Cancer in Young Adults in Canada

PDF Version (1457 KB)

This report is dedicated to the memory of Sir Richard Doll, who had a long and distinguished career as an epidemiologist. Richard Doll died on July 24 2005, at the age of 92. His talk "Progress against cancer: an epidemiologic assessment" at the 1991 Meeting of the Society for Epidemiologic Research in Buffalo, New York, subsequently published in American Journal of Epidemiology (volume 134, 1991) was the inspiration for the work presented in this report.

Table of Contents

Foreword
Introduction
Overview
Individual cancers

• Non-epithelial
  • Non-Hodgkin lymphoma
  • Hodgkin lymphoma
  • Melanoma
  • Sarcoma
  • Kaposi sarcoma
  • Testis
  • Brain
  • Leukemia
• Epithelial
  • Breast
  • Thyroid
  • Cervix
  • Colon and rectum
  • Lung
  • Ovary
  • Kidney
  • Lip, oral cavity and pharynx
  • Bladder
  • Body of uterus

Risk modifiers in relation to young adult cancer patterns
Recommendations
References
Appendix: Materials and methods
For further information

Tables*

1. Summary statistics, adults aged 20-44, Canada, 1990-1999
2. Average annual percent change (AAPC) in age-standardized incidence rates and 95% confidence intervals (CI), CYAC cancers, ages 20-44, 1983-1999
3. Average annual percent change (AAPC) in age-standardized mortality rates and 95% confidence intervals (CI), CYAC cancers, ages 20-44, 1983-1999
4. Five-year relative survival (%) and 95% confidence intervals (CI), CYAC cancers, ages 20-44
5. Summary of epithelial and non-epithelial cancer patterns in young adults: risk modifiers, time trends, incidence across age range
6. Risk modifiers related to high levels of unopposed estrogen
7. Oral contraceptive use by Canadian women
   

A1. Definitions of CYAC cancers and criteria for inclusion
A2. CYAC sarcoma definition
A3. Weights used for age-standardization in CYAC age groups
A4. Subgroup definitions for CYAC cancers

* See individual cancer chapters for additional tables and figures.

Figures*

1. All cancers, Age-specific rates for young adults, Canada, 1990-1999
2. Epithelial and non-epithelial cancers, Age-specific incidence rates for young adults, Canada, 1990-1999
3. All cancers, Age-standardized incidence rates for adults aged 20-44, Selected international regions, 1993-1997
4. All cancers, Age-standardized incidence rates for adults aged 20-44, Canadian regions, 1990-1999
5. Most common cancer diagnoses, Males aged 20-44, 1990-1999, N=38,339
6. Most common cancer diagnoses, Males aged 45 & older, 1990-1999, N=574,630
7. Most common cancer diagnoses, Females aged 20-44, 1990-1999, N=62,035
8. Most common cancer diagnoses, Females aged 45 & older, 1990-1999, N=491,527
9. Most common cancer deaths, Males aged 20-44, 1990-1999, N=10,499
10. Most common cancer deaths, Females aged 20-44, 1990-1999, N=13,529
11. All cancers, Age-standardized rates for adults aged 20-44, Canada, 1983-2005
12. All except sex-specific cancers, Age-standardized rates for adults aged 20-44, Canada, 1983-2005
13. Epithelial and non-epithelial cancers, Age-standardized incidence rates for adults aged 20-44, Canada, 1983-2005
14. Incidence for selected cancers, Age-standardized rates for males aged 20-44, Canada, 1983-2005
15. Incidence for selected cancers, Age-standardized rates for females aged 20-44, Canada, 1983-2005
16. Breast cancer, Age-standardized rates for females aged 20-49, Canada, 1983-2005
17. Mortality for selected cancers, Age-standardized rates for males aged 20-44, Canada, 1983-2005
18. Mortality for selected cancers, Age-standardized rates for females aged 20-44, Canada, 1983-2005
19. Cancers with 20,000 or more potential years of life lost, Adults aged 20-44, Canada, 1990-1999
20. Percentage of Canadians aged 18-44 who are physically inactive
21. Percentage of Canadians aged 18-34 with BMI >=30
22. Percentage of Canadians aged 18-44 exceeding alcohol guidelines
23. Percentage of Canadians aged 15-24 currently smoking, 1945-2005
24. General fertility rates for women aged 15-49, Canada, 1961-1990
25. Median age of mother at first live birth, Canada, 1961-1990

