6

11.2

5.5

12.1

<5 5-9 10-14 15-19 <15 <20

Age (in years) at diagnosis

Average annual rate per million

NHL

Hodgkin's

*Adjusted to the 1970 US standard population

Figure II.4: Hodgkin's disease and NHL age-specific

incidence rates, all races, both sexes

SEER, 1976-84 and 1986-94

(

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Age (in years) at diagnosis

Average annual rate per million

NHL (II b,c,e)

Hodgkin's

(

National Cancer Institute SEER Pediatric Monograph

ICCC IILYMPHOMAS AND RETICULOENDOTHELIAL NEOPLASMS

43.2 per million for 19 year olds (Figure

II.4).

While the overall incidence of

Hodgkin’s disease for children younger than

20 years of age was similar to that for NHL

(12.1 versus 10.5 per million, respectively)

(Figure II.3), the incidence of Hodgkin’s

disease was 2-3 fold higher than NHL

among 15-19 year olds, while the incidence

of NHL was higher among those younger

than 10 years of age (Figure II.4).

Sex-specific incidence

The incidence of Hodgkin’s disease was

slightly higher among females than males

for children younger than 20 years of age

during 1990-95 (M/F = 0.9) (Figure II.5).

However, the differential by sex was age

dependent. Among children younger than

15 years of age, Hodgkin’s disease was

more common among males (M/F = 1.3),

with the differential greatest for children

younger than 5 years of age (M/F = 5.3).

However, for children diagnosed with

Hodgkin’s disease at ages 15-19 years,

Hodgkin’s disease incidence was higher

among females (M/F = 0.8).

Black-white differences in incidence

When considering children younger

than 20 years of age, black children had a

lower incidence of Hodgkin’s disease than

white children (Figure II.6). However, the

incidence was very similar for black and

white children younger than 10 years of

age. For those over 10 years of age, the

ratio of white to black incidence was ap-

proximately 1.4:1.

Figure II.5: Hodgkin's disease age-specific

and age-adjusted* incidence rates by age

and sex with male to female ratios (M/F)

all races, SEER, 1990-95

1.6

4.2

11.9

28.3

6.2

11.6

0.3

10.5

4.9

12.6

<5 5-9 10-14 15-19 <15 < 20

Age (in years) at diagnosis

Average annual rate per million

Male

Female

M/F=5.3

M/F=1.4

M/F=1.1

M/F=0.8

M/F=1.3

M/F=0.9

*Adjusted to the 1970 US standard population

Enumeration of the population at risk by single years of age was

available only for the census years 1980 and 1990. The US Bureau

of the Census provides intercensal population estimates by 5-year

age groups, but not by single years of age. Therefore, the

population estimates for 1980 were used in rate calculations for

cases diagnosed from 1976-84 and the 1990 estimates were used for

cases diagnosed from 1986-94.

W/B=1.3

Figure II.6: Hodgkin's disease age-specific and

age-adjusted* incidence rates by age with white

to black ratios (W/B) both sexes, SEER, 1990-95

4.1

36.5

6.4

13.8

4.5

9.5

26.9

10.4

<5 5-9 10-14 15-19 <15 <20

Age (in years) at diagnosis

Average annual rate per million

White

Black

W/B=0.9

W/B=1.4

W/B=1.3

*Adjusted to the 1970 US standard population

W/B=1.3

National Cancer Institute

SEER Pediatric Monograph

ICCC II LYMPHOMAS AND RETICULOENDOTHELIAL NEOPLASMS

Table II.3: Hodgkin’s and NHL age-adjusted

incidence rates

all races, both sexes, SEER, 1975-79 through 1990-95

1975-79 1980-84 1985-89 1990-95

Hodgkin’s <15 years

7.4 6.7 6.3 5.5

Males

8.4 8.0 7.5 6.2

Females

6.4 5.3 5.1 4.9

Hodgkin’s <20 years

14.5 13.8 13.1 12.1

Males

14.8 14.8 13.4 11.6

Females

14.2 12.8 12.8 12.6

NHL <15 years

8.5 8.1 8.6 8.5

Males

11.7 12.6 11.7 12.7

Females

5.2 3.4 5.3 4.2

NHL <20 years

9.1 9.7 10.0 10.5

Males

12.1 14.0 13.4 14.6

Females

5.9 5.2 6.5 6.1

Adjusted to the 1970 US standard population

TRENDS

The age-adjusted incidence rate for

Hodgkin’s disease for children younger than

20 years of age decreased from 14.5 per

million in 1975-79 to 12.1 per million in

1990-95 (Table II.3). The incidence for

children younger than 15 years of age

declined during this period from approxi-

mately 7.4 per million to 5.5 per million.

The incidence for 15-19 year olds declined

from 35.9 per million (1975-79) to 32.0 per

million (1990-95) (Figure II.7). The decline

in Hodgkin’s incidence rates for the

younger than 15 year age group occurred

for both males and females [estimated

annual percentage change (EAPC) of -2.0%

and -1.4%, respectively]. The decline in

incidence for 15-19 year olds appeared to be

greater for males (EAPC = -1.4%) than for

females (EAPC = -0.1%). A similar pattern

of declining rates for males and stable rates

for females was observed in regions of the

United Kingdom for the years 1984-93 [12].

Another report described little change or

slight decline in the occurrence of Hodgkin’s

disease among 20 populations from around

the world for the period from the early

1970s to the 1980s [13]. In contrast to

these reports, increasing incidence rates for

the 15-24 year olds were reported for males

and females in four geographical areas of

the US for the period from 1947-50 to 1984-

88 [13].

Figure II.7: Trends in Hodgkin's disease age-adjusted*

incidence rates by age, all races, both sexes, SEER, 1975-95

(

1975 1980 1985 1990 1995

Year of diagnosis

Average annual rate per million

<15 Years

15-19 Years

(

*Adjusted to the 1970 US standard population

SURVIVAL

For cases diagnosed from 1985-94, the

5-year survival rate for Hodgkin’s disease

was 91% for children younger than 20

years of age (Figure II.8). White children

had a slight survival advantage over black

children (92% versus 84% 5-year survival),

and males and females had similar out-

come (90% versus 92% 5-year survival).

Outcomes were also similar for children

diagnosed younger than 15 years of age

compared to 15-19 year olds (92% versus

90% 5-year survival).

RISK FACTORS (TABLE II.4)

Available epidemiological and molecu-

lar biological data suggest that Hodgkin’s

disease among children and adolescents

represents at least two distinctive condi-

tions [14-16]. In one condition, EBV ge-

nomic sequences are typically present in

the Reed-Sternberg cells. This form of

Hodgkin’s disease is more common for

National Cancer Institute SEER Pediatric Monograph

ICCC IILYMPHOMAS AND RETICULOENDOTHELIAL NEOPLASMS

Figure II.8: Hodgkin's disease 5 year-relative survival rates

by sex, race, age and time period

SEER (9 areas), 1975-84 and 1985-94

87 87 87 87

92 92

Total Male Female White Black <5 5-10 10-14 15-19

100

Percent surviving 5 years

1975-84

1985-94

Sex

Race Age

# - <25 cases - rate not shown

Table II.4: Current knowledge on causes of Hodgkin’s disease in children

Exposure or Characteristic Comments References

Family history of Hodgkin’s

disease

Monozygotic twins of young adult patients have a 99-fold increased risk.

Other siblings have a 7-fold increased risk.

22,33,34

Epstein-Barr virus infection EBV-associated HD is associated with mixed cellularity vs. nodular

sclerosis histologic subtypes, children from economically less-developed

vs. more-developed regions and young adult males vs. females.

Additionally, history of infectious mononucleosis and high titer

antibodies to EBV are associated with HD among young adults, although

paradoxically HD cases among the young adult population typically do

not have detectable EBV genomic sequences in tumor tissue.

7,8,16-18,20,21,35-37

Socioeconomic status For young adult disease (ages 16-44), risk increases with socioeconomic

status and with related characteristics such as small family size and

single family housing. In children younger than 10 years of age, risk

appears higher for lower socioeconomic status.

15,38-40

Social contacts For young adult disease, having fewer siblings and childhood playmates

is associated with higher risk. These findings suggest that infections in

early childhood may reduce risk of young adult disease.

19,33

Clustering Young adult cases (age 16-45 years) tend to cluster more than older cases

(age 46-70).

Known to increase risk

Factors for which evidence is

inconsistent or limited

National Cancer Institute

SEER Pediatric Monograph

ICCC II LYMPHOMAS AND RETICULOENDOTHELIAL NEOPLASMS

childhood cases occurring in developing

countries compared to cases occurring in

countries with higher socioeconomic status

[7,8]. EBV involvement is also associated

with male sex, mixed cellularity subtype,

and young age at diagnosis [7,8,16-18].

However, the associated environmental

and/or genetic factors that cause EBV

infection, which is common among young

children (particularly in developing coun-

tries), to result in Hodgkin’s disease in a

very small percentage of children are not

understood. A second form of Hodgkin’s

disease is primarily of the nodular sclerosis

subtype, arising in older adolescents and

young adults living in more affluent societ-

ies [15,16]. In spite of observations sug-

gesting a relationship between Hodgkin’s

disease among older adolescents/young

adults and infectious mononucleosis and

high titers of antibodies against EBV [19-

21], cases of Hodgkin’s disease in this

population typically do not have EBV

genomic sequences detectable in tumor

tissue [7,8].

For Hodgkin’s disease arising in young

adults, genetic susceptibility may be a

factor for some cases, based on the greatly

increased relative risk for Hodgkin’s dis-

ease occurring in both members of monozy-

gotic twin pairs compared to the risk of

occurrence in dizygotic twin pairs [22].

Further support for genetic predisposition

to Hodgkin’s disease is the consistently low

incidence of Hodgkin’s disease (especially of

the nodular sclerosis subtype) among

populations of East Asian ethnic origin and

the high incidence (especially of mixed

cellularity subtype) among some popula-

tions of South Asian origin, with these

differences appearing to be independent of

socioeconomic status [15].

National Cancer Institute SEER Pediatric Monograph

ICCC IILYMPHOMAS AND RETICULOENDOTHELIAL NEOPLASMS

(but in the jaws of young children in Af-

rica), and with lymphoblastic lymphomas

commonly present as large mediastinal

masses. The lymphoblastic lymphomas of

children are indistinguishable histologically

from T-cell acute lymphoblastic leukemia

(ALL). The clinical and biological charac-

teristics of each of these categories of

childhood NHL are summarized in Table

II.5.

Histologic subtype

Figure II.9 illustrates the distinctive

distributions by age and by sex of the NHL

subtypes that predominate in US children

in the geographic areas covered by the

SEER Program. For Burkitt’s and

Burkitt’s-like lymphoma, the incidence was

much higher for males than for females.

Five to fourteen year olds had higher rates

than children younger than 5 years of age

and adolescents 15-19 years of age. For

children younger than 20 years of age, the

incidence for the Burkitt’s lymphomas was

nearly 5.0-fold higher for males (3.2 per

million) than for females (0.7 per million).

NON-HODGKIN’S LYMPHOMA

Table II.5: Major histopathological categories of non-Hodgkin’s lymphoma in children*

Category (REAL) Category (Working

Formulation)

Clinical Presentation Chromosomal

Translocation

Genes

Affected

Burkitt’s and Burkitt-like

lymphomas

ML small noncleaved cell Intraabdominal

(sporadic), jaw (endemic)

t(8;14)(q24;q32)

t(2;8)(p11;q24)

t(8;22)(q24;q11)

C-MYC, IgH

IgK,Ig

Lymphoblastic lymphoma,

precursor T

Lymphoblastic

convoluted and

nonconvoluted

Mediastinal MTS1/p16ink4a

Deletion TAL1

t(1;14)(p34;q11)

t(11;14)(p13;q11)

TAL1, TCRad,

RHOMB1,

HOX11

Anaplastic large cell lymphoma ML immunoblastic or ML

large cell

Variable t(2;5)(p23;q35) ALK, NMP

Diffuse large cell lymphoma ML large cell Variable t(8;14)(q24;q32) in

adults, but not

well characterized

in children

BCL2, IgH (in

adults)

* Adapted from Shad and Magrath [31], and from Goldsby and Carroll [6].

The non-Hodgkin’s lymphomas that

develop in children are distinct from the

more common forms of lymphomas in

adults. While the lymphomas in adults are

more commonly of low or intermediate

grade, almost all of those that occur in

children can be classified into one of four

high-grade categories: 1) the Burkitt’s and

Burkitt-like lymphomas (small noncleaved

cell lymphomas); 2) lymphoblastic lympho-

mas; 3) diffuse large B-cell lymphomas;

and 4) anaplastic large cell lymphomas.

Each of these types of childhood NHL is

associated with distinctive molecular

biological characteristics, with transloca-

tions resulting in activation of the C-MYC

gene occurring in the Burkitt’s and Burkitt-

like lymphomas, translocations involving

the TAL1 gene and the T-cell receptor genes

typifying the lymphoblastic lymphomas,

and translocations involving the NPM gene

at chromosome band 5q35 in the anaplastic

large cell lymphomas. These NHL catego-

ries are also associated with characteristic

clinical presentations, with the Burkitt’s

and Burkitt-like lymphomas commonly

present in the abdomen among US children

National Cancer Institute

SEER Pediatric Monograph

ICCC II LYMPHOMAS AND RETICULOENDOTHELIAL NEOPLASMS

In contrast to the Burkitt’s lymphomas, the

incidence of diffuse large cell lymphoma

rose steadily with increasing age, with

males showing somewhat higher incidence

than females for the population younger

than 20 year age group (M/F = 1.4). Lym-

phoblastic lymphoma occurred at similar

frequency across all 5-year age groups, and

like the other lymphoma subtypes for

children younger than 20 years of age was

more common in males than females (M/F

= 2.5).

Age-specific incidence

The contribution of the non-Hodgkin’s

lymphomas to the overall cancer incidence

increased from 3% for children younger

than 5 years of age to 8-9% for 10-14 and

15-19 year olds (Figure II.1). Figure II.3

illustrates that the incidence of NHL varied

much less by age than did Hodgkin’s dis-

ease. The incidence of NHL rapidly in-

creased in the first three years of life,

before reaching a plateau rate of approxi-

mately 10 per million at around 4 years of

age (Figure II.4). A similar age-incidence

Figure II.10: NHL age-specific incidence

rates by sex, all races, SEER

1976-84 and 1986-94 combined

)

01234567891011121314151617181920

Age (in years) at diagnosis

Average annual rate per million

Male

Female

)

Figure II.9: NHL age-specific incidence rates by histologic group

sex, and age, all races, SEER (9 areas), 1977-95

0.6

1.9

5.5

0.5

1.2

2.5

6.1

0.6

0.7

1.4

4.9

1.3

1.6

1.8

1.6

2.2

2.8

2.2

0.9

0.7

0.9

3.6

3.5

3.2

6.1

2.8

0.8

1.1

0.8

1.2

1.9

2.5

3.6

5.6

2.3

3.3

4.3

7.8

1.5

1.6

2.8

3.4

<5 5-9 10-14 15-19 <5 5-9 10-14 15-19 <5 5-9 10-14 15-19

Average annual rate per million

Diffuse large cell

Lymphoblastic

Burkitt's

Other

Both sexes

Male

Female

National Cancer Institute SEER Pediatric Monograph

ICCC IILYMPHOMAS AND RETICULOENDOTHELIAL NEOPLASMS

pattern was observed for males and fe-

males (Figure II.10), but the incidence rate

among children older than 4 years of age

was much higher for males than for fe-

males. While incidence rates for NHL

remained fairly stable through the remain-

der of the first decade, the incidence began

increasing after age 10 (Figures II.4 and

II.10). Part of this increase in incidence

among adolescents was due to higher rates

for diffuse large cell lymphomas compared

to rates for this subtype in younger chil-

dren (Figure II.9).

Sex-specific incidence

There was a notable male predomi-

nance for NHL in children, with 70% of the

cases occurring in males (Table II.1). The

male predominance was seen for all age

groups (Figures II.10 and II.11), although

it was more pronounced for children

younger than 15 years of age (M/F = 3.0)

Figure II.11: NHL age-specific and age-adjusted*

incidence rates by age and sex with male to female

ratios (M/F), all races, SEER, 1990-95

12.2

16.1

20.5

12.7

14.7

3.6

3.9

11.9

4.2

6.1

<5 5-9 10-14 15-19 <15 <20

Age (in years) at diagnosis

Average annual rate per million

Male

Female

M/F=2.5

M/F=3.1

M/F=3.2

M/F=1.7

M/F=3.0

M/F=2.4

*Adjusted to the 1970 US standard population

Figure II.12: NHL age-specific and age-adjusted* incidence rates by age

and race with white to black ratios (W/B) both sexes, SEER, 1990-95

5.9

9.4

10.6

16.1

8.9

10.6

5.1

4.8

10.2

9.4

6.8

7.5

<5 5-9 10-14 15-19 <15 <20

Age (in years) at diagnosis

Average annual rate per million

White

Black

*Adjusted to the 1970 US standard population

W/B=2.0

W/B=1.0

W/B=1.7

W/B=1.4

W/B=1.3

W/B=1.2

National Cancer Institute

SEER Pediatric Monograph

ICCC II LYMPHOMAS AND RETICULOENDOTHELIAL NEOPLASMS

compared to 15-19 year olds (M/F = 1.7).

As noted in the preceding paragraph, the

age-incidence pattern is similar for males

and females, although incidence is higher

at all age groups for males than females

(Figure II.10).

Black-white difference in incidence

The incidence of NHL among white

children was 1.4-fold and 1.3-fold higher

than that for black children for the younger

than 15 years of age group and younger

than 20 years of age group, respectively

(Figure II.12). The difference in incidence

between white and black children appeared

greatest for children 5-9 years of age and

15-19 years of age (Figure II.12).

Trends in incidence rates

The incidence of NHL remained stable

for children younger than 15 years of age

from 1975 through 1995 (Table II.3). How-

Figure II.13: Trends in NHL age-adjusted*

incidence rates, all races

both sexes, SEER, 1975-95

(

1975 1980 1985 1990 1995

Year of diagnosis

Average annual rate per million

<15 Years

15-19 Years

(

*Adjusted to the 1970 US standard population

Figure II.14: NHL 5-year relative survival rates by sex, race, age

and time period, SEER (9 areas), 1975-84 and 1985-94

72 72 72

Total Male Female White Black <5 5-10 10-14 15-19

100

Percent surviving 5 years

1975-84

1985-94

Sex Race

Age

National Cancer Institute SEER Pediatric Monograph

ICCC IILYMPHOMAS AND RETICULOENDOTHELIAL NEOPLASMS

ever, the incidence among 15-19 year olds

increased from 10.7 per million (1975-79) to

16.3 per million (1990-95) (Figure II.13).

This increase among the older adolescents/

young adults is similar to that reported for

young adults older than 20 years of age

[13,23].

SURVIVAL

The 5-year survival rate for children

younger than 20 years of age with NHL

was 72% for the years 1985-94, a substan-

tial improvement from the 56% survival

rate for the years 1975-84 (Figure II.14).

For the most recent 10-year period, survival

was similar for white and black children

and was similar for males and females.

During this 10-year period, children

younger than 10 years of age had slightly

better 5-year survival rates than did those

10-19 years of age (approximately 76%

versus 70%).

RISK FACTORS (TABLE II.6)

The etiology of most cases of childhood

NHL is unknown. In a small proportion of

cases, childhood NHL is linked with various

disorders of immune dysfunction. Congeni-

tal immunodeficiency syndromes (e.g.,

Wiskott-Aldrich, ataxia-telangiectasia, X-

linked lymphoproliferative syndrome, and

severe combined immunodeficiency) and

acquired immunodeficiency syndrome

(AIDS) from human immunodeficiency

virus (HIV) infection are associated with an

increased risk of NHL [24-26]. Persons who

are immuno-compromised as a result of

organ and bone marrow transplantation

are also at increased risk of NHL

[24,27,28]. EBV is associated with the

endemic or ‘African-type’ Burkitt’s lympho-

mas, but much less commonly with the

sporadic Burkitt’s lymphomas in the US

[29-31]. Few epidemiologic studies of

childhood cancer have focused exclusively

on NHL, and most prenatal and perinatal

exposures evaluated to date were not

associated with increased or decreased risk

[32].

SUMMARY

Overall, age-adjusted incidence rates

for Hodgkin’s disease and NHL were simi-

lar for children and adolescents younger

than 20 years of age, although the age-

specific incidence patterns are markedly

different. Among young children, Hodgkin’s

disease is more common among males than

females, whereas for older adolescents

Hodgkin’s disease is more common among

females. Hodgkin’s disease has a high 5-

year relative survival rate, currently

greater than 90%. For unknown reasons,

Table II.6: Current knowledge on causes of non-Hodgkin’s lymphoma (NHL) in children

Exposure or Characteristic Comments References

Immunodeficiency Immunosuppressive therapy, congenital immunodeficiency syndromes (e.g.,

ataxia telangiectasia), acquired immunodeficiency syndrome (AIDS) all

predispose to NHL.

24-27,42,43

Epstein-Barr virus EBV is associated with ‘African-type’ Burkitt’s lymphoma, and chronic

immune suppression due to malaria may be a co-factor in this situation. EBV

is also associated with NHL in patients with immunodeficiency.

27,29-31,44,45

Radiation While a few studies report increased NHL risk in adults or children with

ionizing or electromagnetic field (EMF) radiation, others report no association.

46-50

Known risk factors

Substantial evidence

implicating factor

Factors for which evidence is

inconsistent or limited

National Cancer Institute

SEER Pediatric Monograph

ICCC II LYMPHOMAS AND RETICULOENDOTHELIAL NEOPLASMS

the incidence of Hodgkin’s disease appears

to be slowly decreasing for both younger

children and for older adolescents. Sub-

stantial evidence implicates EBV in the

etiology of a subset of Hodgkin’s disease

(primarily mixed cellularity subtype), but

the mechanism by which EBV results in

development of Hodgkin’s disease and the

potential role of co-factors are not under-

stood.

The non-Hodgkin’s lymphomas of

children are a heterogeneous group of

tumors, with Burkitt’s and Burkitt-like

tumors predominating among 5-14 year

olds, and with diffuse large cell lymphomas

being the most common subtype among 15-

19 year olds. Particularly for the Burkitt’s

lymphomas, there is a marked sex differen-

tial, with males having a much higher

incidence than females. Survival rates

have improved substantially for NHL

between the late 1970s and the late 1980s

are now over 70%, with similar rates for

both sexes and for whites and blacks. The

incidence of NHL among children younger

than 15 years of age appears fairly con-

stant over the past 21 years, while there

appears to have been a slight increase in

incidence for the 15-19 year old age group.

The etiology of childhood NHL is poorly

understood, although a small proportion of

cases arise in children with congenital or

acquired severe immune dysfunction.

Reference List

1. Parkin D, Stiller C, Draper G, et al: Interna-

tional Incidence of Childhood Cancer. Lyon:

World Health Organization, 1988.

2. Kuppers R, Hansmann ML, Rajewsky K:

Clonality and germinal centre B-cell derivation

of Hodgkin/Reed- Sternberg cells in Hodgkin’s

disease. Ann Oncol 9:S17-20, 1998.

3. Izban KF, Nawrocki JF, Alkan S, et al: Mono-

clonal IgH gene rearrangement in

microdissected nodules from nodular sclerosis

Hodgkin disease. Am J Clin Pathol 110:599-

606, 1998.

4. Vockerodt M, Soares M, Kanzler H, et al:

Detection of clonal Hodgkin and Reed-

Sternberg cells with identical somatically

mutated and rearranged VH genes in different

biopsies in relapsed Hodgkin’s disease. Blood

92:2899-907, 1998.

5. Irsch J, Nitsch S, Hansmann ML, et al: Isola-

tion of viable Hodgkin and Reed-Sternberg

cells from Hodgkin disease tissues. Proc Natl

Acad Sci U S A 95:10117-22, 1998.

6. Goldsby RE, Carroll WL: The molecular biology

of pediatric lymphomas. J Pediatr Hematol

Oncol 20:282-96, 1998.

7. Sleckman BG, Mauch PM, Ambinder RF, et al:

Epstein-Barr virus in Hodgkin’s disease:

correlation of risk factors and disease charac-

teristics with molecular evidence of viral

infection. Cancer Epidemiol Biomarkers Prev

7:1117-21, 1998.

8. Glaser SL, Lin RJ, Stewart SL, et al: Epstein-

Barr virus-associated Hodgkin’s disease:

epidemiologic characteristics in international

data. Int J Cancer 70:375-82, 1997.

9. Hudson M, Donaldson S: Hodgkin’s disease. In

Principles and Practice of Pediatric Oncology

(Pizzo P, Poplack D, eds). Philadelphia:

Lippincott-Raven Publishers, 1997, pp 523-543.

10. Harris NL: The many faces of Hodgkin’s

disease around the world: what have we

learned from its pathology? Ann Oncol 9:S45-

56, 1998.

11. Franklin J, Tesch H, Hansmann ML, et al:

Lymphocyte predominant Hodgkin’s disease:

pathology and clinical implication. Ann Oncol

9:S39-44, 1998.

12. Cartwright R, McNally R, Roman E, et al:

Incidence and time trends in Hodgkin’s

disease: from parts of the United Kingdom

(1984-1993) [In Process Citation]. Leuk

Lymphoma 31:367-77, 1998.

13. Hartge P, Devesa SS, Fraumeni JF, Jr.:

Hodgkin’s and non-Hodgkin’s lymphomas.

Cancer Surv 20:423-53, 1994.

14. MacMahon B: Epidemiology of Hodgkin’s

disease. Cancer Res 26:1189-201, 1966.

15. Stiller CA: What causes Hodgkin’s disease in

children? Eur J Cancer 34:523-8, 1998.

16. Armstrong AA, Alexander FE, Cartwright R, et

al: Epstein-Barr virus and Hodgkin’s disease:

further evidence for the three disease hypoth-

esis. Leukemia 12:1272-6, 1998.

17. Armstrong AA, Alexander FE, Paes RP, et al:

Association of Epstein-Barr virus with pediat-

ric Hodgkin’s disease. Am J Pathol 142:1683-8,

1993.

18. Kusuda M, Toriyama K, Kamidigo NO, et al: A

comparison of epidemiologic, histologic, and

virologic studies on Hodgkin’s disease in

western Kenya and Nagasaki, Japan. Am J

Trop Med Hyg 59:801-7, 1998.

19. Gutensohn N, Cole P: Childhood social environ-

ment and Hodgkin’s disease. N Engl J Med

304:135-40, 1981.

National Cancer Institute SEER Pediatric Monograph

ICCC IILYMPHOMAS AND RETICULOENDOTHELIAL NEOPLASMS

20. Evans AS, Gutensohn NM: A population-based

case-control study of EBV and other viral

antibodies among persons with Hodgkin’s

disease and their siblings. Int J Cancer 34:149-

57, 1984.

21. Mueller N, Evans A, Harris NL, et al:

Hodgkin’s disease and Epstein-Barr virus.

Altered antibody pattern before diagnosis. N

Engl J Med 320:689-95, 1989.

22. Mack TM, Cozen W, Shibata DK, et al: Concor-

dance for Hodgkin’s disease in identical twins

suggesting genetic susceptibility to the young-

adult form of the disease [see comments]. N

Engl J Med 332:413-8, 1995.

23. Devesa SS, Blot WJ, Stone BJ, et al: Recent

cancer trends in the United States. J Natl

Cancer Inst 87:175-82, 1995.

24. Kersey JH, Shapiro RS, Filipovich AH: Rela-

tionship of immunodeficiency to lymphoid

malignancy. Pediatr Infect Dis J 7:S10-2, 1988.

25. Filipovich AH, Mathur A, Kamat D, et al:

Primary immunodeficiencies: genetic risk

factors for lymphoma. Cancer Res 52:5465s-

5467s, 1992.

26. McClain KL, Joshi VV, Murphy SB: Cancers in

children with HIV infection. Hematol Oncol

Clin North Am 10:1189-201, 1996.

27. Hanto DW, Najarian JS: Advances in the

diagnosis and treatment of EBV-associated

lymphoproliferative diseases in

immunocompromised hosts. J Surg Oncol

30:215-20, 1985.

28. Penn I: The changing pattern of posttransplant

malignancies. Transplant Proc 23:1101-3, 1991.

29. de-The G, Geser A, Day NE, et al: Epidemio-

logical evidence for causal relationship be-

tween Epstein-Barr virus and Burkitt’s

lymphoma from Ugandan prospective study.

Nature 274:756-61, 1978.

30. Gutierrez MI, Bhatia K, Barriga F, et al:

Molecular epidemiology of Burkitt’s lymphoma

from South America: differences in breakpoint

location and Epstein-Barr virus association

from tumors in other world regions. Blood

79:3261-6, 1992.

31. Shad A, Magrath I: Malignant non-Hodgkin’s

lymphomas in children. In Principles and

Practice of Pediatric Oncology (Pizzo P, Poplack

D, eds). Philadelphia: Lippincott-Raven

Publishers, 1997, pp 545-587.

32. Adami J, Glimelius B, Cnattingius S, et al:

Maternal and perinatal factors associated with

non-Hodgkin’s lymphoma among children. Int

J Cancer 65:774-7, 1996.

33. Grufferman S, Cole P, Smith PG, et al:

Hodgkin’s disease in siblings. N Engl J Med

296:248-50, 1977.

34. Ferraris AM, Racchi O, Rapezzi D, et al:

Familial Hodgkin’s disease: a disease of young

adulthood? Ann Hematol 74:131-4, 1997.

35. Ambinder R, Browning P, Lorenzana I, et al:

Epstein-Barr virus and childhood Hodgkin’s

disease in Honduras and the United States.

Blood 81:462-7, 1993.

36. Chang KL, Albujar PF, Chen YY, et al: High

prevalence of Epstein-Barr virus in the Reed-

Sternberg cells of Hodgkin’s disease occurring

in Peru. Blood 81:496-501, 1993.

37. Pallesen G, Hamilton-Dutoit SJ, Rowe M, et al:

Expression of Epstein-Barr virus latent gene

products in tumour cells of Hodgkin’s disease

[see comments]. Lancet 337:320-2, 1991.

38. Grufferman S, Delzell E: Epidemiology of

Hodgkin’s disease. Epidemiol Rev 6:76-106,

1984.

39. Gutensohn NM: Social class and age at

diagnosis of Hodgkin’s disease: new epidemio-

logic evidence for the “two-disease hypothesis”.

Cancer Treat Rep 66:689-95, 1982.

40. Gutensohn NM, Shapiro DS: Social class risk

factors among children with Hodgkin’s disease.

Int J Cancer 30:433-5, 1982.

41. Greenberg RS, Grufferman S, Cole P: An

evaluation of space-time clustering in Hodgkin’s

disease. J Chronic Dis 36:257-62, 1983.

42. Spector BD, Perry GSd, Kersey JH: Genetically

determined immunodeficiency diseases (GDID)

and malignancy: report from the immunodefi-

ciency—cancer registry. Clin Immunol

Immunopathol 11:12-29, 1978.

43. Taylor AM, Metcalfe JA, Thick J, et al: Leuke-

mia and lymphoma in ataxia telangiectasia.

Blood 87:423-38, 1996.

44. Niedobitek G, Young LS, Herbst H: Epstein-

Barr virus infection and the pathogenesis of

malignant lymphomas. Cancer Surv 30:143-62,

1997.

45. McClain K, Wu W-S, Joshi VV, et al: Lympho-

mas and B-cell leukemias of children with

AIDS are infrequently associated with muta-

tions or translocations of c-myc or with Epstein

Barr virus. Blood 88:A1505, 1996.

46. Boice JD, Jr.: Radiation and non-Hodgkin’s

lymphoma. Cancer Res 52:5489s-5491s, 1992.

47. Pearce N, Bethwaite P: Increasing incidence of

non-Hodgkin’s lymphoma: occupational and

environmental factors. Cancer Res 52:5496s-

5500s, 1992.

48. Tynes T, Haldorsen T: Electromagnetic fields

and cancer in children residing near Norwe-

gian high-voltage power lines. Am J Epidemiol

145:219-26, 1997.

49. Verkasalo PK, Pukkala E, Hongisto MY, et al:

Risk of cancer in Finnish children living close

to power lines [see comments]. BMJ 307:895-9,

1993.

50. Olsen JH, Nielsen A, Schulgen G: Residence

near high voltage facilities and risk of cancer

in children [see comments]. BMJ 307:891-5.

National Cancer Institute

SEER Pediatric Monograph

ICCC IIICNS AND MISCELLANEOUS INTRACRANIAL AND INTRASPINAL NEOPLASMS

National Cancer Institute

SEER Pediatric Monograph

HIGHLIGHTS

Incidence

♦ The CNS malignancies represented 16.6% of all malignancies during childhood

(including adolescence). CNS cancer as a group was the second most frequent

malignancy of childhood and the most common of the solid tumors. In the US

approximately 2,200 children younger than 20 years of age are diagnosed annually

with invasive CNS tumors.

♦ Astrocytomas accounted for 52% of CNS malignancies, PNET comprised 21%,

other gliomas 15% and ependymomas an additional 9% (Figure III.1).

♦ Unlike adults and older children, young children have a relatively high occurrence

of malignancies in the cerebellum and the brain stem. In fact, in children younger

than 10 years of age, brain stem malignancies were nearly as common as cerebral

malignancies, and cerebellum malignancies were far more common than cerebral

malignancies (Figure III.2).