Highlights

• More than 150,000 Canadians are survivors of a cancer diagnosed since 1980 in their young adult years.
• Approximately 10,000 cancers are diagnosed in young adults every year.
• Cancer mortality rates in general fell for young men and women between 1983 and 1999.
• Cancer is the main cause of early death among young adult women.
• Almost two-thirds of young adult cancers occur in young women. Of these, breast cancer is the most common.
• More young women than young men are now diagnosed with, and die from, lung cancer.
• Testicular cancer is the most common cancer in young men, and its incidence rose between 1983 and 1999.
• Melanoma is the second most common cancer among young adults.
• Increases in incidence are most striking for thyroid cancer in young men and women, lymphoma in young women and testicular cancer.
• Incidence is decreasing among young adults for many preventable cancers. Young Canadians appear to be increasingly following three main recommendations -- avoiding smoking, minimizing sun exposure and having regular Pap tests.

Call to Action

Acknowledgements

This report could not have been produced without the technical support and graphic skills of Sandrene Chin Cheong and the administrative support of Yen Borrego, both at Cancer Care Ontario.

The authors acknowledge the contributions of Jennifer Frood (Cancer Care Ontario) during the early development of this work, and the assistance and support of colleagues, in particular Eric Holowaty, Terry Sullivan, Bronwen Waller and Erin Pichora (Cancer Care Ontario) and Yang Mao, Robert Semenciw, Chris Waters, Wei Luo and Wendy Thompson (Public Health Agency of Canada). Mary Hodges and Catherine McGowan assisted Shirley Huchcroft.

Drs. Max Parkin (Oxford University), Anthony Miller (University of Toronto) and Don Wigle (University of Ottawa) critically reviewed a draft of this report. Their insightful comments resulted in important improvements.

Bo Miedema, Ryan Hamilton and Julie Easley (Dalhousie University Family Medicine Teaching Unit, New Brunswick) provided a personal perspective, with quotations from survivors themselves. These were collected as part of a qualitative study of the impact of cancer in young adults.

The following agencies provided the data used in this report:

• The Public Health Agency of Canada (cancer mortality and health survey data)
  • The mortality data were provided to the Public Health Agency of Canada from
    the Canadian Vital Statistics databases at Statistics Canada with the knowledge
    and consent of the provincial and territorial vital statistics registries which
    supply the data to Statistics Canada. Their cooperation is gratefully
    acknowledged.
  • Parts of this analysis are based on the Statistics Canada micro-data which
    contains anonymized data collected in the National Population Health Survey
    1994-1998. All computations on these micro-data were prepared by the Public
    Health Agency of Canada and the responsibility for the use and interpretation
    of these data is entirely that of authors.
• The provincial and territorial cancer registries (cancer incidence data).
• The International Agency for Research on Cancer (pre-publication international
  cancer incidence data for 1993-1997).

Monika Dixon, Canadian Cancer Society, provided excellent administrative support to produce, print, translate and distribute this report.

Foreword

There are few countries with populations the size of Canada's that are covered entirely by population-based cancer registries, and even fewer in which such coverage extends far enough back in time to permit examination of trends in cancer incidence for two decades. This precious resource is what has allowed the preparation of this volume in which the patterns and trends in cancer incidence (and mortality) in young adults in Canada are presented. There is no standard international definition of "young adult". The age range chosen by the authors (ages 20-44) has a sound rationale: it stretches from "adolescence" (generally defined as ages 15-19) to the time of life around which, in women, menopause induces quite marked changes in the cancer profile. The value of studying cancer in young adults has been clearly stated by Sir Richard Doll, in his now-classic article published in 1991. He pointed out that trends in this age group are particularly valuable in identifying the causes of cancer, since they must reflect exposures to carcinogens that have occurred in the relatively recent past, and that when such exposures vary according to specific generations ("birth cohorts"), then the consequent changes in disease risk will be first observed in the young. The authors of this monograph add a third reason--the relative importance of inherited cancer syndromes among young cancer cases--which provides the opportunity to identify the responsible genes and their interaction with risk factor exposures.