♦ The incidence of invasive CNS tumors was higher in males than females and

higher among white children than black children (Figure III.5).

♦ The average annual incidence of CNS cancer varied only slightly by age of diagno-

sis from infancy (36.2 per million) through age 7 years (35.2 per million). From

age 7 to 10, a 40% drop in the incidence rate (to 21.0 per million) was observed.

CNS cancer rates were fairly consistent among children aged 11 through 17 years,

until another substantial decrease occurred at age 18 (Figure III.6).

♦ The increase in CNS cancer rates in the past two decades has been the subject of

numerous reports. One concern is that changes in environmental exposures may

be responsible for the increasing incidence rates, although epidemiologic evidence

to support this hypothesis currently is lacking. An alternative explanation is that

improvements in diagnostic technology and case ascertainment may be

contributing to the increasing trend.

Survival

♦ In general, children with CNS cancer do not share the favorable prognosis of those

with many other common pediatric neoplasms.

♦ Very young children with CNS cancer, especially infants with ependymoma or

PNET, had low survival rates (Table III.2).

Risk factors

♦ There is no specific risk factor that explains a substantial proportion of brain

tumor occurrence, but there are a couple of factors that explain a small proportion

(Table III.3).

INTRODUCTION

Since most of the neoplasms described

in this chapter are in the central nervous

system, the abbreviation CNS will be used

to refer to neoplasms that originate in the

brain, other intracranial sites such as the

pituitary or pineal glands, and the spinal

cord. In the US, approximately 2,200

children and adolescents younger than 20

years of age are diagnosed with malignant

central nervous system tumors each year.

Over 90 percent of primary CNS malignan-

cies in children are located within the

James G. Gurney, Malcolm A. Smith, Greta R. Bunin

ICCC III CNS

National Cancer Institute

SEER Pediatric Monograph

rates for the CNS germ cell malignancies

from 1990-95 were 0.2 per million children

younger than 15 years of age, and 1.9 per

million children younger than 20 years of

age. Fifty-three additional tumors were

excluded because they occurred outside the

brain, intracranium and spinal cord.

It also should be noted that data

reported here are comprised solely of CNS

tumors that are classified as primary and

malignant. Primary CNS neoplasms are

tumors that originated in the central

nervous system. Thus, they exclude cancer

that developed in some other location in the

body and then spread to the CNS. Like-

wise, CNS tumors classified as benign or

with uncertain behavior (nonmalignancies)

are not routinely collected by SEER areas,

and thus are not included in this report.

The pathological distinction between malig-

nant and nonmalignant tumors of the CNS

is not always consistent with clinical behav-

ior, particularly for intracranial tumors.

Depending on the location and the size of

the tumor, some intracranial tumors that

are classified as benign can have a destruc-

tive clinical course (eg. craniopharyngioma).

In contrast, some tumors classified as

malignant may require no treatment and

have little clinical significance (eg. pilocytic

astrocytomas of the optic pathway). Al-

though all central registries will include

malignant neoplasms in their case ascer-

tainment, when comparing CNS incidence

rates across cancer surveillance systems it

is necessary to determine whether a given

registry also includes nonmalignant tu-

mors. An analysis of data from the Central

Brain Tumor Registry of the United States

(a compilation of data from population-

based registries that include case ascertain-

ment of nonmalignant CNS tumors)

showed that the incidence of only malig-

nant CNS tumors underestimates the

incidence of both malignant and non-

malignant CNS tumors by approximately

28% [4].

brain. This report only includes malignant

CNS tumors.

Classification system

CNS tumors are heterogeneous in

regards to histology and clinical course.

Because of the many relatively similar

histopathological types and their rarity, it is

necessary for epidemiologic purposes to

group CNS tumors into rather broad histo-

logic categories. There are several classifi-

cation systems that are used for describing

CNS tumors and no system has yet

emerged as the definitive gold standard

[1,2]. For most of this monograph, malig-

nancies are grouped according to the Inter-

national Classification of Childhood Cancer

(ICCC) system [3]. There are a few minor

discrepancies within the ICCC system for

CNS tumors, however, that somewhat

compromise accurate comparisons with

other published work. Most notable, intrac-

ranial neuroblastoma and pineoblastoma,

which, along with medulloblastoma are

generally considered primitive neuroecto-

dermal tumors (PNET), are not included

with the PNET category of the ICCC for

CNS. For the descriptive analysis that

follows, we modified the ICCC groupings for

CNS tumors in the following manner:

“Other specified intracranial and intraspi-

nal neoplasms excluding pineoblastoma

(IIIe)” and “Unspecified intracranial and

intraspinal neoplasms (IIIf)” were com-

bined into one category, called ‘other CNS’;

the “Ependymoma (IIIa)” category was not

changed; the “PNET (IIIc)” category was

expanded to include intracranial neuroblas-

toma (these were also reported with ICCC

IV) and pineoblastoma. Finally, the ICCC

system places intracranial and intraspinal

germ cell malignancies within the germ cell

category, rather than the CNS tumor

category. We chose to follow the ICCC

system for CNS germ cell tumors, thus we

did not include intracranial and intraspinal

germ cell tumors in this chapter (see ICCC

group X). The average annual incidence

ICCC IIICNS

National Cancer Institute SEER Pediatric Monograph

INCIDENCE

Unless otherwise indicated, the discus-

sion on incidence that follows will pertain

to children younger than 20 years of age

and only malignant tumors. For the 21-

year period of 1975-95, there were 4,945

primary malignant tumors of the CNS

diagnosed among children in SEER areas.

This represented 16.6% of all malignancies

during childhood (including adolescence).

CNS cancer as a group was the second most

frequent malignancy of childhood and the

most common of the solid tumors. Astrocy-

tomas accounted for 52% of CNS malignan-

cies, PNET comprised 21%, other gliomas

15%, and ependymomas an additional 9%

(Figure III.1).

The incidence rates by location within

the brain and other CNS sites as a function

of age are shown in Figure III.2. Unlike

adults and older children, who have higher

rates in the cerebrum, young children have

a relatively high occurrence of malignancies

in the cerebellum and the brain stem. In

fact, in children between the ages of 5 and

9, brain stem malignancies were nearly as

common as cerebral malignancies, and

cerebellum malignancies were far more

common than cerebral malignancies. The

pattern shifted among children between the

ages of 10-19, in that the incidence of both

brain stem and cerebellar cancers de-

creased while cerebral malignancies in-

creased slightly. The “other” brain site

group included the ventricles, where

ependymomas generally develop, and

malignancies with brain sites not otherwise

specified. The “Other CNS” category in-

cludes malignancies of the meninges,

cranial nerves and spinal cord.

Figure III.1: Percent distribution of malignant CNS

tumors by age and histologic group, all races

both sexes, SEER, 1975-95

49.6

22.9

15.4

9.3

2.7

52.2

20.8

15.5

8.6

Astrocytomas

PNET

Other gliomas

Ependymomas

Other CNS

010203040506070 0 10203040506070

<15 years

<20 years

Percent of total CNS cancer

Figure III.2: Malignant CNS tumor age-specific

incidence rates by anatomic site and age

all races, both sexes, SEER, 1975-95

4.7

5.9

2.8

1.7

5.8

5.9

6.8

9.3

9.7

5.7

3.7

8.8

5.9

4.5

2.6

1.9

1.7

1.6

5-9

10-14

15-19

Age (in years) at diagnosis

0123456789101112

Average annual rate per million

Brain Stem

Cerebrum

Cerebellum

Other Brain

Other CNS

ICCC III CNS

National Cancer Institute

SEER Pediatric Monograph

Age-specific incidence

Incidence rates by single year of age

are presented in Figure III.3.

The average

annual incidence of CNS cancer varied only

slightly by age of diagnosis from infancy

(36.2 per million) through age 7 years (35.2

per million). From age 7 to 10, a 40% drop

in the incidence rate (to 21.0 per million)

was observed. CNS cancer rates were

fairly consistent among children aged 11

through 17 years, until another substantial

decrease occurred at age 18.

The incidence of astrocytomas peaked

at age 5 (20.7 per million) and a second

peak occurred at age 13 (19.7 per million).

PNET rates were fairly steady from infancy

through age 3 years (ranging from 11.6 to

10.2 per million) and then steadily declined

thereafter. Rates of ependymomas were

highest through age 3 years, with the age

of peak incidence occurring during the

second year of life (8.6 per million). Among

children aged 5-14, ependymomas are very

rare, averaging only 1.4 per million.

Although in our data the age-specific

rates for black children were fairly unstable

because of small numbers of cases (295

cases from 1986-94), the greatest difference

in rates between whites and blacks was

observed during the first year of life (47.8

vs. 18.7 per million, respectively) (Figure

III.4). In the second year of life, rates

among whites decreased from the first year,

Figure III.3: Malignant CNS tumor age-specific

incidence rates, all races, both sexes

SEER, 1986-94

)

(

((

(

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Age (in years) at diagnosis

Average annual rate per million

All CNS

Astrocytomas

PNET

Other gliomas

Ependymoma

(

)

Figure III.4: Malignant CNS tumor age-specific

incidence rates by race, both sexes

SEER, 1986-94

01234567891011121314151617181920

Age (in years) at diagnosis

Average annual rate per million

White

Black

Enumeration of the population at risk by single years of age was

available only for the census years 1980 and 1990. The US Bureau

of the Census provides intercensal population estimates by 5-year

age groups, but not by single years of age. Therefore, the

population estimates for 1980 were used in rate calculations for

cases diagnosed from 1976-84 and the 1990 estimates were used for

cases diagnosed from 1986-94.

ICCC IIICNS

National Cancer Institute SEER Pediatric Monograph

while rates in blacks increased substan-

tially. To a degree, this could suggest a

pattern in which whites were diagnosed

earlier than blacks (on average) for the

CNS malignancies that occur early in life,

although we are aware of no other evidence

that supports this speculation.

Sex-specific incidence

As will be discussed below, brain cancer

incidence rates in children have increased

in SEER areas over the past 2 decades. For

this reason, the following CNS cancer

incidence rates are reported for the time

period 1990-95, rather than 1975-95, to

reflect recent patterns. The rates that

follow were adjusted to the 1970 US stan-

dard million population. The incidence rate

of primary CNS malignancies was 27.2 per

million children younger than 20 years of

age (if intracranial germ cell malignancies

are included, the rate was 29.1 per million).

Males (30.0 per million) had a 24% higher

incidence rate relative to females (24.2 per

million). Figures III.5 and III.6 illustrate

the sex-specific rates by histologic groups of

children younger than 20 years of age and

younger than 15 years of age, respectively.

A clear male preponderance for both PNET

and ependymomas was evident, but rates

for males and females were similar for the

other histologic groups.

Black-white differences in incidence

White children (28.5 per million) had

an 18% higher average CNS incidence rate

compared with black children (24.2 per

million). Figure III.7 depicts overall inci-

Figure III.5: Malignant CNS tumor age-adjusted*

incidence rates by histologic group and sex

age <20, all races, SEER, 1990-95

14.8

7.3

4.5

0.5

24.2

13.5

4.2

4.4

1.5

0.6

All CNS

Astrocytomas

PNET

Other gliomas

Ependymomas

Other CNS

0 4 8 12162024283236

Average annual rate per million

Males

Females

*Adjusted to the 1970 US standard population

32.7

15.7

8.6

4.5

3.5

0.5

26.8

14.5

1.8

0.4

All CNS

Astrocytomas

PNET

Other gliomas

Ependymomas

Other CNS

0 4 8 12162024283236

Average annual rate per million

Males

Females

*Adjusted to the 1970 US standard population

Figure III.6: Malignant CNS tumor age-adjusted*

incidence rates by histologic group and sex

age <15, all races, SEER, 1990-95

ICCC III CNS

National Cancer Institute

SEER Pediatric Monograph

dence rates by sex for white children, black

children, and all children combined. It is

evident that the racial difference in CNS

rates was primarily concentrated among

males. There was only a slightly higher

CNS cancer incidence rate among white

compared with black females (8%), while

the racial difference in rates for males was

somewhat more pronounced (26%).

TRENDS

The observation that CNS cancer

incidence in children appears to have

increased in the past two decades has been

the subject of numerous previous reports

[5-8]. There is considerable debate regard-

ing the possible reasons for the apparent

trend. One concern is that changes in

environmental exposures may be respon-

sible for the increasing incidence, although

epidemiologic evidence to support this

hypothesis currently is lacking [9]. An

alternative explanation is that changes in

reporting due to improvements in diagnos-

tic technology and case ascertainment may

be contributing to the increasing trend.

Figure III.8 illustrates the increase in

incidence rates of CNS cancer for the years

1975-95 for children younger than 15 years

of age. Based on a model using a constant

rate of increase in incidence over this

period, the estimated annual percentage

change (EAPC) was +1.5% (continuous

green line in Figure III.8). Smith et al [5]

recently evaluated CNS trends for children

in the United States from SEER data using

a more sophisticated statistical modeling

technique. They demonstrated that the

incidence of CNS malignancies did not

increase steadily from 1973 to 1994, but

rather “jumped” to a steady, but higher rate

after 1984-85. When the same methodol-

ogy was applied to the younger than 15

year old age group described in this chapter

for the years 1975 to 1995, this “jump

model”, with the optimal change point from

lower to higher incidence occurring after

1985, produced a significantly better fit

than the model using a constant linear rate

Figure III.7: Malignant CNS tumor age-adjusted*

incidence rates by race and sex

age <20, all races, SEER, 1990-95

31.5

24.2

25.3

23.4

All Races

White

Black

0 4 8 1216202428323640

Average annual rate per million

Males

Females

*Adjusted to the 1970 US standard population

Figure III.8: Temporal trends in malignant CNS tumor

age-adjusted* incidence rates, age <15

all races, both sexes, SEER, 1975-95

)

75 77 79 81 83 85 87 89 91 93 95

Year of diagnosis

Average annual rate per million

Incidence

1975-95

1975-85

1986-95

)

*Adjusted to the 1970 US standard population

ICCC IIICNS

National Cancer Institute SEER Pediatric Monograph

of increase (p = 0.003). The EAPC from

1975-84 was –0.1% (blue line in Figure

III.8) and for 1986-95 the EAPC was also

–0.1% (red line in Figure III.8). The timing

of the jump in incidence is coincident with

the wide-scale availability of magnetic

resonance imaging (MRI) in the United

States [5]. This observation, combined with

the absence of any jump in CNS cancer

mortality during the same period, lends

support to the contention that improved

diagnosis and reporting during the 1980’s is

largely responsible for the temporal trends

in CNS incidence rates that have been

observed since the 1970s. Whether the

relatively stable rates of childhood CNS

cancer observed over the past decade in the

US will continue, however, remains to be

seen.

SURVIVAL

Although survival differs by histology,

behavior, size and location of the malig-

nancy, in general children with CNS cancer

do not share the favorable prognosis of

those with many other common pediatric

neoplasms, such as acute lymphoblastic

leukemia. Additionally, for children who do

survive CNS cancer, long term morbidity

can be substantial. Table III.1 provides 5-

year relative survival probabilities by

histologic group within 2 time periods.

Survival probability improved

somewhat over the two time periods. Nev-

ertheless, other than astrocytomas, many of

which were low grade malignancies such as

Figure III.9: Total malignant CNS tumor 5-year relative

survival rates by sex, race, age and time period

SEER (9 areas), 1975-84 and 1985-94

62 62

Total Male Female White Black <5 5-9 10-14 15-19

100

Percent surviving 5 years

1975-84

1985-94

Sex Race

Age

juvenile pilocytic astrocytomas, survival

probability was less than 60%. While there

were only minimal differences in survival of

CNS cancer by sex and race, age was an

important factor. Table III.2 provides 5-

year relative survival for 1986-94 according

to age and histologic groups.

For all CNS cancer combined, survival

probability increased with increasing age.

Very young children with CNS cancer,

especially infants with ependymoma or

PNET, were at particularly high risk of

Table III.2: 5-year relative survival rates for

CNS cancer by type and age group

all races, both sexes, SEER, 1986-94

ICCC Group <1 1-4 5-9 10-14 15-19

All CNS

Cancer

45% 59% 64% 70% 77%

Astrocytoma

69 79 70 75 75

Other Glioma * 51436479

Ependymoma

25 46 71 76 *

PNET

19 46 69 57 75

* less than 20 cases.

Table III.1: 5-year relative survival rates for

CNS by type and time period

age <20, all races, both sexes

SEER 1975-84 and 1985-94

ICCC Group 1975-84 1985-94

All CNS Cancer 60% 65%

Astrocytoma 70 74

Other Glioma 47 57

Ependymoma 39 56

PNET 52 55

ICCC III CNS

National Cancer Institute

SEER Pediatric Monograph

Figure III.11: Astrocytoma 5-year relative survival rates

by sex, race, age and time period, SEER (9 areas) 1975-84 and 1985-94

70 70 70

75 75

Total Male Female White Black <5 5-9 10-14 15-19

100

Percent surviving 5 years

1975-84

1985-94

Sex

Race

Age

Figure III.10: Ependymoma 5-year relative survival rates

by sex, race, age and time period, SEER (9 areas), 1975-84 and 1985-94

Total Male Female White Black <5 5-9 10-14 15-19

100

Percent surviving 5 years

1975-84

1985-94

Sex

Race

Age

# - <25 cases - rate not shown

###

ICCC IIICNS

National Cancer Institute SEER Pediatric Monograph

Figure III.12: PNET 5-year relative survival rates

by sex, race, age and time period, SEER (9 areas), 1975-84 and 1985-94

48 48

Total Male Female White Black <5 5-9 10-14 15-19

100

Percent surviving 5 years

1975-84

1985-94

Sex

Race

Age

Figure III.13: Other gliomas 5-year relative survival rates

by sex, race, age and time period, SEER (9 areas), 1975-84 and 1985-94

Total Male Female White Black <5 5-9 10-14 15-19

100

Percent surviving 5 years

1975-84

1985-94

Sex

Race

Age

ICCC III CNS

National Cancer Institute

SEER Pediatric Monograph

Table III.3: Current knowledge on causes of childhood brain tumors

Exposure or Characteristic Comments References

Sex Incidence of medulloblastoma and ependymomas in males is higher than

in females. For other types of brain tumors, there is little difference

between males and females.

Therapeutic doses of ionizing

radiation to head

Children treated for tinea capitis experienced 2.5-6-fold increased risk.

Currently, those at risk are children treated with radiation to the head for

leukemia or a previous brain tumor.

11,12

Neurofibromatosis, tuberous

sclerosis, nevoid basal cell

syndrome, Turcot syndrome, Li-

Fraumeni syndrome

Children with these genetic conditions have a greatly increased risk of

brain tumors, for example, 50-fold for neurofibromatosis and 70-fold for

tuberous sclerosis. Together, these conditions account for less than 5% of

all childhood brain tumors.

10,13,14,28

Maternal diet during pregnancy Frequent cured meat consumption has been consistently associated with a

1.5-2.0 fold increased risk. However, it is unclear whether cured meats or

another dietary factor are responsible, since most aspects of diet have not

yet been studied.

10,13,15-17

Parent or sibling with brain

tumor

Having a sibling or parent with a brain tumor has usually been associated

with a 3-9 fold increased risk. It may be that the excess risk is explained

completely by the specific genetic conditions listed above.

10,13,17,18

Family history of bone cancer,

leukemia or lymphoma.

The increased risk seen in some studies may be explained by the Li-

Fraumeni syndrome.

10,13,22,23,

Electromagnetic fields A small increase in risk has been observed in some studies, but not most. 10,13,19,29,

Products containing N-nitroso

compounds: beer, incense,

make-up, antihistamines,

diuretics, rubber baby bottle and

pacifier nipples

The data are inconsistent; associations seen in one study have generally

not been reported in later studies.

10,13,21

Father’s occupation and related

exposures

Many associations have been reported, but few have been replicated:

aircraft industry, agriculture, electronics mfg., petroleum industry,

painter, paper or pulp mill worker, printer, metal-related occupation,

exposure to paint, ionizing radiation, solvents, electromagnetic fields.

10,13,25

Pesticides There has been little focused research on this topic. Two small studies

suggest an association with use of no-pest strips.

10,13,20,31

History of head injury This is difficult to study because of the rarity of serious head injury and

the possibility that mothers of children with brain tumors are more likely

than control mothers to recall minor head injuries.

10,13,26

Family history of epilepsy or

seizures

The data are inconsistent. One study suggests that the effect of family

history of seizures may differ by type of brain tumor and/or type and

circumstances of seizures.

13,18,27

Family history of mental

retardation

Increased risk observed in one study of adults and one of children. 13

Note that the majority of these risk factors have been reviewed recently in references 10 and 13; only selected

references are presented for additional reading.

Factors for which evidence

is suggestive but not conclusive

Known risk factors

Factors for which evidence

is inconsistent or limited

ICCC IIICNS

National Cancer Institute SEER Pediatric Monograph

mortality. Relative to younger children,

adolescents with CNS cancer tended to fare

well (Figures III.9-III.13).

RISK FACTORS

Table III.3 presents a general sum-

mary of the current knowledge on causes of

brain cancer in children. To date, there is

no specific risk factor known to explain a

substantial proportion of brain tumor

occurrence. Some hereditary conditions

that are clearly associated with increased

susceptibility to CNS cancer in children

include neurofibromatosis type 1, nevoid

basal cell syndrome, and tuberous sclerosis.

These diseases are rare, however, and not

all children with genetic predispositions go

on to acquire cancer. Although a somewhat

increased risk has been observed when a

sibling or parent has had a brain tumor, the

association with family history is not strong

or consistent. Thus, from a population

perspective, known inherited genetic factors

explain only a small percentage of child-

hood CNS cancer incidence. The same can

be said for many other exposures that have

been studied. While therapeutic doses of

ionizing radiation to the head are defini-

tively known to increase the risk of brain

tumors in children, this exposure is largely

historical in nature because therapeutic

head x-rays are now used very sparingly

and with much greater caution than in the

past. There is some evidence that certain

dietary components during pregnancy may

either raise or lower risk, but the relevant

aspects have not yet been clarified. For

exposures with inconsistent or limited data

that are listed in the table, it is not yet

possible to say whether they influence risk.

We know a few factors that do not appear

to increase a child’s risk of developing a

brain tumor, including passive cigarette

smoke exposure, electric blanket use, and

ultrasound testing during pregnancy. The

difficulty in identifying CNS cancer risk

factors may stem in part from studying all

childhood brain tumors as a single entity

when many different histologic subtypes

occur. The rarity of any specific histologic

type makes it very difficult to accrue

enough cases for epidemiologic study.

SUMMARY

Cancer of the brain and central ner-

vous system comprises nearly 17% of

malignancies in children younger than 20

years of age. As a group, CNS cancer is the

most common solid tumor and the second

most common malignancy of childhood.

The overall annual incidence in the United

States is about 27 per million children

younger than 20 years of age (29 per mil-

lion with intracranial germ cell malignan-

cies included). The incidence of CNS cancer

is higher in children younger than 8 years

of age than in older children or adolescents.

This difference is largely attributable to

cerebellar PNET (medulloblastoma), brain

stem gliomas and ependymomas, which all

occur primarily before the age of 10 years.

CNS cancer incidence is slightly higher in

males than in females, largely due to the

male predominance of PNET and ependy-

momas. Rates are higher in white children

than in black children, although the differ-

ences are seen primarily in males and in

young children. Survival, which is depen-

dent on the type and location of the CNS

malignancy, tends to be worse in very

young children than in older children. CNS

cancer incidence rates remained essentially

stable from 1986-95. Unfortunately, the

causes of CNS cancer remain largely unde-

termined. The few definitive risk factors

that are known explain only a small propor-

tion of the total case population.

Reference List

1. Heideman RL, Packer RJ, Albright LA, Free-

man CR, Rorke LB; Tumors of the central

nervous system. Pizzo P, Poplack D, Editors.

Principles and Practices of Pediatric Oncology.

3rd ed. Philadelphia, PA: Lippincott-Raven;

1997:633-697.

2. Kleihues P, Burger PC, Scheithauer BW. The

ICCC III CNS

National Cancer Institute

SEER Pediatric Monograph

new WHO classification of brain tumours.

Brain Path. 1993;3:255-268.

3. Kramarova E, Stiller CA. The international

classification of childhood cancer. Int J Cancer

1996;68:759-765.

4. Gurney JG, Wall DA, Jukich PJ, Davis FG. The

contribution of nonmalignant tumors to CNS

tumor incidence rates among children in the

United States. Cancer Causes Control

1999;10:101-105.

5. Smith MA, Feidlin B, Ries LAG, Simon R.

Trends in reported incidence of primary

maligant brain tumors in children in the

United States. J Natl Cancer Inst

1998;90:1269-1277.

6. Gurney JG, Ross JA, Wall DA, Bleyer WA,

Severson RK, Robison LL. Infant cancer in the

U.S.: histology-specific incidence and trends,

1973 to 1992. J Pediat Hematol Oncol

1997;19:428-432.

7. Gurney JG, Davis S, Severson RK, Fang J-Y,

Ross JA, Robison LL. Trends in cancer inci-

dence among children in the U.S. Cancer

1996;78:532-541.

8. Bunin GR, Feuer EJ, Witman PA, Meadows AT.

Increasing incidence of childhood cancer:

report of 20 years experience from the greater

Delaware Valley Pediatric Tumor Registry.

Paediatr Perinat Epidemiol 1996;10:319-338.

9. EPA Conference on Preventable Causes of

Cancer in Children (Minutes). Office of

Children’s Health Protection, U.S. Environ-

mental Protection Agency. Washington, DC

20460, 1997.

10. Preston-Martin S and Mack WJ. Neoplasms of

the nervous system. In: Cancer Epidemiology

and Prevention. D Schottenfeld and JF

Fraumeni eds. Oxford University Press, New

York, 1996;1231-1281.

11. Ron E, Modan B, Boice JD, Alfandary E,

Stovall M, Chetrit A, and Katz L. Tumors of the

brain and nervous system after radiotherapy in

childhood. N Engl J Med 1988;319:1033-1039.

12. Shore RE, Albert RE, and Pasternack BR.

Follow up study of patients treated by X-ray

epilation for tinea capitis: Resurvey of

post-treatment illness and mortality experi-

ence. Arch Environ Health 1976; 31:21-28.

13. Kuijten RR and Bunin GR. Risk factors for

childhood brain tumor: A review. Cancer

Epidemiol Biomarkers Prev 1993;2:277-288.

14. Narod SA, Stiller C, Lenoir GM. An estimate

of the heritable fraction of childhood cancer. Br

J Cancer 1991; 63:993-9.

15. Bunin GR, Kuijten RR, Buckley JD, Rorke LB,

and Meadows AT. Relation between maternal

diet and subsequent primitive neuroectodermal

brain tumors in young children. N Engl J Med

1993; 329:536-541.

16. Bunin GR, Kuijten RR, Boesel CP, Buckley JD,

and Meadows AT. Maternal diet and risk of

astrocytic glioma in children: a report from the

Children’s Cancer Group. Cancer Causes

Control 1994;5:177-187.

17. Preston-Martin S, Pogoda JM, Mueller BA,

Holly EA, Lijinsky W, and Davis RL. Maternal

consumption of cured meats and vitamins in

relation to pediatric brain tumors. Cancer

Epidemiol Biomarkers Prev 1996;5:599-605.

18. Kuijten RR, Strom SS, Rorke LB, Boesel CP,

Buckley JD, Meadows AT, and Bunin GR.

Family history of cancer and seizures in young

children with brain tumors: A report from the

Children’s Cancer Group (United States and

Canada) Cancer Causes and Control

1993;4:455-464.

19. Gurney JG, Mueller BA, Davis S, and Schwartz

SM. Childhood brain tumor occurrence in

relation to residential power line configura-

tions, electric heating sources, and electric

appliance use. Am J Epidemiol 1996;143:120-

128.

20. Davis JR, Brownson RC, Garcia R, Bentz BJ,

and Turner A. Family pesticide use and

childhood brain cancer. Arch Environ Contam

Toxicol 1993;24:87-92.

21. Bunin GR, Buckley JD, Boesel CP, Rorke LB,

and Meadows AT. Risk factors for astrocytic

glioma and primitive neuroectodermal tumor

of brain in young children: A report from the

Children’s Cancer Group. Cancer Epidemiol

Biomarkers Prev 1994;3:197-204.

22. Draper GJ, Heaf MM, and Kinnier Wilson LM.

Occurrence of childhood cancers among sibs

and estimation of familial risks. J Med Genet

1977;14:81-90.

23. Miller RW. Deaths from childhood cancer in

sibs. N Engl J Med 1968;279:122-126.

24. Farwell J and Flannery JT. Cancer in relatives

of children with central-nervous-system

neoplasms. N Engl J Med 1984;311:749-753.

25. McKean Cowdin R, Preston-Martin S, Pogoda

JM, Holly EA, Mueller BA, Davis RL. Parental

occupation and childhood brain tumors –

astroglial and primitive neuroectodermal

tumors. J Occup Environ Med 1998;40:332-

340.

26. Gurney JG, Preston-Martin S, McDaniel AM,

Mueller BA, Holly EA. Head injury as a risk

factor for brain tumors in children: results

from a multicenter case-control study. Epide-

miology 1996;7:485-9.

27. Gurney JG, Mueller BA, Preston-Martin S,

McDaniel AM, Holly EA, Pogoda JM, Davis RL.

A study of pediatric brain tumors and their

association with epilepsy and anticonvulsant

use. Neuroepidemiology 1997;16:248-55.

28. McLendon RE, Tien RD. Genetic syndromes

associated with tumors and/or hamartomas. In

Bigner DD, McLendon RE, Bruner JM: Russell

ICCC IIICNS

National Cancer Institute SEER Pediatric Monograph

and Rubinstein’s Pathology of Tumors of the

Nervous System, 6

Edition. Arnold; London

GB, 1998: 371-417.

29. Preston-Martin S, Gurney JG, Pogoda JM,

Holly EA, Mueller BA. Brain tumor risk in

children in relation to use of electric blankets

and water bed heaters. Am J Epidemiol

1996;143:1116-1122.

30. Preston-Martin S, Navidi W, Thomas D, Lee P,

Bowman J, Pogoda J. Los Angeles study of

residential magnetic fields and childhood brain

tumors. Am J Epidemiol 1996;143:105-119.

31. Pogoda JM, Preston-Martin S. Household

pesticides and risk of pediatric brain tumors.

Environ Health Perspect 1997; 105:1214-1220.

National Cancer Institute

SEER Pediatric Monograph

ICCC IVSYMPATHETIC NERVOUS SYSTEM TUMORS

National Cancer Institute

SEER Pediatric Monograph

Marc T. Goodman, James G. Gurney, Malcolm A. Smith, Andrew F. Olshan

HIGHLIGHTS

Incidence

♦ In the US, approximately 700 children and adolescents younger than 20 years of age

are diagnosed with tumors of the sympathetic nervous system each year, of which

approximately 650 are neuroblastomas.

♦ Sympathetic nervous system tumors accounted for 7.8% of all cancers among

children younger than 15 years of age.

♦ Over 97% of sympathetic nervous system tumors are neuroblastomas, embryonal

malignancies of the sympathetic nervous system that occur almost exclusively in

infants and very young children.

♦ Regardless of age, neuroblastomas most commonly occurred in the adrenal gland.

Mediastinal tumors were more frequent in infants than in older children, while the

opposite age pattern was observed for CNS tumors (Figure IV.1).

♦ The average age-adjusted annual incidence rate for all sympathetic nervous system

cancers was 9.5 per million children.

♦ The occurrence of sympathetic nervous system malignancies was strongly

age-dependent (Figure IV.2). For neuroblastomas alone, the incidence rate for both

sexes combined during the second year of life (29 per million) was less than half that

of infancy (64 per million).

♦ Neuroblastomas were by far the most common cancer of infancy, with an incidence

rate almost double that of leukemia, the next most common malignancy that

occurred during the first year of life.

♦ Sixteen percent of infant neuroblastomas were diagnosed during the first month

following birth and 41% were diagnosed during the first 3 months of life (Figure

IV.3).

♦ Over the 21-year observation period, there was little indication of an increase in the

overall incidence of sympathetic nervous system malignancies (Figure IV.4). The

estimated annual percent change in age-adjusted incidence rates was 0.4%.

Survival

♦ For children aged 1 to 4 years at diagnosis, 5-year survival rate improved from 35%

during 1975-84 to 55% during 1985-94. Survival at 5 years from diagnosis was

essentially unchanged over these time intervals among infants (83%) and children 5

years or older (40%).

Risk factors

♦ Relatively little is known about the etiology of sympathetic nervous system tumors

(Table IV.3). The young age at onset of most cases illustrates the need to investigate

exposure events occurring before conception and during gestation.

ICCC IV SYMPATHETIC NERVOUS SYSTEM TUMORS

National Cancer Institute

SEER Pediatric Monograph

INTRODUCTION

Neuroblastoma is an embryonal malig-

nancy of the sympathetic nervous system

that is derived from primordial neural crest

cells and occurs almost exclusively in

infants and young children [1]. Other

childhood malignancies of the sympathetic

nervous system include

ganglioneuroblastoma, which is a more

differentiated variant of neuroblastoma,

and the histogenetically related pheochro-

mocytoma [2]. Malignant paragangliomas,

medulloepitheliomas, neuroepitheliomas

and olfactory neurogenic tumors are also

cancers of the sympathetic nervous system,

although they are extremely rare in chil-

dren and will not be emphasized. To follow

the convention of the International Classifi-

cation of Childhood Cancer system [3], data

for neuroblastoma and ganglioblastoma are

grouped together as one category (hence-

forth called neuroblastomas), and all other

sympathetic nervous system malignancies

as a second category. Because of important

distinctions in biological characteristics and

prognosis of neuroblastomas in infants (less

than 1 year at diagnosis) compared with

older children (older than 1 year of age at

diagnosis) [1], data are provided to high-

light the epidemiology of both age groups

individually. Additionally, because the

occurrence of neuroblastomas and other

sympathetic nervous system malignancies

are so rare in adolescents, the rate calcula-

tions and discussion are limited to children

younger than 15 years of age. In the US,

approximately 700 children and adoles-

cents younger than 20 years of age are

diagnosed with tumors of the sympathetic

nervous system each year, of which approxi-

mately 650 are neuroblastomas.

INCIDENCE

During the 21-year period from 1975

through 1995, 1,542 children were diag-

nosed with sympathetic nervous system

malignancies in the SEER areas (Table

IV.1). This represented 7.8% of all cancer

Figure IV.1 Percent distribution of neuroblastomas

by primary site and age, all races, both sexes

SEER, 1975-95

Adrenal gland

Connective,

Retroperitonium

Mediastinum

Central

Autonomic

Other sites

01020304050

Relative percent

0 1020304050

subcutaneous,

soft tissue

nervous

system

nervous

system

Age <1 year

at diagnosis

Age 1+ year

at diagnosis

Table IV.1: Number of cases and age-adjusted* incidence rates per million

by ICCC categories of sympathetic nervous system malignancies

and sex, age <15, all races, SEER, 1975-95

Males Females Total

Tumor Type No. Rate No. Rate No. Rate

Neuroblastomas 787 9.4 705 8.9 1492 9.1

Other sympathetic nervous system 28 0.4 22 0.3 50 0.3

Total 815 9.8 727 9.2 1542 9.5

*Adjusted to the 1970 US standard population

ICCC IVSYMPATHETIC NERVOUS SYSTEM TUMORS

National Cancer Institute SEER Pediatric Monograph

in this age group. The majority (97%) of

these malignancies were neuroblastomas;

only 50 children were diagnosed with any

other histological type. Within the neuro-

blastoma category, ganglioneuroblastomas

comprised 15% of tumors (8% among in-

fants and 20% among those 1-14 years of

age).

The distribution of neuroblastomas by

primary site is shown in Figure IV.1. Re-

gardless of age, neuroblastomas most

commonly occurred in the adrenal gland.

Mediastinal tumors were more frequent in

infants than in older children, while the

opposite age pattern was observed for CNS

tumors.

Age-specific incidence

The incidence rate for all sympathetic

nervous system cancers was 9.5 per million

children. The occurrence of sympathetic

nervous system malignancies, however, was

strongly age-dependent. Figure IV.2 illus-

trates the incidence rates by single year of

age

and sex, and shows the predominance

of neuroblastomas during infancy. For

neuroblastomas alone, the incidence rate

for both sexes combined during the second

year of life (29 per million) was less than

half that of infancy (64 per million). The

rates for sympathetic nervous system

tumors other than neuroblastomas were

1.2 per million for infants, and less than 1

per million for all other single years of age.

Neuroblastomas were by far the most

common cancer of infancy with an incidence

rate almost double that of leukemia, the

next most common malignancy that occurs

during the first year of life [4]. As shown in

Figure IV.3, 16% of infant neuroblastomas

were diagnosed during the first month

following birth and 41% were diagnosed

during the first 3 months of life.

Figure IV.2: Sympathetic nervous system age-specific

incidence rates by sex, all races

SEER, 1976-84 and 1986-94

)

0123456789101112131415

Age (in years) at diagnosis

Average annual rate per million

Males

Females

)

Figure IV.3: Percent distribution of infant neuroblastomas

by month of age, all races, both sexes, SEER, 1975-95

0123456789101112

Age (in months) at diagnosis

Percentage of cases

Enumeration of the population at risk by single years of age was

available only for the census years 1980 and 1990. The US Bureau

of the Census provides intercensal population estimates by 5-year

age groups, but not by single years of age. Therefore, the

population estimates for 1980 were used in rate calculations for

cases diagnosed from 1976-84 and the 1990 estimates were used for

cases diagnosed from 1986-94.

ICCC IV SYMPATHETIC NERVOUS SYSTEM TUMORS

National Cancer Institute

SEER Pediatric Monograph

Sex and race-specific incidence

Figure IV.2 also demonstrates that the

incidence of sympathetic nervous system

cancer was slightly higher among males

than females. For neuroblastomas, overall

male rates (9.8 per million) were 6.5%

higher than female rates (9.2 per million)

with the greatest difference occurring

during infancy (69.3 per million versus 59.6

per million for males and females, respec-

tively). There was no discernable sex

difference for sympathetic nervous system

malignancies other than neuroblastomas.

White infants of both sexes had a

higher incidence of sympathetic nervous

system tumors than did black infants, but

little difference by race was observed

among older children (Table IV.2). The

ratio of white to black incidence rates

among infants was 1.7:1 for males and

1:9:1

for females. In Table IV.2, “all races”

includes whites, blacks, and children of

other identified racial or ethnic back-

grounds. There were too few cases of

sympathetic nervous system among any

other races to calculate reliable incidence

rates.

TRENDS

Over the 21-year observation period,

there was little indication of a linear trend

in the overall incidence of sympathetic

nervous system malignancies (Figure IV.4).

The estimated annual percent change in

age-adjusted incidence rates was 0.37%

(p > 0.05). Rates, however, have increased

somewhat among infants during recent

years. Figure IV.5 shows incidence rates of

neuroblastomas by year of age at diagnosis

for the periods 1976-84 versus 1986-94.

Among infants, the rate in the earlier time

period was 53 per million compared to 74

per million in the later time period. No

differences in rates between the time

Figure IV.4: Trends in sympathetic nervous system

age-adjusted* incidence rates by year of diagnosis

age <15, all races, both sexes, SEER, 1975-95

(

((

(

((

(

75 77 79 81 83 85 87 89 91 93 95

Year of diagnosis

Average annual rate per million

*Adjusted to the 1970 US standard population

Table IV.2: Average annual age-specific incidence rates per million for all

sympathetic nervous system tumors by age, sex, and race

SEER 1975-95

Males Females

Age (in years) at diagnosis White Black All White Black All

<1 83.6 50.5 69.3 74.1 38.4 59.6

1-14 8.2 7.4 7.3 7.6 6.5 6.5

<15* 10.1 8.8 9.8 9.6 8.6 9.2

* Adjusted to the 1970 US standard population

ICCC IVSYMPATHETIC NERVOUS SYSTEM TUMORS

National Cancer Institute SEER Pediatric Monograph

periods occurred for children either 1 or 2

years of age at diagnosis. Thus, it does not

appear that the increase among infants can

be explained by a shift towards earlier age

at diagnosis. The increase among infants,

however may be a result of de facto fetal

and neonatal screening. Mass screening of

infants for neuroblastoma has been evalu-

ated in recent years in Japan, Canada, and

some countries in Europe [5,6]. Although

systematic screening for neuroblastoma is

not conducted in the United States, the

awareness of screening in other countries

and the recent widespread availability of

non-invasive diagnostic tests for neuroblas-

toma may have resulted in US physicians

diagnosing cases of neuroblastoma with

minimal clinical symptomatology that

previously were undetected. The docu-

mented ability of some fetal and infant

neuroblastomas to spontaneously regress is

consistent with the hypothesis that the

increased incidence among infants is the

result of detection of cases that were previ-

ously not diagnosed [1,9,10]. Also consis-

tent with this hypothesis is the recent

widespread use of prenatal ultrasound

testing with coincidental detection of adre-

nal neuroblastomas [7,8].

SURVIVAL

Prognosis for neuroblastomas is depen-

dent on age, stage of disease, and the

molecular biologic and cytogenetic charac-

teristics of the tumor [1]. Figure IV.6

illustrates the more favorable prognosis for

infants with neuroblastoma (5-year relative

survival rate, 83%) compared to children

older than 1 year of age. The favorable

outcome for infants with neuroblastoma no

doubt reflects the favorable biological

Figure IV.5: Neuroblastoma age-specifice incidence

rates by age, all races, both sexes

SEER, 1976-84 and 1986-94

53.7

29.3

20.9

9.9 9.9

73.4

28.5

20.7

15.6

6.8

<11234

Age (in years) at diagnosis

Average annual rate per million

1976-84

1986-94

Figure IV.6: Neuroblastoma 5-year relative survival

rates by sex, race, age, and time period

SEER (9 areas), 1975-84 and 1985-94

Total Male Female White Black <1 1-4 5-9

100

Percent surviving 5 years

1975-84

1985-94

Sex

Race

Age

ICCC IV SYMPATHETIC NERVOUS SYSTEM TUMORS

National Cancer Institute

SEER Pediatric Monograph

characteristics of neuroblastomas arising in

this age group [1]. For children aged 1 to 4

years at diagnosis, the 5-year survival rate

improved from 35% during 1975-84 to 55%

during 1985-94. Survival was essentially

unchanged during these time intervals for

children older than 4 years of age (40%).

There were no substantive differences in

survival by sex or race (Figure IV.6).

RISK FACTORS

Relatively little is known about the

etiology of sympathetic nervous system

tumors (Table IV.3). The young age at

onset of most cases illustrates the need to

investigate exposure events occurring

before conception and during gestation.

The few epidemiological investigations of

Table IV.3: Current knowledge on causes of neuroblastoma (NB)

Exposure or Characteristic Comments References

Medications Two studies have reported increased risk when mothers

took medications during pregnancy such as

amphetamines, diuretics, tranquilizers, or muscle

relaxers or for vaginal infection. Other studies have

reported an association with maternal phenytoin

treatment.

11,12,13

Hormones Two studies reported that sex hormones were associated

with an increase in risk. One of the studies reported a 10-

fold increased risk for fertility drug use prior to

pregnancy.

12,13,14

Birth characteristics One study reported increased risk associated with low

birth weight and protective effect for preterm delivery.

This was not confirmed in two other studies.

13,15,16

Congenital anomalies A variety of congenital anomalies has been reported to

occur with NB in a small number of cases, but no

consistent pattern of association has been shown.

Previous spontaneous

abortion/fetal death

Previous spontaneous abortion was associated with

increased risk in one study and decreased risk in another.

13,16

Alcohol One study reported a dose-response relationship between

frequency of alcohol use during pregnancy and NB, but

another reported no effect. An association with fetal

alcohol syndrome has also been reported.

12,13,17

Tobacco An early study reported no effect of maternal smoking on

risk. However, a later study suggested a weak dose-

response relationship between level of maternal smoking

during pregnancy and NB risk.

12,13

Paternal occupational exposures Three studies have reported conflicting results on the risk

associated with paternal employment in electronics,

agriculture, and packaging and materials handling.

Specific associated occupational exposures include

electromagnetic fields, pesticides, hydrocarbons, dusts,

rubber, paint, and radiation.

18-20

Factors for which evidence is

inconsistent or limited

ICCC IVSYMPATHETIC NERVOUS SYSTEM TUMORS

National Cancer Institute SEER Pediatric Monograph

neuroblastoma have not had sufficient

statistical power or detailed data collection

to provide convincing evidence of etiologic

risk factors. Medications [11,12,13] and

hormones used during pregnancy [12,13,14]

are among the most suggestive factors

suspected to increase the risk of neuroblas-

toma. Certain birth characteristics, pesti-

cide exposure, and parental occupational

exposure to electromagnetic fields

[13,15,16,18-20] have been evaluated, but

with conflicting results. In addition, clinical

and molecular characteristics, such as

amplification of the nmyc oncogene, loss of

heterozygosity of the short arm of chromo-

some 1, and hyperdiploidy, may be useful in

establishing homogenous disease sub-

groups for future epidemiological investiga-

tions of neuroblastoma [1].

SUMMARY

Sympathetic nervous system malignancies,

of which neuroblastomas comprised 97% of

the total, represented 7.8% of cancer in

children younger than 15 years of age. The

incidence rate was 9.5 per million children,

however rates were strongly age-depen-

dent. The incidence rate of sympathetic

nervous system malignancies among in-

fants was 65 per million, and the rate

dropped by half in the second year of life.

Overall, incidence rates did not change

substantially during the study period.

Among infants, however, there was an

increase in incidence rates from 1986-94

compared with the period 1976-84. This

increase was not noted in older children,

thus excluding earlier age at diagnosis as a

likely explanation for the trend. Rather,

the increase likely arose from identification

of previously undetected cases with mini-

mal clinical symptomatology through

widespread application of fetal ultrasound

testing and noninvasive diagnostic tests for

neuroblastoma. The known propensity of

the neuroblastomas of infancy to undergo

spontaneous regression supports this

explanation. Five-year relative survival of

neuroblastomas was 83% for infants, 55%

for children 1-4 years of age, and 40% for

older children. Unfortunately, there is very

little known about why neuroblastoma

occurs, or what factors increase risk for

occurrence.

Reference List

1. Brodeur GM, Castleberry RP. Neuroblastoma.

In: Pizzo PA, Poplack DG, Editors. Principles

and Practices of Pediatric Oncology. 3rd ed.

Philadelphia, PA: Lippencott-Raven; 1997:761-

797.

2. Russell DS, Rubinstein LJ. Pathology of

Tumours of the Nervous System, 5

edition.

Williams & Wilkins, Baltimore, MD, 1989; 898-

943.

3. Kramarova E, Stiller CA. The international

classification of childhood cancer. Int J Cancer.

1996;68:759-765.

4. Gurney JG, Ross JA Wall DA Bleyer WA,

Severson RK, Robison LL. Infant cancer in the

U.S.: histology-specific incidence and trends,

1973-1992. J Pediatr Hematol Oncol 1997;

19:428-432.

5. Woods WG, Tuchman M, Robson LL, Bernstein

M, Leclerc JM, Brisson LC, Brossard J, Hill G,

Shuster J, Luepker R, Byrne T, Weitzman S,

Bunin G, Lemieux. A population-based study of

the usefulness of screening for neuroblastoma.

Lancet 1996; 348:1682-1687.

6. Parker L, Powell J. Screening for neuroblas-

toma in infants younger than 1 year of age:

review of the first 30 years. Med Pediatr Onc

1998; 31:455-469.

7. Moore RM, Jeng LL, Kaczmarek RG, Placek

PJ. Use of diagnostic imaging procedures and

fetal and monitoring devices in the care of

pregnant women. Public Health Rep 1990;105:

471-475.

8. Archarya S, Jayabose S, Kogan SJ, Tugal O,

Beneck D, Leslie D, Slim M. Prenatally

diagnosed neuroblastoma. Cancer 1997;

80:304-310.

9. Carlsen NLT. Neuroblastoma: epidemiology

and pattern of regression. Problems in inter-

preting results of mass screening. J Pediatr

Hemat Oncol 1992;14:103-110.

10. Evans AE, Chatten J, D’Angio GJ, Gerson JM,

Robinson J, Schnaufer L. A review of 17 IV-S

neuroblastoma patients at the Children’s

Hospital of Philadelphia. Cancer 1980;45:833-9.

11. Bodeur GM. Neuroblastoma and other periph-

eral neuroectodermal tumors. Chapter 24. In:

Clinical Pediatric Oncology, 4

Ed. Fernbach

DJ, Vietti TJ (Eds.). Mosby Yearbook, St. Louis,

MO., 1991.

ICCC IV SYMPATHETIC NERVOUS SYSTEM TUMORS

National Cancer Institute

SEER Pediatric Monograph

12. Kramer S, Ward E, Meadows AT, and Malone

KE. Medical and drug risk factors associated

with neuroblastoma: A case control study. J

Natl Cancer Inst 1987;78:797-804.

13. Schwartzbaum JA. Influence of mother’s

prenatal drug consumption on risk of neuro-

blastoma in the child. Am J Epidemiol

1992;135:1358-1367.

14. Michalek AM, Buck GM, Nasca PC, Freedman

AN, Baptiste MS, and Mahoney MC. Gravid

health status, medication use, and risk of

neuroblastoma. Am J Epidemiol 1996;143:996-

1001.

15. Johnson CC and Spitz MR. Neuroblastoma: A

case control analysis of birth characteristics. J

Natl Cancer Inst 1985;74:789-192.

16. Neglia JP, Smithson WA, Gunderson P, King

FL, Singher LJ, and Robison LL. Prenatal and

perinatal risk factors for neuroblastoma. A

case-control study. Cancer 1988;61:2202-2206.

17. Kinney H, Faix R, and Brazy J. The fetal

alcohol syndrome and neuroblastoma. Pediat-

rics 1980;66:130-132.

18. Bunin GR, Ward E, Kramer S, Rhee CA, and

Meadows AT. Neuroblastoma and parental

occupation. Am J Epidemiol 1990;131:776-780.

19. Spitz MR and Johnson CC. Neuroblastoma and

paternal occupation. Am J Epidemiol

1985;121:924-929.

20. Wilkins JR and Hundley VD. Paternal occupa-

tional exposure to electromagnetic fields and

neuroblastoma in offspring. Am J Epidemiol

1990;131:995-1008.

ICCC V

National Cancer Institute

SEER Pediatric Monograph

RETINOBLASTOMA

HIGHLIGHTS

Incidence

♦ Retinoblastoma accounted for approximately 11% of cancers developing in the first

year of life, but for only 3% of the cancers developing among children younger than

15 years of age.

♦ In the US, approximately 300 children and adolescents younger than 20 years of age

are diagnosed with retionblastomas each year.

♦ The vast majority of cases of retinoblastoma occur among young children, with

almost two-thirds (63%) of all retinoblastomas occurring before the age of two years

and 95% occurring before the age of five years.

♦ The incidence of bilateral tumors was strongly age dependent with 42% of the retino-

blastomas occurring in children less than one year of age being bilateral compared to

21% of those among children aged one year, and only 9% among older children.

♦ Rates of retinoblastoma were essentially equal among males (3.7 per million) and

females (3.8 per million) and among whites (3.7 per million) and blacks (4.0 per

million) (Table V.2)

♦ There was no substantial sustained change in retinoblastoma incidence during the

21-year period, 1975-95 (Figure V.3 and Table V.2).

Survival

♦ Survival for children with retinoblastoma was quite favorable, with more than 93%

alive at five years after diagnosis. Males and females had similar 5-year survival

rates for the period 1976-94 (93-94%). Black children had slightly lower 5-year

survival rates than white children (89% versus 94%) (Figure V.5).

Risk factors

♦ A retinoblastoma gene has been identified. Each child of a parent with familial

bilateral retinoblastoma has a 50% risk of inheriting the retinoblastoma gene. Some

patients develop the gene as the result of a new mutation (sporadic heritable retino-

blastoma) and can pass the gene on to their children even though they did not

inherit the gene from their parents. Children who inherit the retinoblastoma gene

have a 90% risk of developing retinoblastoma. Genetic retinoblastomas are more

likely to be bilateral and to occur during the first year of life. Little is know about

non-genetic (sporadic) retinoblastomas (Table V.3).

INTRODUCTION

Retinoblastoma is a tumor of childhood

which arises in the retina of the eye and

extremely rarely in the pineal gland [1].

Two types of retinoblastomas have been

described: those linked to genetic muta-

tions and the so-called sporadic retinoblas-

tomas. The genetic-linked retinoblastomas

are divided into two groups, those which

arise in children who carry the retinoblas-

toma gene inherited from one or both

parents (familial retinoblastoma) and those

in which the disease occurs as the result of

a new mutation, usually in their father’s

sperm but sometimes in their mother’s egg

(sporadic heritable retinoblastoma) [2,3].

Both familial retinoblastomas and sporadic

John L. Young, Jr., Malcolm A. Smith, Steven D. Roffers, Jonathan M. Liff, Greta R. Bunin

ICCC V RETINOBLASTOMA

National Cancer Institute

SEER Pediatric Monograph

heritable retinoblastomas are more likely

to be bilateral and to occur during the first

year of life, while the sporadic retinoblasto-

mas are more likely to be unilateral and

occur after the first year of life.

The importance of retinoblastoma to

cancer research far exceeds the low inci-

dence of this uncommon tumor, since it was

through careful analysis and insightful

mathematical modeling of the age distribu-

tion of unilateral and bilateral cases that

the “tumor suppressor gene” concept was

initially developed [4,5]. Subsequent work

led to the localization of the gene respon-

sible for retinoblastoma to a small region

on the long arm of chromosome 13 [6], and

eventually to isolation of the gene itself [7].

The retinoblastoma gene product is now

recognized as a critical element in control-

ling progression through the cell cycle, and

abnormalities of the retinoblastoma gene

are among the most common occurring in

all types of cancer cells [5,8]. In the US,

approximately 300 children and adoles-

cents younger than 20 years of age are

diagnosed with retionblastomas each year.

INCIDENCE

Table V.1 shows the distribution of

retinoblastomas diagnosed among residents

of the SEER areas during 1975-95 by

laterality, race, and sex. Approximately

one-fourth of all retinoblastomas were

bilateral. All bilateral disease is hereditary

whereas unilateral disease may or may not

be hereditary. The percentage of unilateral

and bilateral tumors were similar for black

children and white children and for males

and females.

The vast majority of cases of retino-

blastoma occur among young children

(Figure V.1), with almost two-thirds (63%)

Table V.1: Number of retinoblastomas by laterality, sex, and race, age <15

SEER, 1975-95

Total Unilateral Bilateral Unknown

No. No. % No. % No. %

Total 625 453 72.5 154 24.6 18 2.9

Males 314 226 72.0 76 24.2 12 3.8

Females 311 227 73.0 78 25.1 6 1.9

Whites 474 341 71.9 116 24.5 17 3.6

Blacks 86 61 70.9 25 29.0 - -

Figure V.1: Unilateral and bilateral retinoblastoma

age-specific incidence rates, all races

both sexes, SEER, 1976-84 and 1986-94

(

((

012345678

Age (in years) at diagnosis

Average annual rate per million

Unilateral

Bilateral

Total

(

ICCC VRETINOBLASTOMA

National Cancer Institute SEER Pediatric Monograph

of all retinoblastomas diagnosed between

1975-95 among children residing in the

SEER areas occurring before the age of two

years and with 95% occurring before the

age of five years. Since retinoblastoma was

extremely rare after the age of five years

(only 28 childhood retinoblastomas re-

ported to SEER areas as having been

diagnosed younger than 20 years of age

occurred in children aged 15-19 years in

comparison to 625 cases in children

younger than 15 years of age), information

presented in this chapter will be limited to

children younger than 15 years of age.

The incidence of bilaterality was

strongly age dependent, with 42% (103/248)

of the retinoblastomas occurring in children

less than one year of age being bilateral

compared to 21% (31/147) of those among

children aged one year, and only 9% (20/

230) among older children. The incidence

rates for both unilateral and bilateral

retinoblastoma decrease as age increases.

The incidence rate for bilateral retinoblas-

toma drops to almost zero after age 2, while

the rate for unilateral retinoblastoma

remains higher until after age 7 (Figure

V.1). Figure V.2 provides a more detailed

view of incidence in the first 3 years of life,

with incidence for the first year of life being

estimated by two month age intervals. For

bilateral tumors, the peak incidence is at 4-

5 months of age, with a sharp decline

thereafter and with very low rates by the

third year of life. For unilateral tumors,

peak incidence is also in the first year of life

(at 6-7 months), but the decline in incidence

with increasing age is much more gradual

than for bilateral tumors, with rates re-

maining above 10 per million children for

the first 3 years of life.

The incidence rate of retinoblastoma

for the period 1975-95 was 3.8 per million

(Table V.2). Retinoblastoma accounted for

approximately 11% of the cancers develop-

ing in the first year of life, but for only 3%

of the cancers developing among children

younger than 15 years of age. Rates of

retinoblastoma among males (3.7 per

million) and females (3.8 per million) were

essentially equal. Rates for whites (3.7 per

million) and blacks (4.0 per million) were

also similar.

Figure V.2: Unilateral and bilateral retinoblastoma

age-specific incidence rates, age <3, all races

both sexes, SEER, 1976-84 and 1986-94

Age (in months) at diagnosis

(

0122436

Average annual rate per million

Unilateral

Bilateral

(

< 2

2-3

4-5

6-7

8-9

10-11

Table V.2: Average annual age-adjusted* incidence rates

per million of retinoblastoma, by time period,

race, and sex, age <15 SEER, 1975-95

1975-95 1975-79 1980-84 1985-89 1990-95

All races, Both sexes

3.8 3.6 3.6 3.7 4.0

Male

3.7 3.6 3.3 3.4 4.2

Female

3.8 3.6 4.0 4.0 3.8

White

3.7 3.3 3.5 3.7 4.0

Black

4.0 4.6 4.0 3.7 3.8

* Adjusted to the 1970 US standard population

ICCC V RETINOBLASTOMA

National Cancer Institute

SEER Pediatric Monograph

TRENDS

Figure V.3 shows the incidence for

retinoblastoma among children younger

than 15 years of age for 1975-95. Because

of the relatively small numbers of children

with retinoblastoma diagnosed in SEER

areas annually (approximately 20 per

year), there was considerable variability in

the year-to-year rates. There was no sub-

stantial sustained change in retinoblas-

toma incidence during this 21 year period.

Table V.2 shows the incidence rate for

retinoblastoma among children younger

than 15 years of age for four specific time

periods between 1975-95 by sex and race.

Rates were slightly higher in the last time

period for males and for whites. The esti-

mated overall annual percent change was

about one-half of one percent per year

(0.6%). The change was greater for males

(1.2%) than for females (0.2%), but neither

change was significant. The rates for

blacks were higher than whites for the

earlier time period but by the late 1980s

were similar. Figure V.4 shows the inci-

dence of retinoblastoma by year of age for

an early time period (1976-84) and for a

recent time period (1986-94) and illustrates

that the small increase that did occur

between these two time periods was prima-

rily the result of increased diagnosis of

retinoblastoma in the first year of life.

SURVIVAL

Figure V.5 shows that survival for

children diagnosed with retinoblastoma in

the period 1975-94 was quite favorable,

with more than 93% alive at five years

after diagnosis. Males and females had

similar 5-year survival rates for the period

1975-94 at 93-94%, while black children

had slightly lower 5-year survival rates

than white children (89% versus 94%).

Figure V.3: Trends in retinoblastoma age-adjusted*

incidence rates, age <15, all races

both sexes, SEER 1975-1995

1975 1978 1981 1984 1987 1990 1993 1996

Year of diagnosis

Average annual rate per million

*Adjusted to the 1970 US standard population

Figure V.4: Retinoblastoma age-specific incidence

rates by time period, all races, both sexes

SEER, 1976-84 and 1986-94

(

012345678

Age (in years) at diagnosis

Average annual rate per million

1976-84

1986-94

(

ICCC VRETINOBLASTOMA

National Cancer Institute SEER Pediatric Monograph

RISK FACTORS

While the genetics of retinoblastoma

are well understood, there is much less

known about the role of non-genetic factors

in retinoblastoma (Table V.3). Each child of

a parent with familial bilateral retinoblas-

toma has a 50% risk of inheriting the

retinoblastoma gene. Patients with spo-

radic heritable retinoblastoma carry the

gene for retinoblastoma and can also pass

the gene on to their children even though

they did not inherit the gene from their

parents. Children who inherit the retino-

blastoma gene have a 90% risk of develop-

ing retinoblastoma. Sporadic

(nonheritable) retinoblastoma results from

post-conception events and has been associ-

ated in a single study with parental occupa-

tion (Table V.3) [9].

SUMMARY

Retinoblastomas occur among very

young children, usually before the age of

five years. The incidence is about equal

among males and females and among black

children and white children. Rates have

changed little over the 21-year period,

Figure V.5: Retinoblastoma 5-year relative survival

rates, by sex, race, age and time period

SEER (9 areas), 1975-84 and 1985-94

# - <25 cases - rate not shown

93 93

94 94

Total Male Female White Black <5 5-9 10-1415-19

100

Percent surviving 5 years

1975-84

1985-94

Sex Race

Age

###

Known risk factors

Factors for which evidence is

inconsistent or limited

Table V.3: Current knowledge of causes of retinoblastoma

Exposure or Characteristic Comments References

Parent with history of bilateral

retinoblastoma

Each child has a 50% risk of inheriting the retinoblastoma

gene. If the gene is inherited, the risk of retinoblastoma is

over 90%. A small proportion of unilateral patients also

carry the gene and can pass it on to their children.

4,10

13q deletion syndrome Recognition of this syndrome led to the identification of the

retinoblastoma gene.

Paternal occupation There is a single report of association with employment in

the military, metal manufacturing, and as welder,

machinist, or related occupation.

ICCC V RETINOBLASTOMA

National Cancer Institute

SEER Pediatric Monograph

1975-95. Familial and sporadic heritable

retinoblastomas are caused by genetic

mutations and both usually result in the

bilateral form of the disease. Little is

known about the causes of sporadic non-

heritable retinoblastomas. Survival rates

are greater than 90% for children with

retinoblastoma.

Reference List

1. Donaldson S, Egbert P, Newsham I, et al:

Retinoblastoma. In Principles and Practice of

Pediatric Oncology (Pizzo P, Poplack D, eds).

Philadelphia: Lippincoltt-Raven, 1997, pp 699-

715.

2. Dryja TP, Morrow JF, Rapaport JM: Quantifi-

cation of the paternal allele bias for new

germline mutations in the retionoblastoma

gene. Hum genet 100:446-9, 1997.

3. Dryja TP, Mukai S, Pertersen R, et al: Parental

origin of mutations of the retinoblastoma gene.

Nature 339:556-8, 1989.

4. Knudson Ag, Jr.: Mutation and cancer: statisti-

cal study of retinoblastoma. Proc Natl Acad Sci

U S A 68:820-3, 1971.

5. Knudson AG: Antioncogenes and human

cancer. Proc Natl Acad Sci U S A 90:10914-21,

1993.

6. Dryja TP, Rapaport JM, Joyce JM, et al:

Molecular detection of deletions involving band

q14 of chromosome 13 in retinoblastomas. Proc

Natl Acad Sci U S A 83:7391-4, 1986.

7. Friend, SH, Bernards R, Rogelj S, et al: A

human DNA segment with properties of the

gene that predisposes to retinoblastoma and

osteosarcoma. Nature 323:643-6, 1986.

8. Weinberg RA: The retinoblastoma protein and

cell cycle control. Cell 81:323-30, 1995.

9. Bunin GR, Petrakova A, Meadows AT, et al:

Occupations of parents of children with

retinoblastoma: a report from the Children’s

Cancer Study Group. Cancer Res 50:7129-33,

1990.

10. Li F: Familial aggregation. In Cancer Epidemi-

ology and Prevention (Schottenfeld D,

Fraumeni J, eds). New York: Oxford University

Press, 1996, pp 546-558.

ICCC VI

RENAL TUMORS

National Cancer Institute

SEER Pediatric Monograph

National Cancer Institute

SEER Pediatric Monograph

Leslie Bernstein, Martha Linet, Malcolm A. Smith, Andrew F. Olshan

HIGHLIGHTS

Incidence

♦ Malignancies of the kidney (renal cancers) represented 6.3% of cancer diagnoses

among children younger than 15 years of age (incidence 7.9 per million) (Table

VI.2) and 4.4% of cancer diagnoses for children and adolescents younger than 20

years of age (incidence of 6.2 per million).

♦ In the US approximately 550 children and adolescents younger than 20 years of

age are diagnosed with renal tumors each year, of which approximately 500 are

Wilms’ tumor.

♦ Wilms’ tumor was by far the most common form of renal cancer in children

younger than 15 years of age, representing approximately 95% of diagnoses

(Tables VI.1 and VI.2). Much less common were rhabdoid tumors of the kidney

(1% of renal cancers) and clear cell sarcoma of the kidney (1.6% of renal cancers).

Renal carcinomas, the most common form of renal cancer in adults, represented

only 2.6% of renal cancers in children younger than 15 years of age.

♦ Wilms’ tumor occurred most commonly among children younger than 5 years of

age (Figure VI.1), with very low incidence for 10-14 and 15-19 year olds. The

highest incidence for Wilms’ tumor was in the first 2 years of life, followed by

steadily decreasing rates with increasing age (Figure VI.2).

♦ Rhabdoid tumor of the kidney was diagnosed primarily in infants, while clear cell

sarcoma of the kidney was diagnosed primarily during the first 4 years of life.

Renal carcinomas, by contrast, occurred with highest incidence among 15-19 year

olds (Figure VI.1).

♦ Females had slightly higher incidence than males for Wilms’ tumor during the

period 1975-95 (Table VI.3). For the recent period of 1990-95, however, incidence

rates were similar by sex (Figure VI.3).

♦ Black children had somewhat higher incidence for Wilms’ tumor than white chil-

dren for the period 1975-95. For the time periods 1986-89 and 1990-95, however,

incidence rates by race were similar (Figure VI.4).

♦ Incidence of Wilms’ tumor showed neither substantial increases nor decreases

during the 21-year period from 1975 to 1995 (FigureVI.5).