The monograph is an example of what is often defined as "descriptive epidemiology", presenting the risk of different types of cancer (defined in terms of anatomical site, and often by histological features too) according to sex, age, geography (within Canada, and internationally), and over time. But, as in all good scientific enquiry, the observations are used to make deductions about the possible reasons underlying them, artefactual or causal. Patterns and trends of the different disease entities can be compared with what is known of the prevalence of known or suspected risk factors, and their changes over time. It is in this type of comparison that incidence data from cancer registries are particularly valuable, in giving a direct measure of the risk of disease. Mortality rates are a poor substitute, particularly in this age group, where survival varies between different populations, and, with major advances in chemotherapy for several cancers in the last decades, is changing rapidly over time. These changes in survival mean that variations in mortality cannot be interpreted in terms of differential exposure to possible risk factors. A further obvious limitation is that the deaths occurring in a particular time period and age group relate to cases diagnosed several years earlier, and at younger ages.

It is manifestly not pleasant to develop cancer at any age, even in extreme old age, but in the most active and productive years of life, with maximum family and social responsibilities, it is a particular tragedy. Simple statistics such as the number of new cases of different cancers, the numbers of deaths, and the numbers of years of life lost that these deaths represent are invaluable in setting priorities for cancer control activities. The comprehensive review of numbers and causes in this monograph provides an invaluable planning resource, and is a useful basis from which to evaluate how the occurrence of cancer in this age group should be weighed when considering strategies for cancer control in general. Future projections of incidence and mortality have been used to set targets against which to evaluate progress in prevention and treatment, the monograph incorporates some short-term projections of incidence, and the importance of extending this work is included among the recommendations.

In his keynote address to the 1985 meeting of the International Association of Cancer Registries, Dr. Peter Greenwald of the US NCI stated that cancer registries could only justify their existence if they spent as much time and effort on analyzing, interpreting and presenting information as they did on collection, coding and data management. This volume is an admirable example of how cancer registry data can, and should, be made available, for the benefit of planners, researchers, and the public, with the ultimate aim of reducing the burden of this important cause of death and disability among the young.

Dr. D. Maxwell Parkin
University of Oxford
President, International Association of Cancer Registries

Introduction

Approximately 10,000 Canadians aged 20-44 were expected to be diagnosed with cancer in 2005 and 2,000 of them were expected to die from it. Based on data from 2003, it is estimated that more than 150,000 Canadians are survivors, with all its attendant difficulties, of a cancer diagnosed since 1980 in their young adult years.

Society has a "duty of care" to protect its younger and more vulnerable members, including children, adolescents and to some extent young adults. A diagnosis of cancer in a young adult has far-reaching consequences. Affected individuals may either spend decades living with the effects of cancer and its treatment (physical, reproductive, social, emotional and spiritual), or have tragically shortened lives with major repercussions on their families and society in general. They may still be completing their education, establishing their economic independence and building their own families when they are diagnosed and treated. The short- and long-term economic consequences of their cancer may be substantial.

The importance of studying cancers in young adults

Although cancer can affect individuals at any age, including infants, it is largely a disease associated with older age; 44% of all cancers are diagnosed after the age of 70.25 Consequently, overall cancer patterns reflect cancers that are most prevalent in middle and old age, especially breast, prostate, lung, and colorectal. Cancer patterns at younger ages (childhood, adolescence and young adulthood) differ, however, from patterns at older ages and even among themselves.

There are several reasons to examine cancer patterns specifically in young adults. First, cancers occurring during young adulthood result from exposures relatively early in life. While it is generally believed that many childhood cancers arise as a result of events prior to birth, cancers in young adults are more likely to be related to post-natal exposures. Since the duration of time for relevant exposure is relatively short, young adults will have had fewer accumulated exposures. Potential etiologic factors may
therefore be easier to identify from systematic examination of cancer patterns in this age group. Furthermore, the developing tissues of children and adolescents may be more sensitive to some carcinogenic events, which may manifest as a cancer diagnosed in early adulthood. Some of these cancers may therefore represent "sentinel events" providing a warning of "new" or changing exposures. For example, excess sun exposure in childhood appears to be particularly important in the etiology of cutaneous malignant melanoma, one of the most common cancers in the young adult age group. There is some evidence that this is due to enhanced sensitivity to the carcinogenic effects of sunlight early in life.⁸