Survival

♦ The overall relative 5-year survival rate for children with Wilms’ tumor was ap-

proximately 92% for cases diagnosed from 1985-94 (Figure VI.6), an improvement

from the 81% survival rate for cases diagnosed from 1975-84. Relative survival

rates were slightly higher for females than males and slightly higher for black

children than for white children (Figure VI.6).

Risk factors

♦ Certain congenital anomalies and genetic conditions increase susceptibility for

Wilms’ tumor (Table VI.4). Suggestive, although not conclusive, data indicate that

certain paternal occupations may be associated with increased Wilms’ tumor risk.

ICCC VI RENAL TUMORS

National Cancer Institute SEER Pediatric Monograph

INTRODUCTION

Renal tumors occurring in children

comprise a spectrum of morphologic sub-

types, including some with benign histopa-

thology. Wilms’ tumor (also called

nephroblastoma or renal embryoma) is by

far the most common form of renal cancer

in children. Other rarer forms of childhood

renal cancers are: clear cell sarcoma of the

kidney, rhabdoid tumor of the kidney,

congenital mesoblastic nephroma, mul-

tilocular cystic renal tumor, renal cell

carcinoma, and angiomyolipoma [1,2].

During 1975-95 in regions covered by

SEER cancer registries, malignant forms of

renal tumors represented 6.3% of total

cancer diagnoses among children younger

than 15 years of age and 4.4% for the

younger than 20 years old population. The

contribution of renal cancers to the overall

childhood cancer burden was notably age-

dependent, with renal cancers representing

9.7% of total incident malignancies diag-

nosed among children younger than 5 years

of age, 5.4% in children 5-9 years of age,

1.1% in children 10-14 years of age, and

only 0.6% in adolescents 15-19 years of age.

Wilms’ tumor is believed to arise from

primitive metanephric blastema (i.e., the

tissue from which the normal kidney

arises), though this tumor type often con-

tains tissues not occurring in the develop-

ing kidney, including skeletal muscle,

cartilage, and squamous epithelium [3].

Wilms’ tumor usually arises in only one of

the affected child’s kidneys, although

approximately 12% of affected children may

be diagnosed with Wilms’ tumor that is

multicentric in origin [4]. Approximately

7% of children with Wilms’ tumor have

involvement of both kidneys. Patients with

bilateral forms are generally diagnosed at

younger ages and are more likely to have

associated developmental abnormalities

than patients with unilateral forms [4]. In

the US approximately 550 children and

adolescents younger than 20 years of age

are diagnosed with renal tumors each year,

of which approximately 500 are Wilm’s

tumor.

Classification System

The International Classification for

Childhood Cancers (ICCC) Group VI of

Renal Cancers divides malignant neo-

plasms into three subgroups [5]:

a. Wilms’ tumor, rhabdoid tumor of

the kidney, and clear-cell sar-

coma of the kidney

b. Renal carcinoma

c. Unspecified malignant renal

tumors.

Table VI.1: Number of cases and percent distribution of renal cancers by histologic

subtype and age group, all races, both sexes, SEER, 1975-95

Age (in years) at diagnosis <5 5-9 10-14 15-19 <15 <20

Wilms' tumor 880 260 39 21 1,179 1,200

(96.2%)

(95.9%) (66.1%) (35.0%) (94.7%) (92.0%)

Rhabdoid tumor of the kidney 12 * * * 12 12

(1.3%) (1.0%) (0.9%)

Clear cell sarcoma of the kidney 16 * * * 19 19

(1.8%) (1.6%) (1.6%)

Renal carcinoma

6 7 19 38 32 70

(0.7%) (2.6%) (32.2%) (63.3%) (2.6%) (5.4%)

Unspecified renal cancer

******

Total renal cancers 915 271 59 60 1,245 1,305

Number in parenthesis represents the percentage of all renal cancers for the age group that are

represented by the histologic category.

*Less than 5 cases.

National Cancer Institute SEER Pediatric Monograph

ICCC VIRENAL TUMORS

National Cancer Institute

SEER Pediatric Monograph

The numbers of cases of these histologic

diagnoses among children residing in the

SEER areas for the period 1975-95 are

shown in Table VI.1. Malignant forms of

renal tumors were diagnosed in 1,245

children younger than 15 years of age and

in 1,305 children younger than 20 years of

age. Wilms’ tumor was by far the most

common form of renal cancer, accounting

for 94.7% of the 1,245 renal cancers in

children younger than 15 years of age and

92.0% of the 1,305 renal cancers among the

younger than 20 year olds. Occurring much

less commonly among the total 1,245 cases

of renal cancer in children younger than 15

years of age were rhabdoid tumor of the

kidney (12 cases representing 1.0% of renal

cancers) and clear cell sarcoma of the

kidney (19 cases representing 1.6% of renal

cancers). For renal cell carcinoma, there

were 32 cases among children younger than

15 years of age (2.6% of renal cancers) and

70 cases among children younger than 20

years of age (5.4% of renal cancers).

Wilms’ tumor, rhabdoid tumor of the

kidney, and clear cell sarcoma of the kidney

are classified together in the ICCC category

VIa, while the renal carcinomas are

grouped together in the ICCC category VIb.

In presenting incidence data, the ICCC

category VIa (for which Wilms’ tumor

represents greater than 95% of cases for all

age groups) is designated Wilms’ tumor

(ICCC VIa), since the incidence patterns

and trends for this subcategory are largely

determined by cases of Wilms’ tumor.

When Wilms’ tumor is discussed as a single

diagnosis, the term Wilms’ tumor without

any parenthetic modifier is used. Addition-

ally, since incidence rates for renal cancers

were low and based on very small numbers

in children ages 15-19 (a total of 60 inci-

dent cases diagnosed in the SEER areas

during 1975-95), presentation of incidence

data are generally restricted to renal

cancers diagnosed among children younger

than 15 years of age.

INCIDENCE

The average annual age-adjusted

incidence rates of renal cancer for the years

1975-95 was 7.9 per million for children

younger than 15 years of age (Table VI.2)

and 6.2 per million for children younger

than 20 years of age in the SEER areas. As

discussed in greater detail in subsequent

paragraphs, incidence rates for renal can-

cers in toto and for Wilms’ tumor (ICCC

VIa) for the period 1975-95 were slightly

higher for females than males and for black

children compared to white children. How-

ever, for the most recent time period (1990-

95), rates were similar for both sexes and

for black children and white children.

Age-specific incidence

Age-specific incidence rates of renal

cancers in 5-year age groups were highest

among children younger than 5 years of age

Table VI.2: Age-adjusted* incidence rates for renal cancer

by race and sex, age <15, SEER, 1975-95

Both sexes Males Females

Renal cancers (VI)

All races 7.9 7.4 8.4

Whites 8.0 7.4 8.6

Blacks 9.5 9.5 9.6

Wilms’ tumor (VIa)

All races 7.6 7.1 8.1

Whites 7.9 7.3 8.4

Blacks 8.7 8.4 9.0

*Adjusted to the 1970 US standard population

ICCC VI RENAL TUMORS

National Cancer Institute SEER Pediatric Monograph

(Table VI.3). Incidence declined markedly

with increasing age. The age-incidence

pattern for renal cancer in children was

driven by that for Wilms’ tumor (ICCC

VIa), as illustrated in Figure VI.1. Renal

cell carcinomas (ICCC VIb) occurred very

infrequently among each 5-year age group

younger than 15 years of age, but for 15-19

year olds the incidence rate was higher

(though still only 0.7 per million) (Figure

VI.1). Among the 15-19 year old popula-

tion, renal carcinomas represented the

majority (63%) of cases of renal cancer.

Average annual incidence rates for

Wilms’ tumor by single year of age are

presented in Figure VI.2 for the time peri-

ods 1976-84 and 1986-94.

The age-specific

incidence rates were highest in the first two

years of life at 21 per million, with inci-

dence rates subsequently declining to levels

less than 2 per million for children older

than 9 years of age. Age-specific rates for

the other renal cancers were much lower

than those for Wilms’ tumor. Rhabdoid

tumors of the kidney was present almost

exclusively in the first 2 years of life, with a

peak in infancy of 1.0 per million. Clear

cell sarcoma of the kidney also occurred

much less frequently than Wilms’ tumor,

with age-specific incidence in the first 4

years of life ranging between 0.4 and 0.6

per million, and with very few cases occur-

ring among children older than 3 years of

age. Renal cell carcinoma was also uncom-

mon among children of any age, with most

cases occurring in adolescents 15-19 years

of age, for which age-specific incidence rates

varied between 0.5 and 0.9 per million.

Sex-specific incidence

For the 21 year period from 1975 to

1995, renal cancer incidence rates among

children younger than 15 years of age were

minimally higher for females compared to

males (13 percent higher among females)

(Table VI.2). For Wilms’ tumor (ICCC VIa),

there was also a slight female

predominance when the overall period

1975-95 was considered. However, inci-

dence rates for Wilms’ tumor (ICCC VIa)

were the same for males and females for

the most recent period evaluated (1990-95)

(Figure VI.3). As illustrated in Figure VI.2,

females had slightly higher rates of Wilms’

tumor than males in infancy (22.6 per

Figure VI.1: Renal cancer age-specific

incidence rates by ICCC subcategory,

all races, both sexes, SEER, 1975-95

18.3

5.6

0.8

0.4

0.1

0.4

0.7

<5 5-9 10-14 15-19

Age (in years) at diagnosis

Average annual rate per million

Wilms (VIa)

Renal carcinoma (VIb)

Table VI.3: Age-specific incidence rates per

million for renal cancer by age

and race, SEER, 1975-95

All races Whites Blacks

Males

<5 years 17.8 18.2 21.5

5-9 years 4.9 4.9 6.7

10-14 years 1.2 1.1 2.2

Females

<5 years 19.1 19.6 21.4

5-9 years 6.6 6.8 7.8

10-14 years 1.2 1.3 1.6

Enumeration of the population at risk by single years of age was

available only for the census years 1980 and 1990. The US Bureau

of the Census provides intercensal population estimates by 5-year

age groups, but not by single years of age. Therefore, the

population estimates for 1980 were used in rate calculations for

cases diagnosed from 1976-84 and the 1990 estimates were used for

cases diagnosed from 1986-94.

National Cancer Institute SEER Pediatric Monograph

ICCC VIRENAL TUMORS

National Cancer Institute

SEER Pediatric Monograph

Figure VI.3: Wilms' tumor (VIa) age-adjusted* incidence rates

by sex and year of diagnosis, age <15 all races, both sexes, SEER 1975-95

6.4

7.5

7.3 7.3

8.5

8.7

8.4

7.3

1975-79 1980-84 1985-89 1990-95

Year of Diagnosis

Average annual rate per million

Males

Females

*Adjusted to the 1970 US standard population

Figure VI.2: Wilms' tumor (VIa) age-specific incidence rates

by sex, all races, SEER, 1976-84 and 1986-94

)

0123456789101112131415

Age (in years) at diagnosis

Average annual rate per million

Males

Females

Total

)

ICCC VI RENAL TUMORS

National Cancer Institute SEER Pediatric Monograph

million versus 19.7 per million, respec-

tively). Among children 3-8 years old, the

age-specific incidence rates for females

were generally equal to or greater than

rates for males. For females, but not for

males, the steady decline in Wilms’ tumor

incidence rates with increasing age after

infancy was apparent except in the fourth

year of life during which rates increased in

females to levels approaching those seen in

infancy, then subsequently declined lin-

early.

Black-white differences in incidence

Renal cancer and Wilms’ tumor (ICCC

VIa) incidence rates for the overall period

1975-95 were somewhat higher for black

children than for white children (Table

VI.2). For the two most recent five year

periods (1985-89 and 1990-95), however,

incidence rates for Wilms’ tumor were

similar for black children and white chil-

dren (Figure VI.4).

TRENDS

The age-adjusted incidence rates for

childhood renal cancers did not change

significantly during the period 1975-95.

Incidence rates for Wilms’ tumor (ICCC

VIa) varied from year to year (Figure VI.5),

but there was no trend for increase or

decrease during the 21-year period (esti-

mated annual percentage change = -0.03%).

The incidence of renal carcinoma was very

low throughout the period (Figure VI.5).

For the years 1975-79, Wilms’ tumor (ICCC

VIa) incidence rates for females (8.5 per

million) were higher than for males (6.4 per

million) (Figure VI.3). However, rates for

males rose between 1975-79 and 1980-84 to

7.5 per million, and thereafter remained

fairly stable. Rates for females declined,

particularly between 1985-89 and 1990-95,

so that females and males had the same

incidence rate (7.3 per million) during

1990-95 (Figure VI.3). Incidence rates for

Wilms’ tumor (ICCC VIa) for white children

did not vary much between each 5-6 year

time period from 1975 to 1995. Black

Figure VI.4: Wilms' tumor (VIa) age-adjusted* incidence

rates by race and year of diagnosis, age <15

both sexes, SEER, 1975-95

7.6

7.9

8.2

7.8

8.5

11.7

8.2

7.1

1975-79 1980-84 1985-89 1990-95

Year of diagnosis

Average annual rate per million

White

Black

*Adjusted to the 1970 US standard population

Figure VI.5: Trends in renal cancer age-adjusted*

incidence rates by type, age <15

all races, both sexes, SEER, 1975-95

1975 1980 1985 1990 1995

Year of diagnosis

Average annual rate per million

Wilms' Tumor (VIa)

Renal Carcinoma (VIb)

*Adjusted to the 1970 US standard population

National Cancer Institute SEER Pediatric Monograph

ICCC VIRENAL TUMORS

National Cancer Institute

SEER Pediatric Monograph

children had higher incidence rates for

Wilms’ tumor (ICCC VIa) in 1975-79 and

1980-84 (8.5 and 11.7 per million, respec-

tively), but the rates dropped for the years

1985-89 and 1990-95 (8.2 and 7.1 per

million, respectively) to levels very similar

to those for white children (Figure VI.4).

SURVIVAL

For children of all ages, both sexes, and

all racial/ethnic groups residing in the

SEER areas, the relative 5-year survival

rate for children diagnosed with Wilms’

tumor younger than 15 years of age during

1985-94 was 92%, compared with a rate of

81% among children diagnosed with this

malignancy during 1975-84. Among cases

diagnosed during 1985-94, 5-year survival

was slightly better for females (94%) than

males (91%). For 1985-94, black children

had somewhat better outcome than white

children (95% versus 92% 5-year survival)

(Figure VI.6). Children with rhabdoid

tumor are known to have a much poorer

outcome than children with Wilms’ tumor

[44], and among the small number of

children with rhabdoid tumor of the kidney

followed for survival in SEER areas (n = 8),

all either died (6) or were lost to follow-up

(2). Children with clear cell sarcoma of the

kidney are known to have a somewhat

poorer prognosis than children with Wilms’

tumor, with 6-year relapse-free survival

rates of slightly above 60% based on data

from the US National Wilms’ Tumor Study

Group [45,46]. There were only 13 children

with clear cell sarcoma of the kidney

evaluable for survival from the SEER areas

for the time period 1975-94, and their

relative 5-year survival rate was 84%. For

children and adolescents with renal carci-

nomas, 5-year relative survival rates in-

creased from 48% for cases diagnosed in

1975-84 to 83% for cases diagnosed in

1985-94 (although these estimates are

Figure VI.6: Wilms' tumor 5-year relative survival rates

by race and sex, age <15, SEER (9 Areas), 1975-84 and 1985-94

92 92

All White Black Males Females

100

Percent surviving 5 years

1975-84 1985-94

(

ICCC VI RENAL TUMORS

National Cancer Institute SEER Pediatric Monograph

Table VI.4: Current knowledge on causes of Wilms’ Tumor (WT)

Exposure or Characteristic Comments References

Race Incidence in Asians is about half that in blacks

and whites.

10,11

Aniridia, genitourinary

anomalies, WAGR

syndrome (Wilms’ tumor,

aniridia, genitourinary

abnormalities, mental

retardation), Beckwith-

Wiedemann syndrome,

Perlman syndrome, Denys-

Drash syndrome, Simpson-

Golabi-Behmel syndrome

Risk is increased in children with these

congenital anomalies and genetic conditions.

The study of children with WAGR led to the

identification of one of the WT genes.

12-22

Paternal occupation An increased risk for fathers employed as a

welder or mechanic has been reported in

several studies.

13,26,28

High birth weight Association with birth weight over 4,000 grams

has been reported in some studies.

13,29,30

Parental exposure to

pesticides

One study found an increased risk for parental

occupational exposure to pesticides. Another

study found an association with household

insect extermination.

13,27,31-33

Ionizing radiation (in utero) Prenatal diagnostic x-ray was associated with

increased risk in one study.

Maternal consumption of

coffee and tea during

pregnancy

Three studies reported association with coffee

and/or tea; another did not replicate this

finding.

31,35,36

Maternal hair dye use

during pregnancy

Use was associated with risk in one study, but

not in others.

31,36

Maternal medication use

during pregnancy

Studies reported associations with various

drugs including hormones, antibiotics,

dipyrone, metoclopramide, pethrane anesthesia

during delivery. Most of these results were

found in only a single study.

13,37,38

Maternal occupation One study found an association with job

groupings that included hairdressers,

electronic and clothing manufacturing workers,

laboratory workers, dental assistants.

26,39

Known risk factors

Factors for which

evidence is inconsistent

or limited

Factors for which

evidence is suggestive

but not conclusive

National Cancer Institute SEER Pediatric Monograph

ICCC VIRENAL TUMORS

National Cancer Institute

SEER Pediatric Monograph

based on only 23 and 29 cases, respectively,

from these two time periods).

RISK FACTORS

Despite the rarity of renal cancer in

children, a substantial body of epidemio-

logic, genetic, and molecular studies have

contributed important insights to under-

standing its pathogenesis [3,6,7]. Histori-

cally, Wilms’ tumor was thought to vary

little in incidence throughout the world and

was therefore proposed as an “index tumor”

of childhood cancer [8]. However, interna-

tional comparisons based on data through

the 1980s showed a greater than threefold

difference in age-adjusted incidence rates

among populations, with highest rates

observed in US and Nigerian blacks, fol-

lowed by somewhat lower rates in Sweden

and US whites, and lowest rates in Chinese

and other Asians [9,10]. Data for the years

1973-88 from the US showed similar ethnic

variation, with incidence in Asians about

half that in blacks, and rates for blacks

slightly higher than rates for whites [11]

(Table VI.4).

A small proportion of Wilms’ tumor

cases appear to be heritable including:

those patients with bilateral tumors, those

occurring in association with aniridia and

other congenital anomalies, and those few

cases arising in the small number of fami-

lies with one or more additional cases of

Wilms’ tumor in close family members

[12,13]. Approximately 1.5% of patients in

a large series had one or more family

members (usually siblings or cousins) with

Wilms’ tumor based on interview data [14].

Congenital disorders that have been linked

with Wilms’ tumor include: the Beckwith-

Wiedemann syndrome (an overgrowth

syndrome associated with macrosomia,

omphalocele, macroglossia, and viscerome-

galy and believed to be linked to an as yet

unidentified gene(s) at chromosome region

11p15) [14,15]; the Simpson-Golabi-Behmel

syndrome (an X-linked fetal overgrowth

disorder caused by mutations in the

glypican 3 gene) [15,16]; hemihypertrophy

as an isolated abnormality; the Perlman

and Sotos syndromes [17-19]; the Denys-

Drash syndrome (associated with muta-

tions of the Wilms’ tumor suppressor gene

WT1) [20-22]; and the WAGR syndrome

(Wilms’ tumor, aniridia, genitourinary

malformations, and mental retardation )

that results from deletion of a number of

contiguous genes on chromosome 11 includ-

ing the aniridia gene PAX6 and the WT1

gene [12]. In addition to the heritable

conditions cited above, inherited predisposi-

tion genes associated with some familial

Wilms’ tumor cases appear to exist at two

other loci (and possibly others not yet

identified) [23,24]. However, survivors of

Wilms’ tumor that is unilateral at diagnosis

are at low risk for having children with

Wilms’ tumor [25].

Most of the analytical and epidemio-

logic investigations of childhood renal

cancer have focused on Wilms’ tumor, and

very little is known about risk factors for

childhood renal carcinoma or the other

rarer childhood renal cancer subtypes.

Several epidemiological studies have inves-

tigated occupational, environmental, and

lifestyle characteristics as potential risk

factors for Wilms’ tumor, but findings to

date have been inconsistent [13,26,27]. A

few studies have suggested that children of

fathers employed as welders or mechanics

have increased risk of Wilms’ tumor [13,28],

but occupational exposure assessment was

insufficient to draw firm conclusions [26].

Limited evidence implicates high birth

weight in the etiology [29,30]. Parental

and postnatal exposures to pesticides have

also been linked with increased risk [27,31-

33], but these associations were derived

from interview data only and have not been

confirmed with studies utilizing measure-

ments. A large study in the United King-

dom has reported an association of Wilms’

tumor with exposure of the mother to

ionizing radiation from diagnostic x-rays

ICCC VI RENAL TUMORS

National Cancer Institute SEER Pediatric Monograph

during pregnancy [34.] Some [31,32,35],

but not all [36] investigations have found

associations between maternal consump-

tion of coffee and tea during pregnancy and

risk of Wilms’ tumor in offspring. Inconsis-

tent reports have also implicated maternal

hair dye use and various types of medica-

tions taken or anesthetics to which mothers

have been exposed during pregnancy

[13,37,38]. The role of maternal occupa-

tional exposures has received limited

evaluation [26,39].

Most of the reported associations

described in the preceding paragraph have

not been consistently replicated in multiple,

high quality studies in different popula-

tions. Future epidemiologic studies may

benefit from more detailed exposure assess-

ment, validated by environmental and

biologic measurements. In addition, the

role of genetic susceptibility and assess-

ment of gene-environment interaction

should be considered by evaluation of

appropriate molecular markers to better

define etiologic pathways for Wilms’ tumor.

Recurring molecular abnormalities

have been identified in the tumor cells of

two of the uncommon renal cancers that

occur in young children, rhabdoid tumor of

the kidney and congenital mesoblastic

nephroma. Rhabdoid tumor, which can

develop in the central nervous system and

extrarenal sites as well as in the kidney, is

associated with tumor cell mutations in the

INI1 gene located on chromosome 22

[40,41]. Evaluation of some children with

rhabdoid tumors has revealed germline

mutations of the INI1 gene [41]. Congeni-

tal mesoblastic nephroma is an infantile

spindle cell tumor of the kidney with low

malignant potential that is virtually identi-

cal morphologically to congenital fibrosar-

coma [42]. The tumor cells of both of these

tumors of infancy have been found to

possess fusions of the ETV6 gene (also

known as TEL) on chromosome 12 to the

NTRK3 gene on chromosome 15 [42,43].

SUMMARY

The descriptive epidemiologic features

of Wilms’ tumor have been known for a

number of years. Associated congenital

anomalies and genetic factors have also

been subject of much interest. More recent

studies have further characterized the

specific genetic loci and molecular alter-

ations involved in the development of

Wilms’ tumor. Several epidemiologic stud-

ies have investigated occupational, environ-

mental, and lifestyle factors as risk factors

for Wilms’ tumor. A number of parental

and childhood exposures have been found

to be associated with an increased risk of

Wilms’ tumor. Most of these associations

have not been replicated in multiple high

quality studies. However, some warrant

further evaluation including paternal

occupational exposures, pesticide exposure,

and certain maternal exposures during

pregnancy. Future epidemiologic studies

may benefit from the inclusion of molecular

markers that may better define etiologic

pathways for Wilms’ tumor.

Reference List

1. Charles AK, Vujanic GM, Berry PJ: Renal

tumours of childhood. Histopathology 32:293-

309, 1998.

2. Geller E, Smergel EM, Lowry PA: Renal

neoplasms of childhood. Radiol Clin North Am

35:1391-413, 1997.

3. Green D, Coppes M, Breslow N, et al: Wilms’

tumor. In Principles and Practice of Pediatric

Oncology (Pizzo P, Poplack D, eds). Philadel-

phia: Lippincott-Raven Publishers, 1997, pp

733-59.

4. Breslow N, Beckwith JB, Ciol M, et al: Age

distribution of Wilms’ tumor: Report from the

National Wilms’ Tumor Study. Cancer Res

48:1653-7, 1988.

5. Kramarova E, Stiller CA: The international

classification of childhood cancer. Int J Cancer

68:759-65, 1996.

6. Huff V, Amos CI, Douglass EC, et al: Evidence

for genetic heterogeneity in familial Wilms’

tumor. Cancer Res 57:1859-62, 1997.

7. Rahman N, Arbour L, Tonin P, et al: Evidence

for a familial Wilms’ tumour gene (FWT1) on

chromosome 17q12-q21. Nat Genet 13:461-3,

1996.

National Cancer Institute SEER Pediatric Monograph

ICCC VIRENAL TUMORS

National Cancer Institute

SEER Pediatric Monograph

8. Innis MD: Wilms’ tumour. Med J Aust 1:282,

1972.

9. Parkin D, Stiller C, Draper G, et al: Interna-

tional Incidence of Childhood Cancer. Lyon:

World Health Organization, 1988.

10. Stiller CA, Parkin DM: International varia-

tions in the incidence of childhood renal

tumours. Br J Cancer 62:1026-30, 1990.

11. Breslow N, Olshan A, Beckwith JB, et al:

Ethnic variation in the incidence, diagnosis,

prognosis, and follow-up of children with

Wilms’ tumor. J Natl Cancer Inst 86:49-51,

1994.

12. Coppes MJ, Haber DA, Grundy PE: Genetic

events in the development of Wilms’ tumor. N

Engl J Med 331:586-90, 1994.

13. Sharpe CR, Franco EL: Etiology of Wilms’

tumor. Epidemiol Rev 17:415-32, 1995.

14. Beckwith J: Macroglossia, omphalocele,

adrenal cytomegaly, gigantism and hyperplas-

tic visceromegaly. Birth Defects 5:188-, 1969.

15. Li M, Squire JA, Weksberg R: Overgrowth

syndromes and genomic imprinting: from

mouse to man. Clin Genet 53:165-70, 1998.

16. Neri G, Gurrieri F, Zanni G, et al: Clinical and

molecular aspects of the Simpson-Golabi-

Behmel syndrome [In Process Citation]. Am J

Med Genet 79:279-83, 1998.

17. Miller R, Fraumeni JJ, Manning M: Associa-

tion of Wilms’ tumor with aniridia, hemihyper-

trophy and other congenital malformations. N

Engl J Med 270:922-927, 1964.

18. Perlman M, Levin M, Wittels B: Syndrome of

fetal gigantism, renal hamartomas, and

nephroblastomatosis with Wilms’ tumor.

Cancer 35:1212-7, 1975.

19. Hersh JH, Cole TR, Bloom AS, et al: Risk of

malignancy in Sotos syndrome [see comments].

J Pediatr 120:572-4, 1992.

20. Denys P, Malvaux P, Van Den Berghe H, et al:

[Association of an anatomo-pathological

syndrome of male pseudohermaphroditism,

Wilms’ tumor, parenchymatous nephropathy

and XX/XY mosaicism]. Arch Fr Pediatr

24:729-39, 1967.

21. Drash A, Sherman F, Hartmann WH, et al: A

syndrome of pseudohermaphroditism, Wilms’

tumor, hypertension, and degenerative renal

disease. J Pediatr 76:585-93, 1970.

22. Coppes MJ, Huff V, Pelletier J: Denys-Drash

syndrome: relating a clinical disorder to

genetic alterations in the tumor suppressor

gene WT1. J Pediatr 123:673-8, 1993.

23. Rahman N, Abidi F, Ford D, et al: Confirmation

of FWT1 as a Wilms’ tumour susceptibility

gene and phenotypic characteristics of Wilms’

tumour attributable to FWT1. Hum Genet

103:547-56, 1998.

24. McDonald JM, Douglass EC, Fisher R, et al:

Linkage of familial Wilms’ tumor predisposi-

tion to chromosome 19 and a two-locus model

for the etiology of familial tumors. Cancer

Research 58:1387-90, 1998.

25. Li FP, Williams WR, Gimbrere K, et al: Heri-

table fraction of unilateral Wilms’ tumor.

Pediatrics 81:147-9, 1988.

26. Colt JS, Blair A: Parental occupational expo-

sures and risk of childhood cancer [In Process

Citation]. Environ Health Perspect 106 Suppl

3:909-25, 1998.

27. Zahm SH, Ward MH: Pesticides and childhood

cancer. Environ Health Perspect 106 Suppl

3:893-908, 1998.

28. Breslow N, Olshan A, Beckwith JB, et al:

Epidemiology of Wilms’ tumor. Med Pediatr

Oncol 21:172-81, 1993.

29. Leisenring WM, Breslow NE, Evans IE, et al:

Increased birth weights of National Wilms’

Tumor Study patients suggest a growth factor

excess. Cancer Res 54:4680-3, 1994.

30. Yeazel MW, Ross JA, Buckley JD, et al: High

birth weight and risk of specific childhood

cancers: a report from the Children’s Cancer

Group. J Pediatr 131:671-7, 1997.

31. Olshan AF, Breslow NE, Falletta JM, et al:

Risk factors for Wilms’ tumor. Report from the

National Wilms’ Tumor Study. Cancer 72:938-

44, 1993.

32. Sharpe CR, Franco EL, de Camargo B, et al:

Parental exposures to pesticides and risk of

Wilms’ tumor in Brazil. Am J Epidemiol

141:210-7, 1995.

33. Fear NT, Roman E, Reeves G, et al: Childhood

cancer and paternal employment in agricul-

ture: the role of pesticides. Br J Cancer 77:825-

9, 1998.

34. Bithell JF, Stewart AM: Pre-natal irradiation

and childhood malignancy: a review of British

data from the Oxford Survey. Br J Cancer

31:271-87, 1975.

35. Le Masters G, Bove K: Genetic/environmental

significance of multifocal nodular renal

blastema. Am J Pediatr Hematol Oncol 2:81-

87, 1980.

36. Bunin GR, Kramer S, Marrero O, et al: Gesta-

tional risk factors for Wilms’ tumor: results of

a case-control study. Cancer Res 47:2972-7,

1987.

37. Lindblad P, Zack M, Adami HO, et al: Maternal

and perinatal risk factors for Wilms’ tumor: a

nationwide nested case-control study in

Sweden. Int J Cancer 51:38-41, 1992.

38. Sharpe CR, Franco EL: Use of dipyrone during

pregnancy and risk of Wilms’ tumor. Brazilian

Wilms’ Tumor Study Group. Epidemiology

7:533-5, 1996.

ICCC VI RENAL TUMORS

National Cancer Institute SEER Pediatric Monograph

39. Bunin GR, Nass CC, Kramer S, et al: Parental

occupation and Wilms’ tumor: results of a case-

control study. Cancer Res 49:725-9, 1989.

40. Versteege I, Sevenet N, Lange J, et al: Truncat-

ing mutations of hSNF5/INI1 in aggressive

paediatric cancer. Nature 394:203-6, 1998.

41. Biegel JA, Zhou JY, Rorke LB, et al: Germ-line

and acquired mutations of INI1 in atypical

teratoid and rhabdoid tumors. Cancer Res

59:74-9, 1999.

42. Knezevich SR, Garnett MJ, Pysher TJ, et al:

ETV6-NTRK3 gene fusions and trisomy 11

establish a histogenetic link between meso-

blastic nephroma and congenital fibrosarcoma.

Cancer Res 58:5046-8, 1998.

43. Knezevich SR, McFadden DE, Tao W, et al: A

novel ETV6-NTRK3 gene fusion in congenital

fibrosarcoma. Nat Genet 18:184-7, 1998.

44. Weeks DA, Beckwith JB, Mierau GW, et al:

Rhabdoid tumor of kidney. A report of 111

cases from the National Wilms’ Tumor Study

Pathology Center. Am J Surg Pathol 13:439-58,

1989.

45. Green DM, Breslow NE, Beckwith JB, et al:

Treatment of children with clear-cell sarcoma

of the kidney: a report from the National

Wilms’ Tumor Study Group. Journal of Clinical

Oncology 12:2132-7, 1994.

46. Green DM, Breslow NE, Beckwith JB, et al:

Comparison between single-dose and divided-

dose administration of dactinomycin and

doxorubicin for patients with Wilms’ tumor: a

report from the National Wilms’ Tumor Study

Group. J Clin Oncol 16:237-45, 1998.

ICCC VII

HEPATIC TUMORS

Marc Bulterys, Marc T. Goodman, Malcolm A. Smith, Jonathan D. Buckley

National Cancer Institute

SEER Pediatric Monograph

HIGHLIGHTS

Incidence

♦ Primary neoplasms of the liver are rare in children, comprising only 1.1% of malig-

nancies for children younger than 20 years of age. In the US, 100-150 children are

diagnosed with liver cancer each year.

♦ Primary liver cancer is subdivided into the following histologic subtypes:

hepatoblastoma comprises over two-thirds of the malignant tumors of the liver in

children and adolescents (79% <15 years of age; 66% <20 years of age) and

hepatocellular carcinoma accounts for most of the remaining cases. Hepatoblastoma

occurs primarily in children younger than 5 years of age while hepatocellular

carcinoma occurs primarily after 10 years of age (Figure VII.2).

♦ The rate of hepatoblastoma was highest among infants with rates rapidly declining

with increasing age (Figure VII.3). In contrast, the incidence of hepatocellular

carcinoma increased as age increased (Figure VII.2).

♦ For those younger than 20 years of age, there was little change in liver cancer

incidence during the 21-year period, with rates between 1.4 and 1.6 per million

throughout the time period (Table VII.1).

♦ The incidence of hepatoblastoma for children younger than 15 years of age increased

during the 1975-95 period while the incidence of hepatocellular carcinoma decreased

during the same period (Figure VII.4).

Survival

♦ Five-year survival rates for children with hepatoblastoma improved from 51% to

59% between 1976-84 and 1985-94 (Figure VII.5). Survival rates were substantially

lower for children and adolescents with hepatocellular carcinoma, with an improve-

ment in 5-year survival rates from 31% for the years 1976-84 to 42% for the years

1985-94 (Figure VII.5).

Risk factors

♦ The etiology of hepatoblastoma is as yet unknown but there are some tantalizing

clues (Table VII.2).

INTRODUCTION

Primary neoplasms of the liver are rare

in children, comprising only 1.1% of malig-

nancies in SEER areas for children younger

than 20 years of age. The ICCC category

for liver cancers (VII) is subdivided into the

following histologic subtypes:

hepatoblastoma (VIIa), hepatic carcinomas

(hepatocellular carcinoma) (VIIb), and

“unspecified” tumors of the liver (VIIc) [1].

In the US, 100-150 children younger than

20 years of age are diagnosed with hepatic

tumors each year. Hepatoblastoma com-

prises over two-thirds of the malignant

tumors of the liver in children (79%

younger than 15 years of age; 66% younger

than 20 years of age) and hepatocellular

carcinoma accounts for most of the remain-

ing cases. Most patients with

hepatoblastoma are younger than 4 years

of age at diagnosis, while hepatocellular

carcinoma occurs primarily after 10 years

of age.

ICCC VII

National Cancer Institute

SEER Pediatric Monograph

HEPATIC TUMORS

INCIDENCE

During the 21-year period from 1975

through 1995, there were 316 children

younger than 20 years of age in SEER

areas who were diagnosed with a primary

liver cancer, with 262 (83%) of these chil-

dren being younger than 15 years of age at

the time of diagnosis. Figure VII.1 shows

that the majority of these cancers were

hepatoblastomas, with the remainder being

almost exclusively hepatocellular carcino-

mas. For the entire 21-year period,

hepatoblastoma represented 79% of the

liver cancers for the younger than 15 year

age group, although for the most recent 6-

year period (1990-95) hepatoblastoma

accounted for an even higher proportion

(90%) of childhood liver cancers.

Age-specific incidence

The incidence rates for hepatoblastoma

and hepatocellular carcinomas were very

age-dependent. Among children younger

than 5 years of age, over 95% of liver

cancers were hepatoblastoma, whereas

hepatoblastoma was distinctly uncommon

for older age groups (Figure VII.2). Within

the younger than 5-year age group, the rate

of hepatoblastoma was highest among

infants with rates rapidly declining with

increasing age (Figure VII.3).

During the

most recent period (1986-94), the incidence

rate during infancy was approximately 11.2

per million. In contrast to the age-inci-

dence relationship for hepatoblastoma, the

incidence of hepatocellular carcinoma

increased with each successive 5-year age

group, with rates for 15-19 year olds (0.9

per million) being substantially higher than

for any of the younger age groups (Figure

VII.2).

Figure VII.1: Distribution of liver cancer

by histology and age, all races

both sexes, SEER 1975-95

0.4

Hepatoblastoma

Hepatocellular

Other

0 20406080100

Percent of total liver cancer

<15 years

<20 years

carcinoma

Figure VII.2: Liver cancer age-specific

incidence rates by histology and age

all races, both sexes, SEER, 1986-95

4.6

0.2

0.1

0.04

0.2

0.3

0.9

<5 5-9 10-14 15-19

Age in years

Average annual rate per million

Hepatoblastoma

Hepatocellular carcinoma

4.8

0.4

1.0

Enumeration of the population at risk by single years of age was

available only for the census years 1980 and 1990. The US Bureau

of the Census provides intercensal population estimates by 5-year

age groups, but not by single years of age. Therefore, the

population estimates for 1980 were used in rate calculations for

cases diagnosed from 1976-84 and the 1990 estimates were used for

cases diagnosed from 1986-94.

ICCC VII

National Cancer Institute SEER Pediatric Monograph

HEPATIC TUMORS

Sex and race-specific incidence

For children younger than 15 years of

age, the incidence of liver cancers was

slightly higher in males than females

(male:female ratio = 1.2) and somewhat

lower in black children compared with

white children (1.3 per million versus 1.6

per million). For children younger than 20

years of age, incidence rates were similar

for blacks and whites (1.4 per million

versus 1.5 per million) and were slightly

higher for males than females (male:female

= 1.2). The incidence of hepatoblastoma

was slightly higher in males than females

(male:female = 1.2), while the incidence of

hepatocellular carcinoma was similar in

both sexes (male:female = 1).

TRENDS

The incidence for total liver cancers in

children younger than 15 years of age

increased slightly from 1975 to 1995. The

rate was 1.4 per million for 1975-79 and

increased to 1.7 per million for 1990-95

(Table VII.1). For children younger than 20

years of age, there was little change in liver

cancer incidence during the 21-year period,

with rates between 1.4 and 1.6 per million

throughout the time period (Table VII.1).

The incidence of hepatoblastoma in-

creased markedly during the 1975-95

Figure VII.3: Hepatoblastoma and hepatocellular

carcinoma age-specific incidence rates

all races, both sexes

SEER, 1976-84, 1986-94

'''

+++

01234567891011121314151617181920

Age (in years) at diagnosis

Average annual rate per million

1976-84

1986-94

Table VII.1: Age-adjusted* incidence rates per million of liver cancers by age group, type,

and time period, all races, both sexes, SEER, 1975-95

<15 Years

Diagnosis ICCC

Category

White

Male

Black

Male

White

Female

Black

Female

X a-e All 12.3 3.2 9.0 10.8

Xc Gonadal germ cell tumors 9.1 1.2 4.5 5.6

Testis 9.1 1.2 - -

Ovary - - 4.5 5.6

X a,b,d,e Other than gonadal germ cell

tumors

3.2 2.0 4.5 5.2

*Adjusted to the 1970 US standard population

ICCC X

129

National Cancer Institute SEER Pediatric Monograph

GERM CELL, TROPHOBLASTIC AND OTHER GONADAL NEOPLASMS

Black-white differences in incidence

Black children had a lower incidence of

germ cell tumors than white children (7.0

vs. 10.7 per million). This difference was

primarily the result of a lower rate of

gonadal germ cell tumors among blacks

than whites. Table X.3 shows the incidence

rates for gonadal germ cell tumors for

children younger than 20 years of age for

the years 1975 to 1995 by race and sex.

Remarkably, the lower rates of gonadal

germ cell tumors among black children

were restricted to males. For children

younger than 20 years of age, black males

had a rate of testicular germ cell tumor

that was only one-seventh that for white

males (1.2 versus 9.1 per million), while

black females had slightly higher rates of

ovarian germ cell tumors than white fe-

males (5.6 versus 4.5 per million). The low

rate of testicular germ cell tumors observed

among young black males is consistent

with the reported low incidence for testicu-

lar cancer among adult black males [20-22].

Age-specific incidence

Figure X.1 shows the age-specific inci-

dence of GCTOG tumors by single year of

age and sex.

Rates were relatively high in

the first year of life and then declined to

very low levels before increasing at age 8-

12 years for females and at age 11-14 years

for males. Incidence continued to increase

for both males and females up through age

19. The distribution of tumor types by age

was distinctive for males and females.

For males, the incidence rates of testicu-

lar (Xc) and non-CNS extragonadal (Xb)

germ cell tumors were similar during the

first year of life at approximately 9 per

Figure X.1: GCTOG age-specific incidence rates

by sex, all races, SEER, 1986-94

)

01234567891011121314151617181920

Age (in years) at diagnosis

Average annual rate per million

Males

Females

)

GCTOG - Germ cell, trophoblastic and other gonadal

Enumeration of the population at risk by single years of age was

available only for the census year 1990. The US Bureau of the

Census provides intercensal population estimates by 5-year age

groups, but not by single years of age. Therefore, the population

estimates for 1990 were used in rate calculations for cases

diagnosed from 1986-94.

ICCC X

130

National Cancer Institute

SEER Pediatric Monograph

GERM CELL, TROPHOBLASTIC AND OTHER GONADAL NEOPLASMS

million, and then declined to very low levels

by age 4 years (Figure X.2). Between ages

4 and 15 the rates remained very low, but

between ages 15 and 19 years of age, the

incidence rates increased dramatically.

For females, non-CNS extragonadal

germ cell tumors (Xb) accounted for the

vast majority of cases in the first year of

life, with ovarian germ cell tumors (Xc)

being extremely rare (Figure X.3). Most of

the extragonadal germ cell tumors arising

in the first year of life occurred in pelvic

soft tissue (e.g., the sacrococcygeal region)

and in the retroperitoneum. Gonadal germ

cell tumors (Xc) began to increase in inci-

dence for females at age 8-9 years, while

gonadal carcinomas (Xd) began to increase

after age 12. By age 19, the rate of gonadal

carcinomas (Xd) was similar to ovarian

germ cell tumors (Xc) in females. For

males, the rate of testicular germ cell

tumors (Xc) at age 19 was substantially

higher than that observed for ovarian germ

cell tumors among 19 year old females

(44.5 versus 10.4 per million for age 19).

TRENDS

The age-adjusted incidence rates for

GCTOG tumors increased between 1975-79

and 1990-95 from 3.7 to 5.4 per million for

children younger than 15 years of age and

from 8.5 to 12.0 per million for those

younger than 20 years of age (Table X.4).

For both males and females younger than

15 years of age, the increase in incidence

primarily resulted from higher rates for

intracranial and intraspinal germ cell

tumors (Xa) and for non-CNS extragonadal

germ cell tumors (Xb), while the rates of

gonadal germ cell tumors (Xc) did not

increase. The increased incidence of non-

CNS extragonadal tumors (Xb) for both

males and females was due in large mea-

sure to an increase in incidence in the first

Figure X.2: GCTOG age-specific incidence rates

by selected ICCC subgroups, males

all races, SEER, 1986-94

""""""

'''''

+++

01234567891011121314151617181920

Age (in years) at diagnosis

Average annual rate per million

Xa = Intracranial/intraspinal germ cell tumors

Xb = Other non-gonadal germ cell tumors

Xc = Testicular germ cell tumors

GCTOG - Germ cell, trophoblastic and other gonadal

Figure X.3: GCTOG age-specific incidence rates

by selected ICCC subgroups

females, all races, SEER, 1986-94

,,,,,,,,,

'''

+++

01234567891011121314151617181920

Age (in years) at diagnosis

Average annual rate per million

Xa = Intracranial/intraspinal germ cell tumors

Xb = Other non-gonadal germ cell tumors

Xc = Ovarian germ cell tumors

Xd = Gonadal carcinoma

GCTOG - Germ cell, trophoblastic and other gonadal

ICCC X

131

National Cancer Institute SEER Pediatric Monograph

GERM CELL, TROPHOBLASTIC AND OTHER GONADAL NEOPLASMS

year of life. This increase in non-CNS

extragonadal malignant tumors among

infants must be interpreted with caution,

because non-malignant sacrococcygeal

teratomas diagnosed in the newborn period

outnumber malignant teratomas [3,23-25],

and because careful inspection of mature

and immature sacrococcygeal teratomas

may show microscopic foci of yolk sac tumor

[26,27]. Since nonmalignant sacrococcygeal

teratomas are not reported and yolk sac

tumors are reported, the increase in inci-

dence in the first year of life may be the

result of increasing recognition by patholo-

gists of the need for careful scrutiny of

apparently non-malignant sacrococcygeal

teratomas. Almost all of the increase in the

first year of life for females was in malig-

nant teratomas/embryonal teratomas.

An increase in the age-adjusted inci-

dence for GCTOG tumors was also observed

for both sexes among those younger than 20

years of age. For males younger than 20

years of age, the increase in incidence was

from 9.1 to 12.2 per million, with most of

the increase attributed to intracranial and

intraspinal germ cell tumors (Category Xa)

and to testicular germ cell tumors (Category

Xc). For females, the increase was from 7.8

to 11.7 per million, with most of the increase

attributable to ovarian germ cell tumors

Table X.4:Average annual age-adjusted

incidence rates per million for germ cell

trophoblastic, and other gonadal cancers by sex, age, subtype, and

time period, all races, SEER, 1975-95

Sex/Age Group Years X(total) Xa

Total <15 1975-79 3.7 0.5 0.7 2.2 0.1 0.2

1980-84 4.8 0.9 1.1 2.6 0.1 0.1

1985-89 4.8 0.7 1.3 2.7 0.1 0.1

1990-95 5.4 1.5 1.4 2.1 0.3 0.0

Males <15

1975-79 3.1 0.5 0.6 1.8 0.1 0.1

1980-84 4.1 1.2 0.9 2.0 0.0 0.0

1985-89 4.1 1.1 0.8 2.2 0.0 0.1

1990-95 4.4 1.9 1.1 1.4 0.0 0.0

Females <15

1975-79 4.3 0.4 0.8 2.6 0.2 0.4

1980-84 5.5 0.6 1.3 3.3 0.2 0.2

1985-89 5.6 0.4 1.8 3.2 0.1 0.1

1990-95 6.4 1.2 1.9 2.7 0.7 0.0

Total <20

1975-79 8.5 0.6 1.4 5.4 0.8 0.3

1980-84 9.6 0.9 1.5 6.4 0.7 0.2

1985-89 10.7 1.1 1.7 6.6 1.1 0.2

1990-95 12.0 1.9 1.6 6.8 1.6 0.2

Males <20

1975-79 9.1 1.0 1.2 6.9 0.1 0.1

1980-84 11.0 1.2 1.5 8.1 0.1 0.1

1985-89 11.4 1.7 1.6 7.8 0.1 0.1

1990-95 12.2 2.6 1.3 8.1 0.1 0.1

Female <20 1975-79 7.8 0.3 1.6 3.9 1.5 0.6

1980-84 8.1 0.6 1.5 4.5 1.3 0.3

1985-89 10.0 0.4 1.8 5.3 2.2 0.3

1990-95 11.7 1.1 1.8 5.3 3.2 0.3

Adjusted to the 1970 US standard population

Xa = Intracranial and intraspinal germ cell tumors; Xb = Other and unspecified

non-gonadal germ cell tumors; Xc = Gonadal (ovarian and testicular) germ cell tumors;

Xd = Gonadal carcinoma; Xe = Other and unspecified malignant gonadal tumors.

ICCC X

132

National Cancer Institute

SEER Pediatric Monograph

GERM CELL, TROPHOBLASTIC AND OTHER GONADAL NEOPLASMS

(Category Xc) and to ovarian carcinomas

(Category Xd). Because of the larger num-

ber of cases in the 15-19 year group com-

pared to the younger than 15 year group,

the trends for those younger than 20 years

of age are primarily determined by trends

for the 15-19 year age group.

Figure X.4 illustrates the increase in

incidence of testicular germ cell tumors (Xc)

for the 15-19 year age group between 1975-

79 (22 per million) and 1990-1995 (28 per

million). The increase in incidence of

testicular germ cell tumors for those 15-19

years of age is reminiscent of the increase

in testicular cancer among adult males.

Over the past 30-40 years, increased rates

of testicular cancer have been reported

from developed countries throughout the

world, including the United States [21,22],

European countries [28], Australia [29],

and New Zealand [30].

The overall rate of GCTOG tumors for

females aged 15-19 increased markedly

from 1975-79 to 1990-95 (Figure X.5), but

much of the increase was attributable to

the inclusion of borderline tumors of the

ovary which were not reportable cancers for

the entire time period. Figure X.5 shows

the overall rate for ICCC X for females both

with and without the borderline tumors.

With the borderline tumors excluded, the

overall rate increased only slightly between

1975-79 and 1990-95 (Figure X.5), and this

increase was driven by an increased inci-

dence of ovarian germ cell tumors (8 per

million for 1975-79 to 13 per million for

1990-95). An increased incidence for ova-

rian germ cell tumors in adults has also

been reported [31,32].

SURVIVAL

For the period from 1985 to 1994,

the 5-year survival rate for patients

Figure X.4: Trends in GCTOG age-specific

incidence rates by selected ICCC subgroups

age 15-19, males, all races, SEER 1975-95

1975-79 1980-84 1985-89 1990-95

Year of diagnosis

Average annual rate per million

Total X

Xa = Intracranial and intraspinal germ cell tumors

Xb = Other and unspecified non-gonadal germ cell tumors

Xc = Gonadal (testicular) germ cell tumors

GCTOG - Germ cell, trophoblastic and other gonadal

Figure X.5: Trends in GCTOG age-specific

incidence rates by selected ICCC subgroups

age 15-19, females, all races, SEER, 1975-95

)

1975-79 1980-84 1985-89 1990-95

Year of diagnosis

Average annual rate per million

Total X

excluding borderline

Xb = Other & unspecified non-gonadal germ cell tumors

Xc = Gonadal (ovarian) germ cell tumors

Xd - borderline

Xd - other

)

GCTOG - Germ cell, trophoblastic and other gonadal

ICCC X

133

National Cancer Institute SEER Pediatric Monograph

GERM CELL, TROPHOBLASTIC AND OTHER GONADAL NEOPLASMS

younger than 20 years of age with germ cell

tumors was 87% (Figure X.6). Survival

rates were better for the 15-19 year olds (5-

year survival, 90%) than for the younger

than 15 year olds (5-year survival, 84%).

Other observations about outcome for

children with germ cell tumors are

illustrated in Figures X.6 and X.7 and

include:

• For those younger than 20 years of

age, females had slightly higher 5-

year survival rates than males, and

whites had somewhat higher 5-year

survival rates than blacks (Figure

X.6).

• Survival for patients younger than

20 years of age was better for

gonadal germ cell tumors (ICCC Xc)

than for tumors arising at “other

and unspecified” sites (ICCC

Category Xb), with 5–year survival

rates of 94% and 71%, respectively

in 1985-1994. Outcome was similar

for patients younger than 20 years

of age with intracranial germ cell

tumors (ICCC Xa) and with tumors

arising at “other and unspecified”

sites (ICCC Xb), with both groups

having survival rates for 1985-94 of

approximately 70% (Figure X.7).

• Increasing survival rates were

observed between 1975-84 and

1985-94 for each subgroup of the

ICCC for patients younger than 20

years if age. The overall 5-year

relative survival rate for all

subgroups combined increased from

77% to 87% (Figure X.6). The

largest increase in survival was for

tumors arising at other and

unspecified sites (ICCC Xb): 58

percent compared to 72 (Figure X.7).

Figure X.6: Germ-cell tumor 5-year relative survival

rates by sex, race, age, and time period

SEER (9 areas), 1975-84 and 1985-94

GCTOG - Germ cell, trophoblastic and other gonadal

82 82 82

Total Male Female White Black <5 5-9 10-1415-19

100

Percent surviving 5 years

1975-84

1985-94

Sex

Race Age

Figure X.7: GCTOG tumor 5-year relative survival rates

by sub-group, sex and time period, all races

SEER (9 areas), 1975-84 and 1985-94

# - < 25 cases - rate not shown

GCTOG - Germ cell, trophoblastic and other gonadal

Xa = Intracranial and intraspinal germ cell tumors; Xb = Other and unspecified non-gonadal germ cell

tumors; Xc = Gonadal (ovarian and testicular) germ cell tumors; Xd = Gonadal carcinoma;

Xe - Other and unspecified malignant gonadal tumors.

82 82 82

Xa Xb Xc Male Female Xd Xe

100

Percent surviving 5 years

1975-84

1985-94

ICCC X

134

National Cancer Institute

SEER Pediatric Monograph

GERM CELL, TROPHOBLASTIC AND OTHER GONADAL NEOPLASMS

Table X.5: Current knowledge on causes of childhood malignant germ cell tumors (MGCT)

Exposure or Characteristic Comments

References

Cryptorchidism Risk is increased 2.5 - 11-fold. The contralateral as well

as ipsilateral testis is at increased risk.

34-36

High maternal hormone levels

during pregnancy

Use of oral contraceptives during pregnancy, high pre-

pregnancy weight, bleeding, hyperemesis and spotting

indicate high hormone levels.

34,35,37-39

Family history of germ cell

tumor

When malignant germ cell tumors occur in the same

family, they are usually of the same histologic type.

40,41

Hernia Central nervous system and genitourinary anomalies

have also been observed in germ cell tumor patients.

34,35,42

Pre-term birth Excess risk was not explained by cryptorchidism. 43,44

Trauma The causality of this association is not clear. Trauma

may result in closer scrutiny and earlier detection of an

existing tumor.

45-47

Virus infection, e.g., mumps,

cytomegalovirus, Epstein-B

virus, and parvovirus B19

48-52

High birth weight 35,43,44

Prenatal X-ray exposure 43,53

Parental occupation Associations have been observed with maternal

employment in the medical field, paternal employment

in service stations and aircraft industry, and paternal

exposure to x-rays, maternal exposure to solvents,

plastic and resin fumes.

44,54,55

Constitutional chromosome

abnormalities, particularly sex

chromosome abnormalities

(e.g., Klinefelter syndrome

(47,XXY), inverted Y)

56-60

Known risk factors

Factors for which evidence

is suggestive but not

conclusive

Factors for which evidence

is inconsistent or limtied

ICCC X

135

National Cancer Institute SEER Pediatric Monograph

GERM CELL, TROPHOBLASTIC AND OTHER GONADAL NEOPLASMS

The increase in survival for this

subgroup, ICCC Xb, was dramatic

for children younger than 5 years of

age; the survival rate increased from

38% to 86%.

• The increase in survival between

1975-84 and 1985-94 was similar for

ovarian and testicular germ cell

tumors (Figure X.7). Both increased

from 82% to 93-94%.

The improvement in outcome observed in

the more recent period for children with

germ cell tumors likely represents the

widespread application of platinum-based

chemotherapy, which is particularly

effective against germ cell tumors [33].

RISK FACTORS

The etiology of malignant germ cell

tumors is poorly understood.

Cryptorchidism is the only confirmed risk

factor for testicular germ cell tumors (see

Table X.5 for references). Although rare,

testicular cancer coincidence in father and

son, and in male siblings has been

reported, implying a genetic contribution

in the disease origination. Suggested risk

factors for malignant germ cell tumors,

mainly based on findings from studies of

testicular cancer among adult

populations, include maternal exogenous

hormone use and high endogenous

hormone level during pregnancy, pre-term

birth, high birth weight, hernia, trauma,

pre-natal X-ray exposure, virus infection,

parental occupation and occupational

exposures, and certain constitutional

chromosome abnormalities.

SUMMARY

The ICCC Diagnostic Group X for

GCTOG tumors represents less than 4% of

tumors among children younger than 15

years of age. However, for the 15-19 year

age group, these tumors account for a much

higher proportion (approximately 16%) of

cancer cases. The age-incidence pattern for

the group of GCTOG tumors is character-

ized by relatively high rates in the first

year of life, followed by much lower rates

until puberty, when incidence begins to

increase and reaches rates greater than

those in the first year of life. For males, the

majority of testicular cancers occurring

before age 15 years are diagnosed in the

first 4 years of life. However, because the

incidence of testicular germ cell tumors

increases rapidly after age 15, the vast

majority of testicular cancer cases among

those younger than 20 years of age develop

among 15-19 year olds. Black males have a

much lower incidence of testicular germ cell

tumors than white males, while black

females and white females have similar

rates for ovarian germ cell tumors.

The distinctive nature of the germ cell

tumors of infants and young children

compared to those of adolescents and young

adults complicates analyses of trends in

incidence for the children younger than 20

years of age. However, over the past 20

years there has been a small absolute

increase in incidence for germ cell tumors

for children younger than age 15 years,

with most of the increase due to higher

rates for extragonadal germ cell tumors.

Among children younger than 20 years of

age, the incidence of GCTOG tumors has

increased. The increase has been primarily

driven by higher rates for gonadal germ cell

tumors among 15-19 year olds and by

higher rates for gonadal carcinomas among

15-19 year old females. The latter increase

is attributable to changes in reporting of

ovarian tumors during this time period,

specifically inclusion of borderline tumors of

the ovary. The increases in gonadal germ

cell tumors for adolescents 15-19 years of

age mirrors that observed for young adults

with germ cell tumors. The onset of higher

rates in males and females at the time of

puberty as well as results from epidemio-

logical studies suggest a contributory role

ICCC X

136

National Cancer Institute

SEER Pediatric Monograph

GERM CELL, TROPHOBLASTIC AND OTHER GONADAL NEOPLASMS

for hormonal influences, although the

nature of these influences remains to be

elucidated.

Reference List

1. Dehner LP: Gonadal and extragonadal germ cell

enoplasia of childhood. Hum Pathol 14?493-511,

1983.

2. Castleberry R, Cushing B, Perlman E, et al:

Germ cell tumors. In Principles and Practice of

Pediatric Oncology (Pizzo P, Poplack D, eds).

Philadelphia: Lippincott-Raven, 1997, pp 921-

945.

3. Pinkerton CR: Malignant germ cell tumours in

childhood. Eur J Cancer 33:895-901; discussion

901-2, 1997.

4. Kramarova E, Stiller CA: The international

classification of childhood cancer. Int J Cancer

68:759-65, 1996.

5. Riopel MA, Spellerberg A, Griffin CA, et al:

Genetic analysis of ovarian germ cell tumors by

comparative genomic hybridization. Cancer Res

58:3105-10, 1998.

6. Dal Cin P, Dei Tos AP, Qi H, et al: Immature

teratoma of the pineal gland with isochromo-

some 12p. Acta Neuropathol (Berl) 95:107-10,

1998.

7. Lemos JA, Barbieri-Neto J, Casartelli C:

Primary intracranial germ cell tumors without

an isochromosome 12p. Cancer Genet Cytogenet

100:124-8, 1998.

8. de Bruin TW, Slater RM, Defferrari R, et al:

Isochromosome 12p-positive pineal germ cell

tumor. Cancer Res 54:1542-4, 1994.

9. Shen V, Chaparro M, Choi BH, et al: Absence of

isochromosome 12p in a pineal region malignant

germ cell tumor. Cancer Genet Cytogenet

50:153-60, 1990.

10. Dal Cin P, Drochmans A, Moerman P, et al:

Isochromosome 12p in mediastinal germ cell

tumor. Cancer Genet Cytogenet 42:243-51, 1989.

11. Aly MS, Dal Cin P, Jiskoot P, et al: Competitive

in situ hybridization in a mediastinal germ cell

tumor. Cancer Genet Cytogenet 73:53-6, 1994.

12. Oosterhuis JW, Castedo SM, de Jong B, et al:

Ploidy of primary germ cell tumors of the testis.

Pathogenetic and clinical relevance. Lab Invest

60:14-21, 1989.

13. Silver SA, Wiley JM, Perlman EJ: DNA ploidy

analysis of pediatric germ cell tumors. Mod

Pathol 7:951-6, 1994.

14. Perlman EJ, Valentine MB, Griffin CA, et al:

Deletion of 1p36 in childhood endodermal sinus

tumors by two-color fluorescence in situ hybrid-

ization: a pediatric oncology group study. Genes

Chromosomes Cancer 16:15-20, 1996.

15. Perlman EJ, Cushing B, Hawkins E, et al:

Cytogenetic analysis of childhood endodermal

sinus tumors: a Pediatric Oncology Group study.

Pediatr Pathol 14:695-708, 1994.

16. Rodriguez E, Houldsworth J, Reuter VE, et al:

Molecular cytogenetic analysis of i(12p)-

negative human male germ cell tumors. Genes

Chromosomes Cancer 8:230-6, 1993.

17. Bosl GJ, Ilson DH, Rodriguez E, et al: Clinical

relevance of the i(12p) marker chromosome in

germ cell tumors. J Natl Cancer Inst 86:349-55,

1994.

18. Mostert MC, Verkerk AJ, van de Pol M, et al:

Identification of the critical region of 12p over-

representation in testicular germ cell tumors of

adolescents and adults. Oncogene 16:2617-27,

1998.

19. van Echten J, Oosterhuis JW, Looijenga LH, et

al: No recurrent structural abnormalities apart

from i(12p) in primary germ cell tumors of the

adult testis. Genes Chromosomes Cancer

14:133-44, 1995.

20. Tulinius H, Day N, Muir CS: Rarity of testis

cancer in Negroes. Lancet 1:35-6, 1973.

21. Spitz MR, Sider JG, Pollack ES, et al: Incidence

and descriptive features of testicular cancer

among United States whites, blacks, and

Hispanics, 1973-1982. Cancer 58:1785-90, 1986.

22. Brown LM, Pottern LM, Hoover RN, et al:

Testicular cancer in the United States: trends in

incidence and mortality. Int J Epidemiol 15:164-

70, 1986.

23. Hawkins EP: Germ cell tumors. Am J Clin

Pathol 109:S82-8, 1998.

24. Malogolowkin MH, Mahour GH, Krailo M, et al:

Germ cell tumors in infancy and childhood: a

45-year experience. Pediatr Pathol 10:231-41,

1990.

25. Marsden HB, Birch JM, Swindell R: Germ cell

tumours of childhood: a review of 137 cases. J

Clin Pathol 34:879-83, 1981.

26. Hawkins E, Issacs H, Cushing B, et al: Occult

malignancy in neonatal sacrococcygeal terato-

mas. A report from a Combined Pediatric

Oncology Group and Children’s Cancer Group

study. Am J Pediatr Hematol Oncol 15:406-9,

1993.

27. Gilcrease MZ, Brandt ML, Hawkins EP: Yolk

sac tumor identified at autopsy after surgical

excision of immature sacrococcygeal teratoma. J

Pediatr Surg 30:875-7, 1995.

28. Bergstrom R, Adami HO, Mohner M, et al:

Increase in testicular cancer incidence in six

European countries: a birth cohort phenomenon.

J Natl Cancer Inst 88:727-33, 1996.

29. Stone JM, Cruickshank DG, Sandeman TF, et

al: Trebling of the incidence of testicular cancer

in victoria, Australia (1950-1985). Cancer

68:211-9, 1991.

30. Pearce N, Sheppard RA, Howard JK, et al: Time

trends and occupational differences in cancer of

ICCC X

137

National Cancer Institute SEER Pediatric Monograph

GERM CELL, TROPHOBLASTIC AND OTHER GONADAL NEOPLASMS

the testis in New Zealand. Cancer 59:1677-82,

1987.

31. Walker AH, Ross RK, Pike MC, et al: A possible

rising incidence of malignant germ cell tumours

in young women. Br J Cancer 49:669-72, 1984.

32. dos Santos Silva I, Swerdlow AJ: Ovarian germ

cell malignancies in England: epidemiological

parallels with testicular cancer. Br J Cancer

63:814-8, 1991.

33. Mann JR, Raafat F, Robinson K, et al:

UKCCSG’s germ cell tumour (GCT) studies:

improving outcome for children with malignant

extracranial non-gonadal tumours—carboplatin,

etoposide, and bleomycin are effective and less

toxic than previous regimens. United Kingdom

Children’s Cancer Study Group. Med Pediatr

Oncol 30:217-27, 1998.

34. Depue RH, Pike MC, Henderson BE: Estrogen

exposure during gestation and risk of testicular

cancer. J Natl Cancer Inst 71:1151-5, 1983.

35. Henderson BE, Benton B, Jing J, et al: Risk

factors for cancer of the testis in young men. Int

J Cancer 23:598-602, 1979.

36. Strader CH, Weiss NS, Daling JR, et al: Cryp-

torchism, orchiopexy, and the risk of testicular

cancer. Am J Epidemiol 127:1013-8, 1988.

37. Swerdlow AJ, Stiller CA, Wilson LM: Prenatal

factors in the aetiology of testicular cancer: an

epidemiological study of childhood testicular

cancer deaths in Great Britain, 1953-73. J

Epidemiol Community Health 36:96-101, 1982.

38. Walker AH, Ross RK, Haile RW, et al: Hormonal

factors and risk of ovarian germ cell cancer in

young women. Br J Cancer 57:418-22, 1988.

39. Schottenfeld D, Warshauer ME, Sherlock S, et

al: The epidemiology of testicular cancer in

young adults. Am J Epidemiol 112:232-46, 1980.

40. Klein FA, Pollack MS, Sogani PC, et al: Father-

son testicular (teratocarcinoma) malignancy.

Urology 22:281-3, 1983.

41. Dieckmann KP, Becker T, Jonas D, et al:

Inheritance and testicular cancer. Arguments

based on a report of 3 cases and a review of the

literature. Oncology 44:367-77, 1987.

42. Morrison AS: Cryptorchidism, hernia, and

cancer of the testis. J Natl Cancer Inst 56:731-3,

1976.

43. Brown LM, Pottern LM, Hoover RN: Prenatal

and perinatal risk factors for testicular cancer.

Cancer Res 46:4812-6, 1986.

44. Shu XO, Nesbit ME, Buckley JD, et al: An

exploratory analysis of risk factors for childhood

malignant germ-cell tumors: report from the

Childrens Cancer Group (Canada, United

States). Cancer Causes Control 6:187-98, 1995.

45. Haughey BP, Graham S, Brasure J, et al: The

epidemiology of testicular cancer in upstate

New York. Am J Epidemiol 130:25-36, 1989.

46. Swerdlow AJ, Huttly SR, Smith PG: Testicular

cancer and antecedent diseases. Br J Cancer

55:97-103, 1987.

47. Strader CH, Weiss NS, Daling JR: Vasectomy

and the incidence of testicular cancer. Am J

Epidemiol 128:56-63, 1988.

48. Mueller N, Hinkula J, Wahren B: Elevated

antibody titers against cytomegalovirus among

patients with testicular cancer. Int J Cancer

41:399-403, 1988.

49. Heinzer H, Dieckmann KP, Huland E: Virus-

related serology and in situ hybridization for

the detection of virus DNA among patients with

testicular cancer. Eur Urol 24:271-6, 1993.

50. Birch JM, Marsden HB, Swindell R: Pre-natal

factors in the origin of germ cell tumours of

childhood. Carcinogenesis 3:75-80, 1982.

51. Oliver RT: Atrophy, hormones, genes and

viruses in aetiology germ cell tumours. Cancer

Surv 9:263-86, 1990.

52. Gray A, Guillou L, Zufferey J, et al: Persistence

of parvovirus B19 DNA in testis of patients

with testicular germ cell tumours. J Gen Virol

79:573-9, 1998.

53. Loughlin JE, Robboy SJ, Morrison AS: Risk

factors for cancer of the testis [letter]. N Engl J

Med 303:112-3, 1980.

54. Kardaun JW, Hayes RB, Pottern LM, et al:

Testicular cancer in young men and parental

occupational exposure. Am J Ind Med 20:219-27,

1991.

55. Johnston HE, Mann JR, Williams J, et al: The

Inter-Regional, Epidemiological Study of

Childhood Cancer (IRESCC): case-control study

in children with germ cell tumours. Carcinogen-

esis 7:717-22, 1986.

56. Dmitrovsky E, Bosl GJ, Chaganti RS: Clinical

and genetic features of human germ cell cancer.

Cancer Surv 9:369-86, 1990.

57. Surti U, Hoffner L, Chakravarti A, et al: Genet-

ics and biology of human ovarian teratomas. I.

Cytogenetic analysis and mechanism of origin.

Am J Hum Genet 47:635-43, 1990.

58. Teter J, Boczkowski K: Occurrence of tumors in

dysgenetic gonads. Cancer 20:1301-10, 1967.

59. Atkin NB, Baker MC: X-chromatin, sex chromo-

somes, and ploidy in 37 germ cell tumors of the

testis. Cancer Genet Cytogenet 59:54-6, 1992.

60. Hasle H, Mellemgaard A, Nielsen J, et al:

Cancer incidence in men with Klinefelter

syndrome. Br J Cancer 71:416-20, 1995.

138

National Cancer Institute

SEER Pediatric Monograph

ICCC XI

139

National Cancer Institute

SEER Pediatric Monograph

CARCINOMAS AND OTHER MALIGNANT EPITHELIAL NEOPLASMS

Leslie Bernstein, James G. Gurney

HIGHLIGHTS

Incidence

♦ Among children, particularly before the adolescent years, carcinomas are very rare.

♦ In the US, approximately 1,050 children and adolescents younger than 20 years of

age are diagnosed with carcinomas each year, of which approximately 350 are

thyroid carcinomas and 300-350 are melanomas.

♦ All of the carcinomas combined comprised 9.2% of cancer in children younger than

20. The majority of the carcinomas were either thyroid carcinomas (35.5%) or

melanomas (30.9%). Adrenocortical carcinomas (1.3%), nasopharyngeal carcinomas

(4.5%), and other skin carcinomas (0.5%) combined for only a small proportion of

the total, while other and unspecified carcinomas comprised 27.3%.

♦ The incidence rates for thyroid carcinoma were highest among the 15-19 year olds

and much higher among females (24.4 per million) than males (4.7 per million)

(Table XI.2).

♦ The incidence rates for malignant melanoma were highest among the 15-19 year

olds and higher among females (16.5 per million) than males (10.0 per million)

(Table XI.2).

Survival

♦ The 5-year survival rate was 99% for thyroid carcinomas. Males had a slightly

lower survival rate than females (Table XI.4).

♦ The 5-year survival rate was 91% for malignant melanoma. Females had a 93%

survival rate compared to males with an 87% survival rate (Table XI.4).

Risk factors

♦ The most well established risk factor for thyroid carcinoma is ionizing radiation

exposure, from both environmental and therapeutic sources.

♦ The primary risk factors for melanoma are sun exposure and number of melanocytic

and dysplastic nevi.

INTRODUCTION

Carcinomas are malignancies that

originate in epithelial tissues. Epithelial

cells cover the external surface of the body,

line the internal cavities, and form the

lining of glandular tissues [1,2]. Cancers

that originate from epithelial cells, includ-

ing those of the breast, lung, prostate, and

colon, are by far the most common types of

cancer in adults. Among children, however,

particularly before the adolescent years,

carcinomas are very rare. Leukemias,

central nervous system cancer, lymphomas,

sarcomas, and the embryonal cancers such

as neuroblastoma, retinoblastoma and

Wilms’ tumor, represent a far greater

burden to young children than does any

specific epithelial cancer. Nevertheless, a

variety of carcinomas do occur in children,

especially during late adolescence, and in

this report we provide descriptive epidemio-

logic data to characterize their occurrence.

In the US, approximately 1,050 children

and adolescents younger than 20 years of

age are diagnosed with carcinomas each

year, of which approximately 350 are

thyroid carcinomas and 300-350 are mela-

nomas.

ICCC XI CARCINOMAS AND OTHER MALIGNANT EPITHELIAL NEOPLASMS

140

National Cancer Institute

SEER Pediatric Monograph

Classification system

Although other classification systems

exist [2], the diverse types of epithelial

malignancies that are called ‘carcinomas

and other malignant epithelial neoplasms’

by the ICCC system are classified into six

broad subgroups [3]:

a. adrenocortical carcinoma

b. thyroid carcinoma

c. nasopharyngeal carcinoma

d. malignant melanoma

e. skin carcinoma other than mela-

noma

f. other and unspecified carcinomas

All of the malignancies within the group

‘other and unspecified carcinomas’ are very

rare in children. Neoplasms of the salivary

gland, colon, appendix, lung and bronchus,

uterine cervix, and urinary bladder account

for most of this group. Likewise, the

adrenocortical, nasopharyngeal, and skin

carcinomas (other than melanoma) rarely

occur in children. Because thyroid carcino-

mas and malignant melanoma (henceforth

called melanoma) are the only epithelial

malignancies that occur with any signifi-

cant frequency in children, we will focus

primarily on these two cancers in this

report. It should be noted that the ICCC

system includes germ cell carcinomas

within ‘germ cell, trophoblastic, and other

gonadal neoplasms’, rather than ‘carcino-

mas and other malignant epithelial neo-

plasms’. The incidence rate of gonadal

carcinomas for children younger than 20

years of age is only 1.0 per million, so this

omission will not appreciably influence our

results. Please note also that we have

calculated frequencies and rates from

SEER data for the years 1975 through

1995, and, unless otherwise specified, we

report incidence rates as average annual

rates per million children younger than 20

years of age, adjusted to the 1970 US

standard population.

Thyroid carcinomas are endocrine

tumors, although they do not necessarily

exhibit hormonal activity. There are four

types of thyroid carcinomas: papillary,

follicular, medullary and anaplastic. In

children, papillary tumors represent

greater than 70% of thyroid carcinomas,

and follicular tumors another 20%. Only a

small proportion of thyroid carcinomas are

medullary (5-10%), or anaplastic (extremely

rare). The vast majority of childhood

thyroid carcinomas are well differentiated

tumors, and despite their malignant pathol-

ogy, the clinical course of most thyroid

carcinomas is relatively benign [4-6].

Malignant cutaneous melanomas arise

from epidermal melanocytes and are often

classified into one of three histopathological

types according to the presence and pattern

of intraepidermal growth: superficial

spreading melanoma, nodular melanoma,

and lentigo maligna melanoma. Melano-

mas occur most frequently on the trunk in

white males, the lower limbs in white

females, and the soles of the feet in blacks

[7,8].

INCIDENCE

All of the carcinomas combined com-

prised 9.2% of cancer in children during the

time period of this study. The majority of

the 2,735 epithelial cancers were either

thyroid carcinomas (35.5%) or melanomas

(30.9%). Adrenocortical carcinomas (1.3%),

nasopharyngeal carcinomas (4.5%), and

other skin carcinomas (0.5%) combined for

only a small proportion of the total, while

other and unspecified carcinomas com-

prised 27.3%. Nearly 75% (2,047) of the

childhood carcinomas occurred in adoles-

cents 15-19 years of age, including 75% of

the thyroid carcinomas, 80% of the melano-

mas, 63% of the nasopharyngeal carcino-

mas, and 74% of the other and unspecified

carcinomas. Although there were only 36

adrenocortical carcinomas reported to

ICCC XICARCINOMAS AND OTHER MALIGNANT EPITHELIAL NEOPLASMS

141

National Cancer Institute SEER Pediatric Monograph

SEER areas, 18 of them occurred in chil-

dren younger than 5 years of age.

Age-specific incidence

The impressive age differences in inci-

dence rates of carcinomas are shown in

Table XI.1. Incidence rates of thyroid

carcinoma and melanoma were practically

negligible in very young children. Among

15-19 year olds, however, both melanoma

and thyroid cancer substantially increased

in occurrence over younger ages, particu-

larly for females. The male-to-female ratios

Table XI.1: Average annual age-specific incidence rates per million by histology and

sex, all races, SEER, 1975-95

Tumor type Age Total Females Males F/M Ratio

All carcinomas 0-4 1.6 1.9 1.3 1.5

5-9 2.7 3.1 2.3 1.3

10-14 10.0 12.1 8.0 1.5

15-19 40.3 55.5 25.9 2.1

Adrenocortical carcinoma 0-4 0.4 0.5 0.3 1.7

5-9 0.1 0.1 0

10-14 0.1 0.1 0.1 1.0

15-19 0.2 0.2 0.1 2.0

Thyroid carcinoma 0-4 0.1 0.2 0

5-9 1.0 1.3 0.8 1.6

10-14 3.9 6.0 1.8 3.3

15-19 14.4 24.4 4.7 5.2

Nasopharyngeal carcinoma 0-4 0.1 0 0.1

5-9 0.1 0 0.1

10-14 0.8 0.3 1.30 0.2

15-19 1.5 1.2 1.8 0.7

Malignant melanoma 0-4 0.7 0.7 0.7 1.0

5-9 0.7 0.8 0.7 1.1

10-14 2.2 2.5 1.9 1.3

15-19 13.2 16.5 10.0 1.7

Skin carcinoma other than melanoma 0-4 0 0 0

5-9 0 0 0

10-14 0.1 0.1 0.1 1.0

15-19 0.2 0.2 0.1 2.0

Other and unspecified carcinomas 0-4 0.4 0.5 0.3 1.7

5-9 0.8 0.9 0.6 1.5

10-14 2.8 3.0 2.7 1.1

15-19 11.0 12.8 9.2 1.4

ICCC XI CARCINOMAS AND OTHER MALIGNANT EPITHELIAL NEOPLASMS

142

National Cancer Institute

SEER Pediatric Monograph

in rates were greatest among adolescents

15-19 years, and the sex difference was

most pronounced for thyroid carcinoma.

Figure XI.1 provides greater detail on the

age-specific patterns of incidence for all

carcinomas combined

. Carcinoma inci-

dence was quite low through age 10 years,

but the rates increased dramatically in

older children. At age 19, incidence rates

were 35 per million males and 75 per

million females. For thyroid carcinoma

(Figure XI.2), the age-specific incidence

rates for males and females began to

diverge at age 10 years. Beginning at age

13, the rates increased substantially for

females, while the increase in male rates

was more modest. Incidence rates of mela-

noma are shown in Figure XI.3. Rates

differed minimally by sex until age 16,

when the rates for females became greater

than those for males.

Sex-specific and race-specific incidence

As shown in Table XI.2, the incidence

rate of all childhood carcinomas combined

was 13.8 per million. The ratio of white to

black rates was 1.5 to 1. The magnitude of

this racial difference is explained by the 2.5

to 1 ratio for thyroid cancer and the dra-

Figure XI.1: Total carcinoma age-specific incidence rates

by sex, all races, SEER, 1976-84 and 1986-94

)

))

)

01234567891011121314151617181920

Age (in years) at diagnosis

Average annual rate per million

Male

Female

)

Enumeration of the population at risk by single years of age was

available only for the census years 1980 and 1990. The US Bureau

of the Census provides intercensal population estimates by 5-year

age groups, but not by single years of age. Therefore, the

population estimates for 1980 were used in rate calculations for

cases diagnosed from 1976-84 and the 1990 estimates were used for

cases diagnosed from 1986-94.

ICCC XICARCINOMAS AND OTHER MALIGNANT EPITHELIAL NEOPLASMS

143

National Cancer Institute SEER Pediatric Monograph

Figure XI.2: Thyroid carcinoma age-specific

incidence rates, by sex, all races

SEER, 1976-84 and 1986-94

)

))

)

))

)

"""""

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1617 18 19 20

Age (in years) at diagnosis

Average annual rate per million

Male

Female

)

Figure XI.3: Malignant melanoma age-specific

incidence rates, by sex, all races

SEER, 1976-84 and 1986-94

)

)))

)

01234567891011121314151617181920

Age (in years) at diagnosis

Average annual rate per million

Male

Female

)

matic 16 to 1 ratio for melanoma. The

variation in carcinoma rates between

whites and blacks was greater among

females than males.

Table XI.2 further illustrates that

incidence rates of thyroid carcinoma were

over 4-fold higher in females than males for

both white and black children. Among

whites, melanoma was more common in

females than males, but not nearly to the

extent that was observed for thyroid cancer.

In fact, among whites, male rates of mela-

noma were higher than male rates of

thyroid cancer. As is well known, occur-

rence of melanoma in blacks is extremely

unusual. Although not shown in the table,

black children had a slightly higher rate of

other and unspecified carcinomas than

white children (4.7 versus 3.6 per million,

respectively).

Table XI.2: Average annual age-adjusted* incidence rates per million

by histology, race, and sex, age <20, SEER, 1975-95

Tumor type Race Total Females Males F/M Ratio

All carcinomas All races 13.8 18.2 9.5 1.9

Whites 14.2 19.3 9.4 2.1

Blacks 9.3 10.7 7.8 1.4

Thyroid carcinoma All races 4.9 8.1 1.9 4.3

Whites 5.2 8.6 2.0 4.3

Blacks 2.1 3.4 0.7 4.9

Malignant melanoma All races 4.2 5.1 3.3 1.5

Whites 4.8 5.8 3.8 1.5

Blacks 0.3 0.2 0.3 0.7

* Adjusted to the 1970 US standard population

ICCC XI CARCINOMAS AND OTHER MALIGNANT EPITHELIAL NEOPLASMS

144

National Cancer Institute

SEER Pediatric Monograph

TRENDS

Figure XI.4 illustrates yearly rates of

carcinomas from 1975-95. Age-adjusted

incidence rates did not increase appreciably

over this time frame for either melanoma

or thyroid carcinoma. Incidence rates over

time are further illustrated in Table XI.3,

where rates by sex are shown within spe-

cific time periods.

SURVIVAL

Survival probability is excellent for

children with either thyroid carcinoma or

melanoma. Table XI.4 provides 5-year

relative survival probabilities for both

major types of childhood carcinoma, as well

as for all carcinomas combined. Virtually

no change in survival was observed for

thyroid carcinoma over the time periods

1975-84 versus 1985-94 (Figure XI.5).

Survival improved slightly for melanoma,

from 85% for those diagnosed during 1975-

84, to 91% during 1985-94 (Figure XI.6).

Figure XI.4: Trends in carcinoma age-adjusted*

incidence rates, by histology, age <20

all races, both sexes, SEER 1975-95

(

+++

75 77 79 81 83 85 87 89 91 93 95

Year of diagnosis

Average annual rate per million

All carcinomas

Thyroid carcinoma

Malignant melanoma

(

*Adjusted to the 1970 US standard population

Table XI.3: Trends in average annual age-adjusted* incidence rates

per million, by histology and sex, age <20, all races

SEER, 1975-95

Tumor type Year of diagnosis Total Females Males

All carcinomas 1975-79 13.6 18.7 8.7

1980-84 12.9 16.1 9.4

1985-89 14.1 18.4 10.1

1990-95 14.3 19.3 9.7

Thyroid carcinoma 1975-79 4.8 7.8 1.8

1980-84 4.7 7.2 2.4

1985-89 5.2 8.8 1.8

1990-95 4.9 8.6 1.5

Malignant melanoma 1975-79 4.1 5.1 3.1

1980-84 3.6 4.4 2.8

1985-89 4.6 5.5 3.9

1990-95 4.6 5.6 3.6

* Adjusted to the 1970 US standard population

ICCC XICARCINOMAS AND OTHER MALIGNANT EPITHELIAL NEOPLASMS

145

National Cancer Institute SEER Pediatric Monograph

For each type of carcinoma shown, females

had slightly better survival probabilities

than males. Black children appeared to

fare worse than white children for all

carcinomas combined, however there were

only 29 black males and 34 black females

during the time period, so the results

should be viewed very cautiously. Simi-

larly, there were too few black children with

either thyroid carcinoma or melanoma on

whom to base relative survival rates.

RISK FACTORS

Risk factors for thyroid carcinoma have

been reviewed in several previous reports

[9-15] and will only be briefly summarized

here. The most well established risk factor

for thyroid carcinoma is ionizing radiation

exposure, from both environmental and

therapeutic sources. Irradiation treatment

for conditions such as tinea capitis, en-

larged thymus, acne, and cancer have

clearly been shown to increase risk for

thyroid carcinoma development. Other

than cancer treatment, however, these

causes are primarily of historical concern.

Environmental exposures to ionizing radia-

tion from the atomic bombings in Japan

and from the nuclear disaster at Chernobyl

also have definitively been shown to cause

substantial increases in thyroid carcinoma.

The preponderance of thyroid cancer in

females suggests that hormonal factors

may mediate disease occurrence. Other

potential etiologic factors include benign

thyroid diseases and certain inherited

cancer susceptibility syndromes, such as

familial adenomatous polyposis, and mul-

tiple endocrine neoplasia (MEN) types I,

IIA and IIB.

The primary risk factors for melanoma

are sun exposure and number of

melanocytic and dysplastic nevi. An exten-

sive review of studies related to sun expo-

Table XI.4: Five-year relative survival rates by histology, race and sex

age <20, SEER, 1985-94

Tumor type Total Females Males

Percent Percent Percent

All carcinomas

All races 89 93 83

White 91 93 86

Black 77 85 65

Thyroid carcinoma

All races 99 99 95

White 98 99 94

Black * * *

Malignant melanoma

All races 91 93 87

White 91 93 88

Black * * *

*Less than 25 cases.

ICCC XI CARCINOMAS AND OTHER MALIGNANT EPITHELIAL NEOPLASMS

146

National Cancer Institute

SEER Pediatric Monograph

Figure XI.5: Thyroid carcinoma 5-year relative survival rates

by sex, race, age and time period, SEER (9 areas), 1975-84 and 1985-94

# - <25 cases - rate not shown

99 99 99 99

100

9999

98 98

Total Male Female White Black <5 5-9 10-14 15-19

100

Percent surviving 5 years

1975-84

1985-94

Sex

Race

Age

Figure XI.6: Malignant melanoma 5-year relative survival rates

by sex, race, age and time period, SEER (9 areas), 1975-84 and 1985-94

# - <25 cases - rate not shown

Total Male Female White Black <5 5-9 10-14 15-19

100

Percent surviving 5 years

1975-84

1985-94

Sex

Race

Age

###

ICCC XICARCINOMAS AND OTHER MALIGNANT EPITHELIAL NEOPLASMS

147

National Cancer Institute SEER Pediatric Monograph

sure, nevi, and other potential risk factors

can be found in reference 6.

SUMMARY

In stark contrast to cancer incidence in

adults, carcinomas were very rare in chil-

dren, especially those younger than 15

years of age. Rates increased quite sub-

stantially with increasing age, however,

particularly among females aged 10-19

years. The most common types of epithelial

cancer in children were thyroid carcinomas

(4.9 per million younger than 20 years of

age) and melanomas (4.2 per million

younger than 20 years of age). Females,

however, had 4 times the thyroid cancer

rate observed for males. Additionally, white

children had 2.5 times the thyroid cancer

rate and 16 times the melanoma rate

observed for black children. From 1975-95

incidence rates of both thyroid cancer and

melanoma remained fairly stable. The

strongest known risk factor for thyroid

carcinoma is ionizing radiation exposure.

Sun exposure and number of nevi are the

best described risk factors for melanoma

occurrence. Fortunately, 5-year survival

probability is excellent for both thyroid

cancer (99%) and melanoma (91%).

Reference List

1. Junqueira LC, Carneiro J, Kelly, RO. Basic

Histology, 7

Edition. Appleton & Lange,

Norwalk, CT, 1992; 66-94.

2. Berg JW. Morphologic classification of human

cancer. In: Shottenfeld D, Fraumeni JF (eds):

Cancer Epidemiology and Prevention, 2

edition. New York: Oxford University Press,

1996; 28-44.

3. Kramarova E, Stiller CA. The international

classification of childhood cancer. Int J Cancer.

1996;68:759-765.

4. Stratakis CA, Chrousos GP. Endocrine tumors.

In: Pizzo PA, Poplack DG, Editors. Principles

and Practices of Pediatric Oncology. 3rd ed.

Philadelphia, PA: Lippencott-Raven; 1997: 947-

976.

5. Danese D, Gardini A, Farsetti A, Sciacchitano S,

Andreoli M, Ponecorvi A. Thyroid carcinoma in

children and adolescents. Eur J Pediatr; 1997;

156:190-194.

6. McClollan DR, Francis GL. Thyroid cancer in

children, pregnant women, and patients with

Graves’ disease. Endocrin Metab Clin North

Am, 1996; 25:27-47.

7. Armstrong BK, English DR. Cutaneous malig-

nant melanoma. In: Schottenfeld D, Fraumeni

JF, Editors. Cancer Epidemiology and Preven-

tion. 2nd ed. New York: Oxford University

Press; 1996; 1282-1312.

8. Armstrong BK, Kricker A.. Skin Cancer.

Dermatologic Clin. 1995; 13:583-594.

9. Ron E. Thyroid Cancer. In: Schottenfeld D,

Fraumeni JF, Editors. Cancer Epidemiology and

Prevention. 2nd ed. New York: Oxford Univer-

sity Press; 1996; 1000-1021.

10. Nikiforov YE, Fagin JA. Risk factors for thyroid

cancer. Trends Endocrinol Metab, 1997: 8:20-25.

11. Fraker DL. Radiation exposure and other

factors that predispose to human thyroid

neoplasia. Surg Clin North Am, 1995; 3:365-375.

12. Sokic SI, Adanja BJ, Vlajinac HD, Jankovic RR,

Marinkovic JP, Zivaljevic VR. Risk factors for

thyroid cancer. Neoplasma, 1994; 41:371-374.

13. Akslen LA. Thyroid cancer: some aspects of

epidemiology and etiological factors, pathologi-

cal features and tumour biology (review). Int J

Onc 1994; 4:931-942.

14. Geirger JD, Thompson NW. Thyroid tumors in

children. Otolaryngologic Clinics of North

America. 1996; 29:711-719.

15. Tucker MA, Morris Jones PH, Boice JD, Robison

LL, Stone BJ, Stovall M, Jenkin RDT, Lubin JH,

Baum ES, Siegel SE, Meadows AT, Hoover RN,

Fraumeni JF. Therapeutic radiation at a young

age is linked to secondary thyroid cancer.

Cancer Res, 1991; 51:2885-2888.

148

National Cancer Institute

SEER Pediatric Monograph

XII

149

National Cancer Institute

SEER Pediatric Monograph

CANCER AMONG INFANTS

HIGHLIGHTS

Incidence

♦ The age of peak cancer incidence among children occurred during the first year of

life (Figure XII.1).

♦ Malignancies of infancy represented 10% of all cancer that was diagnosed among

children younger than 15 years of age. The average annual incidence rate of all

infant cancer combined was 233 per million infants, which was 12% higher than the

age (2 years) with the next highest incidence.

♦ The rate among females (234 per million infants) was essentially the same as that

in males (232 per million infants). This is notable because infancy was the only age

among children younger than 15 years of age in which female rates were not lower

than male rates.

♦ Neuroblastoma comprised 28% of infant cancer cases and was the most common

malignancy among these young children (65 per million infants).

♦ The leukemias as a group (41 per million infants) represented the next most com-

mon type of cancer, comprising 17% of all cases (Figure XII.2).

♦ Central nervous system malignancies comprised 13% of infant cancer, with an

average annual incidence rate of nearly 30 per million infants.

♦ The average annual incidence rates for malignant germ cell and malignant soft

tissue tumors were essentially the same at 15 per million infants. Each comprised

about 6% of infant cancer (Figure XII.2).

♦ Leukemias accounted for a substantial proportion of the racial difference, in that the

average annual rate for white infants (48.7 per million) was 66% higher than for

black infants (29.4 per million).

Survival

♦ The prognosis for infants with cancer is often worse than in children of older ages,

even when comparing the same histologic diagnosis. For instance, the 5-year rela-

tive survival for children younger than 15 years of age who were diagnosed with

acute lymphoid leukemia from 1975-94 was well over 70%, but for infants the sur-

vival rate was 33%.

♦ Over 80% of children diagnosed with neuroblastoma during infancy were alive 5

years following diagnosis. In contrast, for children diagnosed with neuroblastoma at

age 1 year or older, the 5-year relative survival was about 45%.

INTRODUCTION

Adult cancers usually form in epithelial

tissues and are believed often to be the

result of a long biological process related to

the interaction of exogenous exposures with

genetic and other endogenous characteris-

tics among susceptible people. However, in

young children, particularly infants, the

aberrant genetic processes that fail to

safeguard against the clonal proliferation of

cells with unregulated growth potential

occur very early in life and progress very

quickly. Due to the unique clinical, genetic,

and epidemiologic characteristics of cancers

in infants [1,2], it is becoming increasingly

James G. Gurney, Malcolm A. Smith, Julie A. Ross

XII

150

National Cancer Institute

SEER Pediatric Monograph

CANCER AMONG INFANTS

apparent that the study of infant cancer

may lead to further understanding of the

mechanisms of carcinogenesis. With that in

mind, this chapter will briefly summarize

and discuss data on cancer that occurs

among children who are diagnosed within

the first year of life.

INCIDENCE

The cancer cases used to calculate

incidence rates in this discussion were

limited to primary malignancies that were

registered in SEER areas of the United

States during the time periods 1976-84 and

1986-94. This time restriction was imposed

because enumeration of the population at

risk by single years of age was available

only for the census years of 1980 and 1990.

The US Bureau of the Census provides

intercensal population estimates by 5-year

age groups, but not by single years of age.

Therefore, the population estimates for

1980 were used in rate calculations for

cases diagnosed from 1976-84, and the

1990 estimates were used for cases

diagnosed from 1986-94.

The age of peak cancer incidence

among children occurred during the first

year of life, as shown in Figure XII.1.

Malignancies of infancy represented 10% of

all cancer that was diagnosed among

children younger than 15 years of age. The

incidence rate of all infant cancer combined

was 233 per million infants, which was 12%

higher than the age (2 years) with the next

highest incidence. The rate among females

(234 per million infants) was essentially the

same as that in males (232 per million

infants). This is notable because infancy

was the only age among children younger

than 15 years of age in which female rates

were not lower than male rates. Differ-

ences in rates by sex will be discussed in

more detail below.

Figure XII.1: Total childhood cancer age-specific incidence rates

by sex, all races, SEER, 1976-84 and 1986-94 combined

)

0123456789101112131415

Age (in years) at diagnosis

100

120

140

160

180

200

220

240

260

280

Average annual rate per million

Males

Females

Both sexes

)

XII

151

National Cancer Institute SEER Pediatric Monograph

CANCER AMONG INFANTS

Histology-specific incidence

Figure XII.2 illustrates the incidence

rate for the most predominant types of

infant cancer. Although neuroblastoma

represented less than 8% of cancer cases

among children younger than 15 years of

age, neuroblastoma comprised 28% of

infant cancer cases and was the most

common malignancy among these young

children (65 per million infants). The

leukemias as a group (41 per million in-

fants) represented the next most common

type of cancer, comprising 17% of all cases.

As with older children, acute lymphoid

leukemia was the most frequently occur-

ring leukemia. The average annual inci-

dence rate of acute lymphoid leukemia was

21 per million infants, while the rate for the

acute non-lymphoid leukemias was 11 per

million infants. For juvenile chronic my-

eloid leukemia, which by international

consensus is now called juvenile

myelomonocytic leukemia (JMML), the

average annual incidence rate was 3 per

million infants. The combined rate for

other and unspecified leukemias was about

5 per million infants.

Central nervous system malignancies

comprised 13% of infant cancer, with an

incidence rate of nearly 30 per million

infants. Astrocytomas and other gliomas

(combined) accounted for half of the CNS

malignancies (15 per million), followed by

primitive neuroectodermal tumors/medullo-

blastomas (PNET, 9 per million infants)

and ependymomas (5 per million). Retino-

blastoma and Wilms’ tumor followed CNS

cancer in order of occurrence among in-

fants. Retinoblastoma accounted for about

12% of infant cancer and Wilms’ tumor an

additional 9%.

The incidence rates for malignant germ

cell tumors (including intracranial) and

malignant soft tissue tumors were essen-

tially the same at 15 per million infants.

64.6

40.5

29.7

26.7

22.5

15.3

15.2

9.5

4.4

2.8

1.2

0.5

Neuroblastoma

Leukemias

CNS

Retinoblastoma

Wilms'

Germ cell

Soft tissue

Hepatic

Lymphomas

Epithelial

Other/unspecified

Bone

0 10203040506070

Average annual rate per million

Figure XII.2: Infant age-specific incidence rates by type, all races,

both sexes, SEER, 1976-84 and 1986-94 combined

XII

152

National Cancer Institute

SEER Pediatric Monograph

CANCER AMONG INFANTS

Each comprised about 6% of infant cancer.

About 29% of the germ cell malignancies

diagnosed during infancy were gonadal,

which is half the percentage that occurs

overall in children younger than 15 years of

age. Unlike soft tissue tumors in older

children, the rate of fibrosarcomas (5 per

million infants) was similar to that of

rhabdomyosarcomas (6 per million infants).

This contrasts with the rhabdomyosarcoma

and fibrosarcoma rates among children

younger than 15 years of age of 4.6 per

million and 2.3 per million, respectively.

The rate for malignancies of the liver,

almost exclusively hepatoblastoma, was 9.5

per million infants. Hepatoblastoma is

similar to neuroblastoma, retinoblastoma,

and Wilms’ tumor (nephroblastoma) in that

it is an embryonal malignancy with the age

of peak incidence occurring during very

early childhood. Lymphomas and espe-

cially bone cancers, which are quite impor-

tant cancers among adolescents, are ex-

tremely rare in infants.

Sex-specific incidence

The female to male ratios of incidence

rates for selected cancer types are illus-

trated in Figure XII.3. None of the sex

differences were very pronounced. For

leukemias and CNS cancer, however, the

sex ratios differed by histologic subtype.

Figure XII.4 provides female to male ratios

for major subgroups of leukemia and CNS

cancer. This illustration reveals that most

types of infant leukemia and CNS cancer

were more common in females than males,

but JMML and ependymoma were notable

exceptions. For both types of neoplasms,

there is around a 2-fold higher average

annual incidence in males than in females.

The JMML rates were 4.0 per million male

infants compared with 1.7 per million

female infants. The ependymoma rates

0.86

1.17

1.07

1.19

1.1

0.92

0.91

0.8

Neuroblastoma

Leukemias

CNS

Retinoblastoma

Wilms'

Germ cell

Soft tissue

Hepatic

0 0.5 1 1.5 2

Ratio of female to male incidence rates

Figure XII.3: Ratios of female to male cancer incidence

rates among infants by type, all races

SEER, 1976-84 and 1986-94 combined

Figure XII.4: Ratios of female to male cancer incidence

rates among infants by type, all races

SEER, 1976-84 and 1986-94 combined

1.17

1.4

1.26

0.43

0.8

1.07

1.25

1.27

0.56

0.63

LEUKEMIAS

Acute lymphoid

Acute non-lymphoid

JMML

Other/Unspecified

CNS

Astrocytomas

PNET

Ependymoma

Other/Unspecified

0 0.5 1 1.5 2

Ratio of female to male incidence rates

XII

153

National Cancer Institute SEER Pediatric Monograph

CANCER AMONG INFANTS

were 6.2 per million male infants compared

with 3.5 per million female infants.

Black-white differences in incidence

Figure XII.5 demonstrates the sub-

stantial discrepancy in black-white cancer

incidence rates among children, especially

young children. During infancy this racial

variation is quite pronounced, in that white

children (275 per million white infants) had

a 40% higher malignancy rate than black

children (196 per million black infants).

This difference does not appear to be a

result of earlier diagnosis for white chil-

dren, based on the similar age pattern of

incidence that is illustrated in Figure XII.5.

The ratios of incidence rates for white

relative to black infants are shown for

selected cancer types in Figure XII.6.

Leukemias accounted for a substantial

proportion of the racial difference, in that

the average annual rate for white infants

(48.7 per million) was 66% higher than for

black infants (29.4 per million). For neuro-

blastoma, white rates were 78% higher

than black rates (79.0 vs. 44.5 per million,

respectively). The relative difference was

even greater for both liver cancer and CNS

cancer. The infant incidence rate for

hepatoblastoma was 95% higher in whites

than in blacks (11.1 vs. 5.7 per million,

respectively) and for CNS malignancies the

rate was 120% higher for whites (37.7 per

million) than for blacks (17.1 per million).

Of the more common infant cancers, only

rates of malignant germ cell tumors and

retinoblastoma were higher in black in-

fants, and these differences were slight.

The small number of cases for black infants

precluded reliable subgroup comparisons

for the CNS cancers and leukemias.

Distribution by month of age at diagnosis

Although we lack a valid denominator

(the number of children at risk by month of

Figure XII.5: Total childhood cancer age-specific

incidence rates by race, both sexes

SEER, 1976-84 and 1986-94 combined

)

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Age (in years) at diagnosis

100

120

140

160

180

200

220

240

260

280

Average annual rate per million

White

Black

All races

)

Figure XII.6: Ratios of white to black incidence rates

among infants by type, both sexes

SEER, 1976-84 and 1986-94 combined

1.78

1.66

2.2

0.89

1.07

0.93

1.03

1.95

Neuroblastoma

Leukemias

CNS

Retinoblastoma

Wilms'

Germ cell

Soft tissue

Hepatic

0 0.5 1 1.5 2 2.5

Ratio of white to black incidence rates

XII

154

National Cancer Institute

SEER Pediatric Monograph

CANCER AMONG INFANTS

age) to accurately calculate month-specific

age incidence rates, the percentage of

cancer cases (all types combined) by month

of age at diagnosis is presented in Figure

XII.7. This distribution shows that 13% of

the infant malignancies were diagnosed

during the first month following birth.

Separate distributions for all leukemias

combined and for neuroblastoma are also

illustrated. Approximately half of neuro-

blastomas were diagnosed within the first 4

months of life, with 16% diagnosed during

the month following birth. In fact, many of

these tumors were likely detected in utero.

Unlike neuroblastoma, the pattern of

diagnosis for leukemias shows that the

peak month of diagnosis occurred during

the latter part of infancy, in the 7

month

of life. Although not shown in the figure,

the majority of infant germ cell tumors

(56%) were diagnosed very soon after birth,

before 2 months of age. There was no other

cancer type that presented with such a

large percentage of cases so early after

birth.

TRENDS

As discussed in a previous report of

trends in infant cancer [2], considerable

caution must be exercised when interpret-

ing temporal changes in rates for a single

age group. Any trend analysis of incidence

rates may be confounded by changes in

population characteristics, the accuracy of

census estimates, screening practices,

diagnostic technology, morphology classifi-

cations, and case ascertainment. One or

more of these factors could effectively

conspire to show increasing incidence over

time that is not reflective of more cancer,

but rather of earlier diagnosis or better

case identification. Neuroblastoma is an

excellent example of this because of the

recent introduction of (controversial)

screening practices for earlier detection in

Japan and Canada [3,4] and because of the

Figure XII.7: Percent distribution of infant cancer cases

by month of age at diagnosis and type, all races

both sexes, SEER, 1976-84 and 1986-94 combined

0123456789101112

Age (in months) at diagnosis

Relative percent

All sites

Neuroblastoma

Leukemias

XII

155

National Cancer Institute SEER Pediatric Monograph

CANCER AMONG INFANTS

fairly recent advent of fetal ultrasound

diagnosis [5]. With that caveat in place, a

table of incidence rates for two time periods

(1976-84 and 1986-94) is presented. (Table

XII.1) The percentage change in the inci-

dence rates is also shown.

These data imply that incidence rates

of infant cancer were higher during the

time period 1986-94 than the time period

1976-84. To what degree these data repre-

sent a true increase in cancer incidence in

the US, compared with the influence of the

potentially misleading factors that were

mentioned above, has not been determined

and requires more detailed assessment.

For instance, the 125% increase in germ cell

tumors is largely concentrated among non-

gonadal malignancies (primarily sacrococ-

cygeal and pelvic) among female infants

(see the monograph chapter entitled:

“Germ Cell, Trophoblastic and Other Go-

nadal Tumors” for additional details).

While this could indicate a true increase in

the incidence of this disease, it is likely that

a substantial proportion of the increase

reflects changes over time in the identifica-

tion by pathologists of malignant elements

in teratomas with an otherwise benign

appearance, leading to increased reporting

of malignant teratomas among infants [6].

Likewise, the increase in reported rates of

infant neuroblastoma, a disease with an

exceptionally high survival rate and with a

propensity for spontaneous regression, may

reflect increased diagnoses and reporting of

tumors that previously regressed before

being detected [7].

SURVIVAL

The prognosis for infants with cancer is

often worse than in children of older ages,

even when comparing the same histologic

diagnosis. For instance, the 5-year relative

survival for children younger than 15 years

of age who were diagnosed with acute

lymphoid leukemia from 1975-94 was well

over 70%, but for infants the survival rate

was 35%. Survival was bleak at all ages for

acute non-lymphoid leukemia, but it was

the poorest for infants, with a relative 5-

year survival of 30%. This pattern was also

evident for rhabdomyosarcomas and CNS

tumors, particularly ependymomas and

PNET. For ependymomas, 5-year relative

survival for infants was less than 20% and

for PNET it was less than 30%. Infant

neuroblastoma was an exception. Over

80% of children diagnosed with neuroblas-

toma during infancy were alive 5 years

following diagnosis. In contrast, for chil-

dren diagnosed with neuroblastoma at age

1 year or older, the 5-year relative survival

was about 45%. The 5-year survival from

1975-94 was also very good for children

diagnosed with Wilms’ tumor (86%) and

retinoblastoma (over 90%).

SUMMARY

The age of peak incidence of cancer in

children occurs during the first year of life.

Neuroblastoma is the most common infant

malignancy, followed by the leukemias and

the CNS cancers. Female infants and male

infants have essentially the same overall

cancer incidence rates. White infants have

substantially higher cancer rates than

black infants for most cancer types. Inci-

Table XII.1: Average annual incidence rates

per million by type, age <1, all races

both sexes, SEER, 1976-84 and 1986-94

Cancer type 1976-84 1986-94 % Change

All cancer 197.9 269.3 36

Neuroblastoma 55.2 74.4 35

All leukemia 35.9 45.4 26

All CNS 23.3 36.5 57

Retinoblastoma 22.1 31.5 43

Wilms’ 21.4 23.6 10

Germ cell 9.5 21.3 124

Soft tissue 13.6 16.6 22

Hepatic 7.6 11.4 50

Lymphoma 4.5 4.2 - 7

Epithelial 2.6 3.0 15

Other/Unspecified 1.4 1.0 -29

Bone 0.5 0.5 0

XII

156

National Cancer Institute

SEER Pediatric Monograph

CANCER AMONG INFANTS

dence rates were notably higher for the

period 1986-94 compared with 1976-84,

although many factors other than a real

increase in incidence may be influencing

this trend. Relative survival for infants is

very good for neuroblastoma, Wilms’ tumor

and retinoblastoma, but not for most other

types of cancer.

The distribution of malignant disease

in infants is quite different from that which

is found in older children, adolescents, or

adults. For instance, embryonal tumors

such as neuroblastoma, Wilms’ tumor,

retinoblastoma, medulloblastoma and

hepatoblastoma are more prevalent in

infants than in humans of any other age.

The descriptive epidemiologic data that is

presented here may serve to stimulate

ideas for further etiologic research into the

multifactorial nature of cancer occurrence.

For example, the initial two-hit theory for

carcinogenesis was developed primarily

from clinical observations of a higher

frequency of bilaterality of retinoblastoma

in infants than in older children [8,9]. The

study of infant cancer can aid in developing

new hypotheses related to how aberrant

genetic processes, early developmental

abnormalities and gene-environment

interactions contribute to the carcinogenic

process. The study of retinoblastoma, and

later of Wilms’ tumor, led to the discovery of

two important tumor suppressor genes that

are related to adult as well as pediatric

malignancy [10]. Recent work has shown

that hematologic malignancies manifest

differently in infants than in older children

[11]. All these factors speak to the impor-

tance of further research into the epidemi-

ology and biology of cancer in very young

children.

Reference List

1. Kenney LB, Reaman GH; Special consider-

ations for the infant with cancer. Pizzo P,

Poplack D, Editors. Principles and Practices of

Pediatric Oncology. 3rd ed. Philadelphia, PA:

Lippincott-Raven; 1997:343-356.

2. Gurney JG, Ross JA, Wall DA, Bleyer WA,

Severson RK, Robison LL. Infant cancer in the

US: histology-specific incidence and trends,

1973 to 1992. J Pediat Hematol Oncol.

1997;19:428-432.

3. Bessho F. Effects of mass screening on age-

specific incidence of neuroblastoma. Int J

Cancer. 1996;67:520-522.

4. Woods WG, Tuchman M, Robison LL, et al.

Screening for neuroblastoma is ineffective in

reducing the incidence of infavourable ad-

vanced stage disease in older children. Eur J

Cancer. 1997;33:2106-2112.

5. Holgersen LO, Subramanian S, Kirpekar M,

Mootabar H, Marcus JR. Spontaneous resolu-

tion of antenatally diagnosed adrenal masses.

J Pediatr Surgery. 1996;31:153-155.

6. Hawkins, e, Issacs, H. Et al. (1993). “Occult

malignancy in neonatal sacrococcygeal terato-

mas. A report from a Combined Pediatric

Oncology Group and Children’s Cancer Group

study”. Am J Pediatr Hematol Oncol 15(4);406-

7. Brodeur GM, Castleberry RP. Neuroblastoma.

Pizzo P, Poplack D, Editors. Principles and

Practices of Pediatric Oncology. 3rd ed. Phila-

delphia, PA: Lippincott-Raven; 1997:761-.797.

8. Knudson AG. Mutation and cancer: statistical

study of retinoblastoma. Proc Natl Acad Sci

USA. 1971;68:820.

9. Pollock BH, Knudson AG; Preventing cancer in

adulthood: advise for the pediatrician. Pizzo P,

Poplack D, Editors. Principles and Practices of

Pediatric Oncology. 3rd ed. Philadelphia, PA:

Lippincott-Raven; 1997:1421-1435.

10. Malkin D, Portwine C. The genetics of child-

hood cancer. Eur J Cancer. 1994;30A:1942-

1946.

11. Ross JA, Davies SM, Potter JD, Robison LL.

Epidemiology of childhood leukemia, with a

focus on infants. Epidemiol Rev. 1994;16:243-

272.

XIII

157

National Cancer Institute

SEER Pediatric Monograph

CANCER AMONG ADOLESCENTS 15-19 YEARS OLD

HIGHLIGHTS

Incidence

♦ The incidence of cancer among adolescents (i.e., 15-19 year-olds) in SEER areas for

1986-95 was 202.2 per million, which was similar to the incidence of cancer among

0-4 year-olds and substantially greater than the incidence for 5-9 and 10-14 year-

olds (Table XIII.1).

♦ The spectrum of cancers that occurred among 15-19 year-olds was distinctive from

those that occurred in young children. For SEER areas from 1986-95, the most

common tumors among adolescents were Hodgkin’s disease (16.1%), germ cell

tumors (15.2%), CNS tumors (10.0%), non-Hodgkin’s lymphoma (NHL) (7.6%),

thyroid cancer (7.2%), malignant melanoma (7.0%), and acute lymphoblastic leuke-

mia (ALL) (6.4%) (Figure XIII.1 and Table XIII.1).

♦ The embryonal cancers that predominated among young children (e.g., neuroblas-

toma, Wilms tumor, retinoblastoma, and hepatoblastoma) (Figure XIII.2) were

distinctly uncommon among 15-19 year olds (Figure XIII.1 and Table XIII.1).

Trends in Incidence

♦ The annual incidence of cancer for adolescents increased from 183.0 per million in

1975-79 and to 203.8 per million in 1990-95 (Table XIII.4 and Figure XIII.3).

♦ The largest contributor to this increase was the germ cell, trophoblastic, and other

gonadal tumor category (specifically testicular and ovarian germ cell tumors).

♦ Smaller increases in incidence were observed for non-Hodgkin’s lymphoma (NHL),

osteosarcoma, and acute lymphoblastic leukemia (ALL).

♦ No significant increases in incidence were observed for CNS tumors, melanoma,

thyroid cancer, Hodgkin’s disease, or soft tissue sarcomas.

Incidence by Gender and Race

♦ Rates of specific cancer types differed substantially by gender and by race among

adolescents.

♦ For gender, these differences were most remarkable for thyroid cancer (much more

common in females) and for the bone tumors, ALL, and NHL (the latter three more

common among males) (Table XIII.2).

♦ Black 15-19 year olds had much lower incidence rates for Ewing’s sarcoma, testicu-

lar germ cell tumors, and melanoma than did whites. Black adolescents also had

modestly lower incidence of ALL and thyroid cancer compared to white 15-19 year

olds (Table XIII.3).

Survival

♦ Overall 5-year survival rates for adolescents with cancer improved from 69% to 77%

from 1975-84 to 1985-94 (Table XIII.5).

♦ For some cancer types (Hodgkin’s disease, germ cell tumors, thyroid cancer, and

melanoma), 5-year survival rates were 90% or better for the most recent time period

(1985-94).

♦ For other cancer types (e.g., osteosarcoma, Ewing’s sarcoma, ALL, and AML)

survival rates for adolescents remained less than 60%.

Malcolm A. Smith, James G. Gurney, Lynn A. Gloeckler Ries

XIII

158

National Cancer Institute

SEER Pediatric Monograph

CANCER AMONG ADOLESCENTS 15-19 YEARS OLD

Table XIII.1: Age-specific cancer incidence rates per million and percentage of total cases by ICCC

category and age group, all races, both sexes, SEER, 1986-95

Age (in years) at diagnosis

Tumor category

Rate

5-9

Rate

10-14

Rate

15-19

Rate

% of Total for

15-19 Group

All Sites 199.9 110.2 117.3 202.2 100.0%

Acute lymphoblastic leukemia (ALL) 58.2 30.3 17.8 12.9 6.4%

Acute myeloid leukemia (AML) (Ib) 10.1 4.5 5.7 8.5 4.2%

Hodgkin's disease (IIa) 0.8 3.9 11.7 32.5 16.1%

Non-Hodgkins lymphoma (NHL) (IIb,c,e) 5.9 8.9 10.3 15.3 7.6%

CNS tumors (III(total)) 36.0 31.9 24.6 20.2 10.0%

Ependymoma (IIIa) 5.6 1.6 1.3 1.1 0.5%

Astrocytoma (IIIb) 15.0 15.9 15.1 12.3 6.1%

Medulloblastoma/PNET (IIIc) 9.6 7.3 4.0 2.5 1.2%

Neuroblastoma & ganglioneuroblastoma (IVa) 27.4 2.6 0.8 0.5 0.2%

Retinoblastoma (V(total)) 12.5 0.5 0.0 0.1 0.0%

Wilms', rhabdoid, clear cell sarcoma (VIa) 18.0 5.8 0.6 0.4 0.2%

Hepatic tumors (VII(total)) 4.8 0.4 0.4 1.0 0.5%

Hepatoblastoma (VIIa) 4.6 0.2 0.1 0.0 0.0%

Osteosarcoma (VIIIa) 0.3 2.8 8.3 9.4 4.6%

Ewing's sarcoma (VIIIc) 0.3 1.9 4.1 4.6 2.3%

Soft tissue sarcoma (IX(total)) 10.9 8.3 10.9 15.9 7.9%

Rhabdomyosarcoma and embryonal sarcoma (IXa) 6.5 4.4 3.5 3.9 1.9%

Non-rhabdo soft tissue sarcoma (IXb-e) 4.4 4.0 7.4 11.9 5.9%

Germ Cell, trophoblastic, & other gonadal tumors (X (total)) 6.9 2.4 6.7 30.8 15.2%

Thyroid carcinoma (XIb) 0.1 1.0 4.1 14.6 7.2%

Malignant melanoma (XId) 0.8 0.6 2.8 14.1 7.0%

Other and unspecified carcinomas (XIf) 0.4 0.8 2.8 10.5 5.2%

INTRODUCTION

The adolescent population (here de-

fined as age 15-19 years) have variably

been included in analyses and reports of

childhood cancer. An NIH Policy concerning

inclusion of children in clinical research

defines children as being younger than 21

years of age, while the Food and Drug

Administration considers children to be 15

years and younger. Regardless of the

definition of children that is applied for

regulatory or reporting purposes, it is

instructive to consider the 15-19 year old

population separately because the types of

tumors that occur in this population differ

substantially from those that predominate

in younger children and in adults. Addi-

tionally, the 15-19 year old age group is one

whose participation rate in cancer clinical

trials has been noted to be much lower

than that for younger children [1].

In this chapter, differences in cancer

types and their incidence between the 15-

19 year group and younger children will be

highlighted. Additional points for emphasis

are the changes in cancer incidence for this

older age group from 1975 to 1995 and the

distinctive sex distribution for individual

tumor types. The chapter concludes with a

summary of survival rates for the 15-19

year old population, illustrating that

survival for many tumor types has im-

proved from 1975-84 to 1985-94.

INCIDENCE

Distribution of tumor types by 5-year age

groups

The incidence of specific cancers by

International Classification of Childhood

Cancer (ICCC) codes for the period 1986-

1995 is shown in Table XIII.1 by 5-year age

groups. Figure XIII.1 shows that the most

common tumors in the adolescent popula-

tion were Hodgkin’s disease (16.1%), germ

cell tumors (15.2%), CNS tumors (10.0%),

XIII

159

National Cancer Institute SEER Pediatric Monograph

CANCER AMONG ADOLESCENTS 15-19 YEARS OLD

Figure XIII.1: Distribution of cancer types, age 15-19

all races, both sexes, SEER, 1986-95

Hodgkin's

16.1%

NHL

7.6%

Germ cell

15.2%

CNS

10.0%

Thyroid

7.2%

Melanoma

7.0%

ALL

6.4%

AML

4.2%

Non-RMS sarcoma

5.9%

Osteosarcoma

4.6%

Ewing's

2.3%

Rhabdomyosarcoma (RMS)

1.9%

Other

11.6%

non-Hodgkin’s lymphoma (NHL) (7.6%),

thyroid cancer (7.2%), malignant melanoma

(7.0%), and acute lymphoblastic leukemia

(ALL) (6.4%). Table XIII.1, as well as

comparison of Figures XIII.1 and XIII.2,

illustrates that a group of tumors that

occurred commonly among children

younger than 5 years of age were virtually

absent among 15-19 year olds, including:

neuroblastoma, Wilms’ tumor, retinoblas-

toma, ependymoma, and hepatoblastoma.

These 5 tumor types accounted for approxi-

mately 35% of cases among children

younger than 5 years of age (Figure XIII.2),

Figure XIII.2: Distribution of cancer types, age <5

all races, both sexes, SEER, 1986-95

ALL

29.1%

AML

2.7%

CNS

18.0%

Neuroblastoma

13.7%

Wilms'

9.0%

Retinoblastoma

6.3%

Other

7.0%

Germ cell

3.4%

Rhabdomyosarcoma (RMS)

3.3%

NHL

3.0%

Hepatoblastoma

2.3%

non-RMS sarcoma

2.2%

XIII

160

National Cancer Institute

SEER Pediatric Monograph

CANCER AMONG ADOLESCENTS 15-19 YEARS OLD

but less than 1% of cases among 15-19 year

olds.

The distribution of tumor types within

several ICCC categories for 15-19 year olds

compared with children younger than 15

years of age warrants specific comment.

The distribution of soft tissue sarcoma

diagnoses differed by age. Rhabdomyosar-

coma accounted for 60% of soft tissue

sarcoma cases among children younger

than 5 years of age. However, the incidence

of rhabdomyosarcoma decreased with age,

while the incidence of non-rhabdomyosar-

coma soft tissue sarcomas increased with

age, so that among 15-19 year olds, rhab-

domyosarcoma accounted for only 25% of

soft tissue sarcoma diagnoses (Table

XIII.1). The annual incidence of ALL also

decreased with age: among children

younger than 5 years of age the rate was

58.2 per million, while for 15-19 year olds

the rate was nearly 5-fold less (12.9 per

million) (Table XIII.1). Although ALL

accounted for nearly 30% of cancer cases

among children younger than 5 years of

age, it represented only 6.4% of cases

among the 15-19 year olds. The incidence

of NHL was higher among 15-19 year olds

than among younger age groups (Table

XIII.1). This increase was largely the

result of much higher rates for diffuse large

cell lymphoma among 15-19 year olds,

while rates for Burkitt’s lymphoma and

lymphoblastic lymphoma for 15-19 year

olds were similar to or less than those

observed for children less than 15 years of

age (see Lymphoma chapter for details).

Sex-specific incidence

The overall incidence of cancer cases

was similar among males and females in

the 15-19 year old age group for the years

1986 to 1995 (Table XIII.2). However, the

overall similarity masked marked differ-

ences in rates for individual tumor types.

There was a strong male predominance for

ALL, NHL, osteosarcoma, and Ewing’s

sarcoma, with 60% to over 100% higher

rates occurring in males than females.

Additionally, rates of CNS tumors and germ

cell tumors were 30-40% higher in males

than in females. On the other hand, there

was a female predominance for Hodgkin’s

disease, thyroid carcinoma, and melanoma.

The rates by sex for the 15-19 year olds

were distinct from those for children

younger than 15 years of age. For ALL,

there was only a 20% excess of male cases

among the children younger than 15 years

of age, compared to a nearly 120% excess

for the 15-19 year olds. There was a male

predominance for Hodgkin’s disease in the

0-14 year olds, compared to a female

predominance in 15-19 year olds. The

situation was reversed for the germ cell,

trophoblastic, and other gonadal tumors,

which were more common among females

Table XIII.2: Age-adjusted cancer incidence rates per million by ICCC group

sex, and age, all races, both sexes, SEER, 1986-95

Age (in years) at diagnosis

Tumor category

0-14

Male*

0-14

Female*

0-14

Ratio: M/F

15-19

Male

15-19

Female

15-19

Ratio: M/F

ALL SITES 149.5 128.7 1.2 204.3 199.9 1.0

Acute lymphoblastic leukemia 37.1 30.9 1.2 17.5 8.0 2.2

Acute myeloid leukemia (Ib) 6.6 6.5 1.0 8.4 8.5 1.0

Hodgkin's disease (IIa) 6.5 5.0 1.3 28.8 36.5 0.8

Non-Hodgkin's lymphoma (IIb,c,e) 12.3 4.5 2.7 19.4 11.0 1.8

CNS (III) 33.0 27.9 1.2 23.0 17.3 1.3

Osteosarcoma (VIIIa) 3.8 4.3 0.9 11.5 7.1 1.6

Ewing's sarcoma (VIIIc) 2.3 2.2 1.1 5.8 3.3 1.8

Soft tissue sarcomas (IX) 10.9 9.1 1.2 17.4 14.3 1.2

Germ cell tumors (X) 4.3 6.2 0.7 35.2 26.1 1.4

Thyroid carcinoma (XIb) 0.9 2.9 0.3 3.7 26.2 0.1

Melanoma (XId) 1.3 1.6 0.8 10.5 17.9 0.6

*Adjusted to the 1970 US standard population

XIII

161

National Cancer Institute SEER Pediatric Monograph

CANCER AMONG ADOLESCENTS 15-19 YEARS OLD

in children younger than 15 years of age,

but more common among males in the 15-

19 year old age group. The male predomi-

nance for bone tumors observed in adoles-

cents was absent in the younger than 15

year old age group. For NHL, there was a

marked male predominance for both 0-14

year olds and for 15-19 year olds.

Black-white differences in incidence

The incidence of cancer among whites

age 15-19 years for 1986-95 was

approximately 1.5-fold higher than that

among blacks age 15-19 years (Table

XIII.3). In comparing cancer incidence for

white and black 15-19 year olds, incidence

rates at least 2-fold higher were observed

among whites, compared to blacks for ALL,

germ cell tumors, thyroid cancer, Ewing’s

sarcoma, and melanoma. The low incidence

for germ cell tumors among blacks was

restricted to testicular germ cell tumors in

males. White females and black females

had similar rates for germ cell tumors (see

Germ Cell, Trophoblastic, and Other

Gonadal Tumor chapter for additional

details). While the very high ratio of white

to black cases for melanoma may be

explained by the protection afforded from

ultraviolet light by melanin, the reasons

that blacks have lower rates of Ewing’s

sarcoma, ALL, testicular germ cell tumors,

and thyroid cancer are not apparent.

TRENDS

The average annual age-adjusted

cancer incidence among 15-19 year olds

increased from 183 per million in 1975-79

to slightly over 203.8 per million in 1990-95

(Figure XIII.3 and Table XIII.4). By

comparison, the incidence of cancer for

children younger than 15 years of age

increased from 124.3 per million in 1975-79

to 139.9 per million in 1990-95. The

greatest numeric increase in annual

incidence for the 15-19 year group occurred

for the germ cell,trophoblastic, and other

gonadal (GCTOG) tumors. This increase

was primarily the result of an increase in

the incidence of testicular germ cell tumors

among males (increasing from 22.1 to 28.4

per million) and ovarian germ cell tumors

among females (increasing from 7.9 to 13.3

Table XIII.3: Age-specific cancer incidence rates

per million by ICCC group and race

age 15-19, SEER, 1986-95

Tumor category White Black

W/B

Ratio

Total 213.5 144.8 1.5

Acute lymphoblastic leukemia 14.3 6.4 2.2

Acute myeloid leukemia (Ib) 8.3 7.1 1.2

Hodgkin's disease (IIa) 36.5 26.9 1.4

Non-Hodgkin's lymphoma (IIb,c,e) 16.1 9.4 1.7

CNS (III) 21.8 15.8 1.4

Osteosarcoma (VIIIa) 9.2 8.4 1.1

Ewing's sarcoma (VIIIc) 5.4 0.3 18.0

Soft tissue sarcomas (IX) 14.5 20.5 0.7

Germ cell tumors (X) 33.9 13.8 2.5

Thyroid carcinoma (XIb) 15.5 6.7 2.3

Melanoma (XId) 16.1 0.3 53.7

Figure XIII.3: Trends in age-adjusted* cancer incidence

rates by age group, all races, both sexes, SEER, 1975-95

(

75 80 85 90 95

Year of diagnosis

100

150

200

Average annual rate per million

Incidence 15-19

Incidence <15

(

*Adjusted to the 1970 US standard population

XIII

162

National Cancer Institute

SEER Pediatric Monograph

CANCER AMONG ADOLESCENTS 15-19 YEARS OLD

per million). As discussed in the chapter for

GCTOG tumors, the increase in gonadal

carcinomas is likely artifactual and

attributable to changes in reporting of

ovarian tumors during this time period,

specifically inclusion of borderline tumors of

the ovary. The incidence of ALL, NHL, and

osteosarcoma also increased from 1975-79

to 1990-95 (Table XIII.4). These four tumor

types accounted for the majority of the

increase in cancer incidence for the 15-19

year old group. No significant increases or

decreases in incidence were observed for

CNS tumors, melanoma, thyroid cancer,

Hodgkin’s disease, or soft tissue sarcomas.

SURVIVAL

Table XIII.5 shows 5-year relative

survival rates for different cancer types for

15-19 year olds , with comparison made

between an earlier time period (1975-84)

Table XIII.4: Average annual age-specific incidence rates per million adolescents

15-19 years old for selected tumors, all races, both sexes, SEER, 1975-95

Tumor type (ICCC Category) 1975-79

Rate

1980-84

Rate

1985-89

Rate

1990-95

Rate

All Sites 183.0 187.7 199.3 203.8

Acute Lymphoblastic Leukemia (Ia) 10.6 13.2 12.4 13.0

Non-Hodgkin’s lymphoma (IIb,c,e) 10.7 14.5 14.4 16.3

Osteosarcoma (VIIIa) 6.6 8.9 9.7 9.3

Germ cell, trophoblastic and other gonadal tumors (X) 23.2 24.0 28.6 32.0

Testicular germ cell tumor (Xc, male) 22.1 26.7 24.9 28.4

Ovarian germ cell tumor (Xc, female) 7.9 8.3 11.8 13.3

Gonadal carcinoma (Xd) 2.7 2.4 4.3 5.3

Rates are per 1,000,000.

Table XIII.5: 5-Year Relative Survival Rates by ICCC group

and time period, age 15-19, all races, both sexes

SEER, 1975-84 and 1985-94

TUMOR CATEGORY 1975-84 1985-94

Total 69% 77%

Acute lymphoblastic leukemia (ALL) 35% 51%

Acute myeloid leukemia (AML) (Ib) 22% 42%

Hodgkin's (IIa) 88% 90%

NHL (IIb,c,e) 56% 69%

Astrocytoma (IIIb) 62% 75%

Medulloblastoma (IIIc) 63% 75%

Osteosarcoma (VIIIa) 49% 59%

Ewing's sarcoma (VIIIc) 36% 56%

Soft tissue sarcoma (IX) 70% 63%

Rhabdomyosarcoma (IXa) 40% 45%

Germ cell tumors (X) 79% 90%

Thyroid carcinoma (XIb) 99% 99%

Melanoma (XId) 84% 92%

XIII

163

National Cancer Institute SEER Pediatric Monograph

CANCER AMONG ADOLESCENTS 15-19 YEARS OLD

and a recent reporting period (1985-94).

Important observations concerning survival

rates include:

• For all cancer diagnoses in the 15-

19 year old age group, the 5-year

survival rate for the recent

reporting period was 77%, which

was higher than that for the other

five-year age groups younger than

20 years of age.

• Five-year survival rates of 90% or

higher were observed for Hodgkin’s

disease, germ cell tumors, thyroid

carcinoma, and melanoma. For

germ cell tumors and melanoma,

the survival rates improved

between the earlier and recent

time period.

• Five-year survival rates for NHL

improved between 1975-84 and

1985-94 from approximately 56%

to 69%.

• Five-year survival rates for both

ALL and AML improved substan-

tially for the 15-19 year old group,

with survival rates for 1985-94 of

51% and 42%, respectively, com-

pared to only 35% and 22% for

1975-84.

Five-year survival rates for Ewing’s

sarcoma improved from 36% to 56%

between the earlier and the recent report-

ing period.

The mortality burden is a function of

the survival and the incidence rates. The

leukemias are the primary contributor to

the cancer mortality burden for cancers

developing in the 15-19 year olds. In

addition to leukemia, bone cancer, soft

tissue sarcoma, CNS cancer, NHL, and

Hodgkin’s disease are the most common

causes of cancer death among this group -

see the mortality chapter. Although thyroid

carcinoma and melanoma are among the

more common cancers in this age group,

they contribute little to the overall cancer

mortality burden for the 15-19 year old age

group.

SUMMARY

The spectrum of malignancies that

occur in adolescents is distinctive when

compared to those that occur in young

children and those that occur in older

adults. The embryonal cancers that

predominate among young children (e.g.,

neuroblastoma, Wilms’ tumor, retinoblas-

toma, ependymoma, and hepatoblastoma)

are very uncommon among 15-19 year olds.

Similarly, the epithelial carcinomas of

adults (e.g., lung, breast, colon) rarely occur

in 15-19 year olds. While some types of

acute leukemias and CNS cancers are

shared with both the older adult and the

young childhood populations, the 15-19

year old group experiences high rates of a

set of tumors (including germ cell tumors,

Hodgkin’s disease, and the bone cancers)

that are relatively characteristic of the

adolescent/young adult age group.

The annual incidence of cancer in

adolescents increased from 183 per million

for 1975-79 to 203.8 per million from 1990-

95. The largest contributor to this increase

was the germ cell, trophoblastic, and other

gonadal tumor category (specifically

testicular and ovarian germ cell tumors),

with smaller contributions from NHL,

osteosarcoma, and ALL.

Rates of specific cancer types differed

substantially by sex and by race. For sex,

these differences were most remarkable for

thyroid cancer (much more common in

females) and for the bone tumors, ALL, and

NHL (the latter three more common among

males). Black 15-19 year olds had much

lower incidence rates of Ewing’s sarcoma,

testicular germ cell tumors, and melanoma

than did whites, and modestly lower

XIII

164

National Cancer Institute

SEER Pediatric Monograph

CANCER AMONG ADOLESCENTS 15-19 YEARS OLD

incidence rates of ALL and thyroid cancer.

Five-year survival for 15-19 year olds

increased from 69% to 77% from 1975-84 to

1985-94, with a 90% survival rate or better

for several diagnoses (Hodgkin’s disease,

germ cell tumors, thyroid cancer, and

melanoma). However, for some cancers, the

survival rates remained less than 60%

(including osteosarcoma, Ewing’s sarcoma,

ALL, and AML).

Reference List

1. Bleyer W, Tejada H, Murphy S, et al: National

cancer clinical trials: Children have equal

access; adolescents do not. J Adolesc Health

21:366-373, 1997.

XIV

165

National Cancer Institute SEER Pediatric Monograph

CHILDHOOD CANCER MORTALITY

Introduction

Over the past two decades, childhood

cancer mortality in the United States has

declined dramatically. To present a compre-

hensive picture of childhood cancer occur-

rence and outcome, it would be ideal to

include cancer-specific data on incidence,

survival, and mortality within each indi-

vidual chapter of the monograph. The

available data on mortality, however, are

obtained from death certificates and col-

lected by the National Center for Health

Statistics (NCHS) for the entire United

States. In addition to the difference in

geographic coverage between NCHS and

SEER areas, the cancer classification used

by NCHS for mortality is less specific than

that used by SEER areas. Therefore, we

are presenting this separate chapter on

cancer mortality and have included inci-

dence [1,2] comparisons based on compa-

rable definitions to the mortality data [3].

A further explanation on differences be-

tween the incidence definitions used in the

other chapters and mortality is included at

the end of this chapter. The mortality data

are provided by the National Center for

Health Statistics to the National Cancer

Institute on public-use tapes.

All Sites

In contrast to incidence rates, cancer

mortality declined substantially between

1975 and 1995 (Figure XIV.1). There were

statistically significant declines in

mortality for each of the five-year age

groups ( <5, 5-9, 10-14, and 15-19) for

cancers combined. The declines by age

group ranged from 2.0 to 3.2 percent per

year. The overall decline in mortality was

nearly 40 percent between 1975 and 1995,

a statistically significant decrease of 2.6

percent per year. The overall incidence

increased 0.8 percent per year. There were

2,275 cancer deaths among children in

1995. Except for those 15-19, leukemia and

brain/other nervous system comprised more

than 50 percent of the deaths due to cancer

(Figure XIV.2). The relative difference for

Lynn A. Gloeckler Ries

Figure XIV.1: Trends in childhood cancer age-adjusted*

rates, all races, both sexes, age <20

SEER incidence & US mortality, 1975-95

1975 1980 1985 1990 1995

Year of diagnosis/death

100

125

150

175

Average annual rate per million

Incidence

Mortality

*Adjusted to the 1970 US standard population

Figure XIV.2: Percent distribution of childhood cancer

mortality by type and age group, age <20

all races, both sexes, United States, 1995

5-9

10-14

15-19

<20

Age (in years) at diagnosis

0 20406080100

Percent

Leukemias Brain/ONS Other

XIV

166

National Cancer Institute

SEER Pediatric Monograph

CHILDHOOD CANCER MORTALITY

the 15-19 year olds was due to deaths from

lymphoma (14%), bone (13%), and soft

tissue sarcomas (9%). Leukemias and

brain cancer, however, accounted for 57% of

cancer deaths for all children combined.

Leukemia

In 1995, thirty-four percent of the

childhood cancer deaths were due to leuke-

mia. The death rate from leukemia fell

nearly 50 percent between 1975 and 1995

(Figure XIV.3), a statistically significant

decline of 3.4 percent per year while the

incidence increased between 1975 and

1995. Mortality rates declined significantly

for each of the age groups (<5, 5-9,10-14,

15-19, <20) and for both males and females.

Brain/other central nervous system (brain/

CNS)

In 1995, nearly one-fourth of childhood

cancer deaths were due to invasive malig-

nancies of the central nervous system,

primarily the brain. Mortality from brain

and other CNS cancer declined an average

of 1.1 percent per year. This was an overall

decline of 23 percent between 1975 and

1995 (Figure XIV.4). This mortality decline

occurred while the incidence rate increased

mainly in the mid-1980s [4].

Unlike most benign tumors,

noninvasive tumors of the brain/CNS have

the potential to be fatal. Figure XIV.5

illustrates mortality rates for brain tumors

classified as invasive, unspecified or uncer-

tain, and benign. If the behavior of the

Figure XIV.3: Trends in childhood leukemia age-adjusted*

rates, age <20, all races, both sexes

SEER incidence, and US mortaility, 1975-95

1975 1980 1985 1990 1995

Year of diagnosis/death

Average annual rate per million

Incidence

Mortality

*Adjusted to the 1970 US standard population

Figure XIV.4: Trends in brain/other nervous system

cancer age-adjusted* rates, all races, both sexes

age <20, SEER incidence & US mortality, 1975-95

1975 1980 1985 1990 1995

Year of diagnosis/death

Average annual rate per million

Mortality

Incidence

*Adjusted to the 1970 US standard population

XIV

167

National Cancer Institute SEER Pediatric Monograph

CHILDHOOD CANCER MORTALITY

tumor is not clear from the death certifi-

cate, it is considered “unspecified or uncer-

tain”. Some of these tumors will be inva-

sive and some will not. Although mortality

from invasive tumors has declined some-

what over the past decade, there appears to

be no change over time in the rates of death

from brain tumors classified as either

“benign” or “unspecified or uncertain”.

Thus, the reduction in mortality from

invasive brain cancer does not appear to be

an artifact due to changes in the reporting

of the other categories of brain tumors. To

avoid changes in death classification be-

tween 1978 and 1979, this figure begins in

1979.

Ages 0-4

From 1975 to 1995, death rates from

cancer declined 2.9 percent per year among

children younger than 5 years of age. The

Figure XIV.6 shows the mortality rates for

the four leading causes of cancer death

among young children. The death rates

have declined for each. For leukemias, the

death rates declined by an average of 3.5

percent each year or more than 50 percent

between 1975 and 1995. After leukemia

and brain/CNS cancer, endocrine malignan-

cies were responsible for the most cancer

deaths. Most of the cancers classified as

“endocrine” in this age group were neuro-

blastomas. In 1995,there were 558 deaths

due to cancer among children younger than

5 years of age in the entire United States.

Ages 5-9

There were 523 deaths due to cancer

among children 5-9 years of age in the

entire United States in 1995. The age

group 5-9 years of age had the largest

decline in cancer mortality. The top four

mortality sites were leukemia, brain/CNS,

endocrine and non-Hodgkin’s lymphoma.

Figure XIV.6: Trends in age-specific cancer mortality

rates by type, age <5, all races, both sexes

United States, 1975-95

(

((

(

1975-77 1981-83 1987-89 1993-95

Year of death

Average annual rate per million

Brain & ONS

Leukemia

Endocrine

Soft tissue

(

Figure XIV.5: Trends in age-adjusted* brain tumor

mortality rates by behavior, age <20, United States, 1979-95

)

1979 1984 1989 1994

Year of death

Average annual rate per million

Invasive

Benign

Unspecified & uncertain

)

*Adjusted to the 1970 US standard population

XIV

168

National Cancer Institute

SEER Pediatric Monograph

CHILDHOOD CANCER MORTALITY

The decline in leukemia deaths was 5

percent per year (Figure XIV.7).

Ages 10-14

There were 503 deaths due to cancer

among children 10-14 years of age in the

entire United States in 1995. The death

rate declined 2.5 percent per year. The

decline for leukemias was 3 percent per

year. The top four mortality sites were

leukemia, brain/CNS, bone/joints, and non-

Hodgkin’s lymphoma (Figure XIV.8).

Ages 15-19

There were 691 deaths due to cancer

among children 15-19 in the entire United

States in 1995. The overall cancer death

rate declined 2 percent per year. The top

five cancer mortality sites are shown for

this age group since the death rates for soft

tissue and non-Hodgkin’s lymphoma were

similar for the most recent time period

(Figure XIV.9).

Recent cancer mortality (1990-1995) by

race/ethnicity

The cancer mortality rates for all races

combined and for white children declined

2.4 and 3.0 percent per year, respectively.

The mortality rates for black and for His-

panic children declined 0.5 percent per year

between 1990 and 1995. For American

Indian children and Asian Pacific Islander

children, the death rates increased slightly

at 0.5 percent per year. The cancer death

rates for American Indian children (23.8

per million) and for Asian Pacific Islander

children (29.2 per million) were less than

those for white children (32.9), black chil-

dren (32.5) or Hispanic children (33.5 per

million children). The mortality data are

for the whole United States except for

Hispanics for which four states (New

Hampshire, Oklahoma, Connecticut and

Louisiana) are excluded. Hispanics can be

of any race and are therefore, not mutually

exclusive from the other categories.

Figure XIV.7: Trends in age-specific cancer mortality

rates by type, age 5-9, all races, both sexes

United States, 1975-95

(

((

(

((

(

1975-77 1981-83 1987-89 1993-95

Year of death

Average annual rate per million

Brain & ONS

Leukemia

Endocrine

Non-Hodgkin's lymphomas

(

Figure XIV.8: Trends in age-specific cancer mortality

rates by type, age 10-14, all races

both sexes, United States, 1975-95

(

1975-77 1981-83 1987-89 1993-95

Year of death

Average annual rate per million

Brain & ONS

Leukemia

Bones & joints

Non-Hodgkin's lymphomas

(

XIV

169

National Cancer Institute SEER Pediatric Monograph

CHILDHOOD CANCER MORTALITY

Problems comparing incidence to mortality

The histology site groupings presented in

other chapters of this monograph are based

on the International Childhood Cancer

Classification (ICCC) [5]. While they are

useful groupings for incidence, there are

problems when comparing incidence to

mortality. The ICCC uses histology as its

main criteria and secondarily primary site.

The underlying cause of death, on the other

hand, is coded by the International Classifi-

cation of Diseases, which is based primarily

on site of origin rather than histology

especially for solid tumors [3]. For example,

mortality data would use kidney cancer but

the ICCC grouping would be Wilms’ tumor.

Therefore, all of the incidence rates pre-

sented in this chapter are based primarily

on site rather than histology. Note, that

this does not effect non-solid tumors such

as leukemia which would have comparable

groups in each. More incidence and mortal-

ity rates using comparable categories can

be found in the SEER Cancer Statistics

Review: 1973-1996 [6].

Summary

Cancer mortality has declined dramatically

for children. In the United States today

few children die from cancer in comparison

to other causes of death. In 1995, for

children younger than 20 years of age, the

major causes of death were:

• conditions from the perinatal

period (13,449);

• accidents (13,234);

• congenital anomalies (7,949);

• homicides (4,617);

• SIDS (3,397);

• cancer (2,275);

• suicides (2,227 deaths).

Of the nearly 60,000 deaths among

children younger than 20 years of age, less

than 4% were due to neoplasms (cancer). If

infants are excluded, the number one cause

of death was accidents followed by

homicides, suicides and then cancer.

Reference List

1. World Health Organization, International

Classification of Diseases for Oncology, First

Edition, Geneva, 1976.

2. Percy C, Van Holten V, and Muir C, Eds.

International Classification of Diseases for

Oncology, Second Ed., World Health Organiza-

tion, Geneva, 1990.

3. World Health Organization, International

Classification of Diseases, 1975 Revision, vols.1

and 2, Geneva, 1977.

4. Smith MA, et al: Trends in reported incidence

of primary malignant brain tumors in children

in the United States. J Natl Cancer Inst.

90:1269-77, 1998.

5. Kramarova E, Stiller CA: The international

classification of childhood cancer. Int J Cancer:

68:759-65, 1996.

6. Ries LAG, Kosary CL, Hankey BF, Miller BA,

Clegg L, Edwards BK (eds). SEER Cancer

Statistics Review 1973-1996, National Cancer

Institute, http://www-seer.ims.nci.nih.gov,

1998.

Figure XIV.9: Trends in age-specific cancer mortality

rates by type, age 15-19, all races

both sexes, United States, 1975-95

(

1975-77 1981-83 1987-89 1993-95

Year of death

Average annual rate per million

Brain & ONS

Leukemia

Bones & joints

Soft tissue

Non-Hodgkins lymphomas

(

170

National Cancer Institute

SEER Pediatric Monograph

171

National Cancer Institute SEER Pediatric Monograph

OTHER RESOURCES AVAILABLE AT NCI/NIH

Cancer Information Service (CIS) – 1-800-4-CANCER or

http://cis.nci.nih.gov/contact/faqform.html

♦ The Cancer Information Service is the National Cancer Institute’s link to the public,

providing current scientific information in understandable language to patients, their

families, the general public and health professionals. Through a network of 19 regional

offices located throughout the country, the CIS serves the entire United States and

Puerto Rico. CIS staff are available Monday through Friday from 9:00 a.m. to 4:30 p.m.

local time. Callers with TTY equipment may call 1–800–332–8615. Recorded informa-

tion on cancer topics is available 24 hours a day.

NCI publications on childhood cancer: Available by calling 1-800-4-CANCER (1-800-

422-6237).

Also, these publications may be viewed on the Cancer Net’s Kids Home page at

http://cancernet.nci.nih.gov/occdocs/KidsHome.html

♦ Managing Your Child’s Eating Problems: During Cancer Treatment This booklet in-

cludes information about the importance of nutrition, the side effects of cancer and its

treatment, ways to encourage your child to eat, and special diets.

♦ Talking With Your Child About Cancer: This booklet is for a parent whose child has

been diagnosed with cancer. It addresses the health-related concerns of young people of

different ages; it suggests ways to discuss disease-related issues with a child.

♦ When Someone In Your Family Has Cancer: This booklet is for young people whose

parent or sibling has cancer. It includes sections on the disease, its treatment, and

emotional concerns.

♦ Young People With Cancer: A Handbook For Parents: This booklet discusses the most

common types of childhood cancer, treatments, side effects, and issues that may arise

when a child is diagnosed with cancer. It offers medical information and practical tips

gathered from parents.

National Cancer Institute Information (http://cis.nci.nih.gov/resources/resources.html)

♦ The National Cancer Institute is the Federal Government’s principal agency for cancer

research and training. The Web site contains information about the latest news on

cancer research and treatment, upcoming events at NCI, job opportunities, and links to

cancer information resources. A publication index, containing some full-text NCI publi-

cations, is also available.

PDQ

(http://cancernet.nci.nih.gov/pdq.htm):

♦ NCI’s comprehensive cancer database, including summaries on cancer treatment,

screening, prevention, and supportive care, and information on ongoing clinical trials.

♦ PDQ

is a dynamic database that is updated regularly to ensure that the information it

contains is consistent with the results of the latest cancer research. PDQ

contains:

172

National Cancer Institute

SEER Pediatric Monograph

OTHER RESOURCES AVAILABLE AT NCI/NIH

cancer information summaries describing the latest advances in cancer treatment,

supportive care, screening, and prevention; an extensive register of over 1,500 ongoing

clinical trials, with information about studies around the world; and directories of over

23,000 physicians and over 11,000 organizations active in cancer treatment and care.

Most cancer information summaries appear in two versions: a technical version for the

health professional and a non-technical version for patients, their families, and the

public. Both are available in English and Spanish. The information in the database is

peer reviewed by editorial boards of oncology experts and updated monthly.

cancerTrials™ Web site (http://cancertrials.nci.nih.gov/):

♦ NCI’s comprehensive clinical trials information center for patients, health professionals,

and the public. Includes information on understanding trials, deciding whether to

participate in trials, finding specific trials, plus research news and other resources.

CancerNet™ (http://cancernet.nci.nih.gov/index.html):

♦ The NCI International Cancer Information Center’s World Wide Web site, CancerNet™,

provides both health professionals and the public access to a variety of information on

cancer.

MEDLINEplus (http://www.nlm.nih.gov/medlineplus/ )

♦ The National Library of Medicine’s MEDLINEplus Web site includes links to informa-

tion about a number of health topics, medical dictionaries, databases (including

MEDLINE), clearing houses, directories, organizations, publications and health news,

and consumer health libraries.

NCI Office of Liaison Activities:

♦ Throughout the Nation, hundreds of cancer advocacy and outreach organizations pro-

vide education and support to their communities. The Office of Liaison Activities is

NCI’s central point of contact to the national advocacy organizations and, through them,

to the community-based groups. This office maintains ongoing communications and

information exchange between the cancer advocacy organizations and NCI, encouraging

input and feedback from them, and cooperates and collaborates with these groups in

areas of mutual interest. The Office of Liaison Activities serves as a catalyst and re-

source to link advocates with NCI programs, working groups and advisory committees

and helps to integrate consumer advocate representatives throughout the NCI. The

Office of Liaison Activities also builds relationships with professional societies and

federal agencies, and provides input and perspective to NCI on complex issues relevant

to cancer patients and the public. Telephone number 301-594-3194 and FAX 301-480-

7558.

173

National Cancer Institute SEER Pediatric Monograph

OTHER RESOURCES AVAILABLE AT NCI/NIH

On the SEER Web page, http://www-seer.ims.nci.nih.gov/, the Cancer Statistics Branch

(CSB) provides additional information on cancer statistics. This monograph can be viewed

at this Web address under Publications. Other SEER publications/monographs (SEER

Cancer Statistics Review (CSR); SEER Prostate Cancer Trends, 1973-1995; Racial/Ethnic

Patterns of Cancer in the United States; and Cancer Rates and Risks) are also included in

this area of the Web page. The CSR is an annual compendium of the most recent cancer

statistcis available. The current version, 1973-1996, includes two chapters on childhood

cancer: one by primary cancer site and a second by the International Classification of

Childhood Cancer.

The CSB provides additional SEER data on the SEER Web page under Scientific Systems:

• CANQUES is an interactive system on the Web that allows the user to access over 10

million pre-calculated cancer statistics.

• A SEER public-use file with SEER*Stat provides an easy to use PC desktop system for

the production of a myriad of cancer statistics such as incidence rates, survival rates by

various demographic and tumor variables. This CD-ROM can be ordered from the Web

page.

The Applied Research Branch (ARB) Web site,

http://www-dccps.ims.nci.nih.gov/ARB/index.html, provides information on the following

topics: Risk Factor Monitoring & Methods, SEER-Medicare, Cancer Statistics Methods &

Models, Dietary Assessment, Breast Cancer Surveillance Consortium, Health Services &

Economics, HMO Cancer Research Network, Outcomes Research. Under Cancer Statistics

Methods & Models, there is information about the methodology and estimation for cancer

prevalence and the probability of developing cancer.

174

National Cancer Institute

SEER Pediatric Monograph

175

National Cancer Institute SEER Pediatric Monograph

INTERNATIONAL CLASSIFICATION OF CHILDHOOD CANCER (ICCC)

Source: Kramárová E, Stiller CA, Ferlay J, Parkin DM, Draper GJ, Michaelis J,

Neglia J, Qureshi S (1996) International Classification of Childhood

Cancer1996. IARC Technical Report No.29, International Agency for

Research of Cancer, Lyon.

ICCC GROUP MORPHOLOGY TOPOGRAPHY

I Leukemia

(a) Lymphoid Leukemia

Excluding ALL 9820, 9822-9827, 9850 C00.0-C80.9

ALL 9821 C00.0-C80.9

(b) Acute Leukemia

Excluding AML 9840, 9841, 9864, 9866, 9867, 9891,

9894, 9910

C00.0-C80.9

AML 9861 C00.0-C80.9

Leukemia

9863, 9868 C00.0-C80.9

(d) Other Specified

Leukemias

9830, 9842, 9860, 9862, 9870-9890,

9892, 9893, 9900, 9930-9941

C00.0-C80.9

(e) Unspecified Leukemias 9800-9804 C00.0-C80.9

II Lymphomas and

Reticuloendothelial

Neoplasms

(a) Hodgkin’s disease 9650-9667 C00.0-C80.9

(b) Non-Hodgkin’s

lymphomas

9591-9595, 9670-9686, 9690-9717,

9723, 9688

C00.0-C80.9

(d) Miscellaneous

lymphoreticular

neoplasms

9720, 9731-9764 C00.0-C80.9

(e) Unspecified lymphomas 9590 C00.0-C80.9

III CNS and

Miscellaneous

Intracranial and

Intraspinal Neoplasms

(a) Ependymoma 9383, 9390-9394 C00.0-C80.9

(b) Astrocytoma 9380 C72.3

9381, 9400-9441 C00.0-C80.9

neuroectodermal tumors

9470-9473 C00.0-C80.9*

(d) Other gliomas 9380 C70.0-C72.2, C72.4-

C72.9

9382, 9384, 9442-9460, 9481 C00.0-C80.9

(e) Miscellaneous

intracranial and

intraspinal neoplasms

8270-8281, 8300, 9350-9362, 9480,

9505, 9530-9539

C00.0-C80.9

(f) Unspecified intracranial

and intraspinal

neoplasms

8000-8004 C70.0-C72.9, C75.1-

C75.3

*For this monograph, any cases with site codes C00.0-C69.9, C73.9-C75.0, C75.4-C77.9

were removed from this group.

INTERNATIONAL CLASSIFICATION OF CHILDHOOD CANCER (ICCC)

176

National Cancer Institute

SEER Pediatric Monograph

Source: Kramárová E, Stiller CA, Ferlay J, Parkin DM, Draper GJ, Michaelis J,

Neglia J, Qureshi S (1996) International Classification of Childhood

Cancer1996. IARC Technical Report No.29, International Agency for

Research of Cancer, Lyon.

ICCC GROUP MORPHOLOGY TOPOGRAPHY

IV Sympathetic Nervous

System Tumors

(a) Neuroblastoma and

ganglioneuroblastoma

9490, 9500 C00.0-C80.9

(b) Other sympathetic

nervous system tumors

8680, 8693-8710, 9501-9504, 9520-

9523

C00.0-C80.9

V Retinoblastoma

9510-9512 C00.0-C80.9

VI Renal Tumors

(a) Wilms’ tumor, rhabdoid

and clear cell sarcoma

8963 C64.9, C80.9

8960, 8964 C00.0-C80.9

(b) Renal carcinoma 8010-8041, 8050-8075, 8082, 8120-

8122, 8130-8141, 8143, 8155, 8190-

8201, 8210, 8211, 8221-8231,8240,

8241, 8244-8246, 8260-8263, 8290,

8310, 8320, 8323, 8401, 8430, 8440,

8480-8490, 8504, 8510, 8550, 8560-

8573

C64.9

8312 C00.0-C80.9

renal tumors

8000-8004 C64.9

VII Hepatic Tumors

(a) Hepatoblastoma 8970 C00.0-C80.9

(b) Hepatic carcinoma 8010-8041, 8050-8075, 8082, 8120-

8122, 8140, 8141, 8143, 8155, 8190-

8201, 8210, 8211, 8230, 8231,8240,

8241, 8244-8246, 8260-8263, 8310,

8320, 8323, 8401, 8430, 8440, 8480-

8490, 8504, 8510, 8550, 8560-8573

C22.0, C22.1

8160-8180 C00.0-C80.9

hepatic tumors

8000-8004 C22.0, C22.1

VIII Malignant Bone

Tumors

(a) Osteosarcoma 9180-9200 C00.0-C80.9

(b) Chrondosarcoma 9220-9230 C00.0-C80.9

9231, 9240 C40.0-C41.9

9363, 9364 C40.0-C41.9

(d) Other specified

malignant bone tumors

8812, 9250, 9261-9330, 9370 C00.0-C80.9

(e) Unspecified malignant

bone tumors

8000-8004, 8800, 8801, 8803, 8804 C40.0-C41.9

177

National Cancer Institute SEER Pediatric Monograph

INTERNATIONAL CLASSIFICATION OF CHILDHOOD CANCER (ICCC)

Source: Kramárová E, Stiller CA, Ferlay J, Parkin DM, Draper GJ, Michaelis J,

Neglia J, Qureshi S (1996) International Classification of Childhood

Cancer1996. IARC Technical Report No.29, International Agency for

Research of Cancer, Lyon.

ICCC GROUP MORPHOLOGY TOPOGRAPHY

IX Soft-Tissue Sarcomas

(a) Rhabdomyosarcoma and

embryonal sarcoma

8900-8920, 8991 C00.0-C80.9

(b) Fibrosarcoma,

neurofibrosarcoma and

other

fibromatous neoplasms

8810, 8811, 8813-8833, 9540-9561 C00.0-C80.9

(d) Other specified soft-

tissue sarcomas

8840-8896, 8982, 8990, 9040-9044,

9120-9134, 9150-9170, 9251, 9581

C00.0-C80.9

8963 C00.0-C63.9, C65.9-

C76.8

9231, 9240, 9363, 9364 C00.0-C39.9, C44.0-

C80.9

9260 C00.0-C39.9, C47.0-

C76.8

(e) Unspecified soft-tissue

sarcomas

8800-8804 C00.0-C39.9, C44.0-

C80.9

X Germ-Cell,

Trophoblastic and other

Gonadal Neoplasms

(a) Intracranial and

intraspinal germ-cell

tumors

9060-9102 C70.0-C72.9, C75.1-

C75.3

(b) Other and unspecified

non-gonadal germ-cell

tumors

9060-9102 C00.0-C55.9, C57.0-

C61.9, C63.0-C69.9,

C73.9-C75.0, C75.4-

C80.9

tumors

9060-9102 C56.9, C62.0-C62.9

(d) Gonadal carcinomas 8010-8041, 8050-8075, 8082, 8120-

8122, 8130-8141, 8143, 8155, 8190-

8201, 8210, 8211, 8221-8241, 8244-

8246, 8260-8263, 8290, 8310, 8320,

8323, 8430, 8440, 8480-8490, 8504,

8510, 8550, 8560-8573

C56.9, C62.0-C62.9

8380, 8381, 8441-8473 C00.0-C80.9

(e) Other and unspecified

malignant gonadal

tumors

8590-8670, 9000 C00.0-C80.9

8000-8004 C56.9, C62.0-C62.9

INTERNATIONAL CLASSIFICATION OF CHILDHOOD CANCER (ICCC)

178

National Cancer Institute

SEER Pediatric Monograph

Source: Kramárová E, Stiller CA, Ferlay J, Parkin DM, Draper GJ, Michaelis J,

Neglia J, Qureshi S (1996) International Classification of Childhood

Cancer1996. IARC Technical Report No.29, International Agency for

Research of Cancer, Lyon.

ICCC GROUP MORPHOLOGY TOPOGRAPHY

XI Carcinomas and other

Malignant Epithelial

Neoplasms

(a) Adrenocortical carcinoma 8370-8375 C00.0-C80.9

(b) Thyroid carcinoma 8010- 8041, 8050-8075, 8082, 8120-

8122, 8130-8141, 8155, 8190, 8200,

8201, 8211, 8230, 8231, 8244-8246,

8260-8263, 8290, 8310, 8320, 8323,

8430, 8440, 8480, 8481, 8500-8573

C73.9

8330-8350 C00.0-C80.9

carcinoma

8010- 8041, 8050-8075, 8082, 8120-

8122, 8130-8141, 8155, 8190, 8200,

8201, 8211, 8230, 8231, 8244-8246,

8260-8263, 8290, 8310, 8320, 8323,

8430, 8440, 8480, 8481, 8504, 8510,

8550, 8560-8573

C11.0-C11.9

(d) Malignant melanoma 8720-8780 C00.0-C80.9

(e) Skin carcinoma 8010- 8041, 8050-8075, 8082, 8090-

8110, 8140, 8143, 8147, 8190, 8200,

8240, 8246, 8247, 8260, 8310, 8320,

8323, 8390-8420, 8430, 8480, 8542,

8560, 8570-8573, 8940

C44.0-C44.9

(f) Other and unspecified

carcinomas

8010-8082, 8120-8155, 8190-8263,

8290, 8310, 8314-8323, 8430-8440,

8480-8580, 8940, 8941

C00.0-C10.9, C12.9-

C21.8, C23.9-C39.9,

C48.0-C48.8, C50.0-

C55.9, C57.0-C61.9,

C63.0-C63.9, C65.9-

C72.9, C75.0-C80.9

XII Other and

Unspecified Malignant

Neoplasms

(a) Other specified

malignant tumors

8930, 8933, 8950, 8951, 8971-8981,

9020, 9050-9053, 9110, 9580

C00.0-C80.9

(b) Other unspecified

malignant tumors

8000-8004 C00.0-C21.8, C23.9-

C39.9, C42.0-C55.9,

C57.0-C61.9, C63.0-

C63.9, C65.9-C69.9,

C73.9-C75.0, C75.4-

C80.9

179

National Cancer Institute SEER Pediatric Monograph

INDEX

Topic Chapter

Acute erythremia I - Table I.1

Acute lymphoblastic leukemia (ALL) I

Acute myelofibrosis I - Table I.1

Acute myeloid leukemia (AML) I

Acute non-lymphocytic leukemia I

Acute panmyelosis I - Table I.1

Adrenal gland IV

Adrenocortical carcinoma XI

Age-adjusted rate (definition) Intro

Age-specific rate (definition) Intro

ALL (Acute lymphoblastic leukemia) I

Alveolar rhabdomyosarcoma IX

AML (Acute myeloid leukemia) I

Angiosarcoma IX

Astrocytomas III

Astrocytomas in infants Infants

Bilateral retinoblastoma V

Bone VIII

Bone tumors in infants Infants

Burkitt's lymphoma II

Cancer, liver VII

Cancer, renal VI

Cancers, renal VI

Case-control study (definition Intro

Central nervous system (CNS) III

Central nervous system malignancies III

Chondrosarcoma IX

Chondrosarcoma VIII

Chronic erythremia I - Table I.1

Chronic myeloid leukemia (CML) I

CML (Chronic myeloid leukemia) I

CNS (Central nervous system) III

CNS malignancies in infants Infants

CNS tumors among adolescents 15-19 year olds

Cohort study (definition) Intro

Dermatofibrosarcoma IX

Dysgerminomas X

EAPC (Estimated annual percent change) (definition) Intro

Embryonal rhabdomyosarcoma IX

Ependymoma III

Erythremia, acute I - Table I.1

Erythremia, chronic I - Table I.1

Estimated annual percent change (EAPC) (definition) Intro

Etiology, Multifactorial (definition) Intro

Ewing's (extraosseous) sarcoma IX

Ewing's sarcoma VIII

180

National Cancer Institute

SEER Pediatric Monograph

INDEX

Topic Chapter

Eye, retina V

Fibrosarcoma IX

Germ cell tumor X

Germ cell tumor, gonadal X

Germ cell tumor, intracranial X

Germ cell tumor, Intraspinal X

Germ cell tumor, non-gonadal X

Germ cell tumor, ovarian X

Germ cell tumor, testicular X

Germ cell tumors among adolescents 15-19 year olds

Germ cell tumors in infants Infants

Gliomas III

Gonadal carcinoma X

Gonadal germ cell tumor X

Gonadal tumor X

HD (Hodgkin's Disease) II

Hemangiopericytoma, malignant IX

Hemangiosarcoma IX

Hepatic carcinomas VII

Hepatic tumors VII

Hepatic tumors in infants Infants

Hepatoblastoma VII

Hepatocellular carcinoma VII

Histiocytoma, malignant fibrous IX

Histologic confirmation Intro

Hodgkin's Disease (HD) II

Hodgkin's disease among adolescents 15-19 year olds

ICCC Classification (International Classification of Childhood Cancers) Intro

Incidence (definition) Intro

Infants Infants

International Classification of Childhood Cancers (ICCC) Intro

Intracranial germ cell tumor X

Intraspinal germ cell tumor X

Juvenile chronic myelomonocytic leukemia (JMML) Infants

Kaposi's sarcoma IX

Kindey VI

Leiomyosarcoma IX

Leukemia I

Leukemia in infants Infants

Leukemias I

Liposarcoma IX

Liver cancer VII

Lymphoctytic depletion II

Lymphocytic predominance II

Lymphoid leukemia I

Lymphoma II

Lymphomas II

Lymphomas in infants Infants

Lymphoreticular neoplasms II

Melanoma XI

181

National Cancer Institute SEER Pediatric Monograph

INDEX

Topic Chapter

Melanoma among adolescents 15-19 year olds

Mixed cellularity II

Mortality Mortality

Mortality rate (definition) Intro

Myelofibrosis, acute I - Table I.1

Myeloid sarcoma I - Table I.1

Nasopharyngeal carcinoma XI

Neoplasms, lymphoreticular II

Nerve sheath tumor, malignant peripheral IX

Neuroblastoma in infants Infants

Neuroblastomas IV

NHL (non-Hodgkin's lymphoma) II

Nodular scerosis II

non-Hodgkin's lymphoma (NHL) II

Non-Hodgkin's lymphoma among adolescents 15-19 year olds

Osteosarcoma VIII

Ovarian germ cell tumor X

Ovary X

Panmyelosis, acute I - Table I.1

PNET (Primitive neuroectodermal tumors) III

Population Intro

Primitive neuroectodermal tumors (PNET) III

Reed-Sternberg II

Relative risk (definition) Intro

Relative survival (definition) Intro

Renal cancers VI

Retina of the eye V

Retinoblastoma in infants Infants

Retinoblastomas V

Rhabdomyosarcoma (RMS) IX

Risk factor (definition) Intro

RMS (Rhabdomyosarcoma) IX

Rye classification II

Sarcoma, myeloid I - Table I.1

Sarcoma, soft tissue IX

Sarcomas of the bone VIII

Sarcomas, soft tissue IX

SEER (Surveillance, Epidemiology and End Results) Intro

Skin carcinoma XI

Soft tissue sarcomas IX

Soft tissue sarcomas in infants Infants

Surveillance, Epidemiology and End Results (SEER) Intro

Sympathetic nervous system IV

Synovial sarcoma IX

Teratomas X

Testicular germ cell tumor X

Testis X

Thyroid cancer XI

Thyroid cancer among adolescents 15-19 year olds

Thyroid carcinomas XI

182

National Cancer Institute

SEER Pediatric Monograph

INDEX

Topic Chapter

Trophoblastic tumor X

Wilms' tumor VI

Wilms' tumor in infants Infants

Yolk sac tumors X