Second, lifetime cancer risks tend to be shared to some extent by individuals who are born about the same time (a "birth cohort"). For example, today's adolescent boys have a much lower prevalence of smoking than their grandfathers did, and these boys will likely continue, through their lives, to smoke less. Thus, as they age they will have lower rates of lung cancer compared to their grandfathers at comparable ages. This "cohort effect" will first be evident during young adulthood, when lung cancer begins to appear. The incidence of lung cancer in today's young adults will therefore be important for forecasting future lung cancer rates and for planning preventive strategies and cancer services.

Third, cancers that occur in this age group may reflect heightened genetic susceptibility to the disease, for example, in individuals who inherit a cancer predisposition gene (e.g., BRCA 1 or BRCA 2 mutation).6 Cancer patterns in young adults may therefore suggest directions for investigation of genetic factors and interactions between genetic factors and risk factor exposures.

Sir Richard Doll argued strongly for surveillance of cancer trends in young adults in 1991, during his address to the Society for Epidemiologic Research. He challenged epidemiologists by saying:

The trends in young adults are, I suggest, by far the most important for assessing our progress against cancer for two reasons. First, because the trends can reflect only relatively recent changes in the prevalence of carcinogenic agents and are not confused by the effect of changes in the distant past and, second, because young people tend to adopt new habits before the old.⁴¹

In addition to the theoretical reasons noted above, systematic surveillance can serve a number of more immediate purposes. First, it can highlight etiologic research priorities that would not otherwise be evident. For example, increasing trends in the incidence of testicular cancer are alarming and should be stimulating research aimed at identifying exposure changes in the early lives of males. Surveillance could also help to identify areas where prevention should be focussed in attempts to stem both the current and future tides of cancer. For example, the increasing incidence of melanoma, for which the primary risk factor is known to be overexposure to the sun, would highlight the need for emphasis on sun safety early in life.

Finally, knowing more about the size and characteristics of the affected population should stimulate research into the unknown consequences of cancer in this age group, and thence to efforts to reduce harmful late effects. It can also serve as a foundation for related health system planning.

Canada's unique contribution

Canada is fortunate to have a mature nationwide system of cancer registration that covers the entire population. A registry exists in every province and territory in Canada, and all have good quality data from at least the early 1980s.¹⁰ Canada has a history of using its cancer registry data for monitoring the occurrence of cancer and for projecting the future burden of cancer for Canada and its regions. Canadian Cancer Statistics, an annual publication of data from the Canadian cancer registries, has been produced since 1987.25 Data from all the registries combined permits the study of relatively rare events, such as cancer in young adults, not only currently but also over time. No one jurisdiction within Canada and few around the world could carry out this work for a wide range of cancer types.

Specific objectives

The primary objectives of this monograph are to:

1. identify and describe the most important forms of cancer in young men and young
   women aged 20-44 years in Canada in a recent decade, 1990-1999;
2. document time trends in incidence and mortality between 1983 and 1999 for these
   cancers and for important sub groupings thereof;
3. interpret the patterns and trends; and
4. recommend priority areas for research, surveillance and policy stemming from the
   results.

The information in this report was analyzed and interpreted by:

Beth Theis¹, Diane Nishri¹, Saira Bahl¹, Anne-Marie Ugnat² and Loraine Marrett1 ¹Cancer Care Ontario and ²Public Health Agency of Canada (formerly part of Health Canada)

with support from the Cancer in Young Adults in Canada (CYAC) Working Group:

Christofer Balram, Provincial Epidemiology Service, New Brunswick
Sharon Buehler, Memorial University of Newfoundland
Ron Dewar, Cancer Care Nova Scotia
Douglas C. Dover, Alberta Cancer Board
Dagny Dryer, Prince Edward Island Cancer Registry
Juanita Hatcher, Alberta Cancer Board (currently at Aga Khan University,
Pakistan)
Eric Holowaty, Cancer Care Ontario
Claire Infante-Rivard, McGill University, Quebec
Yang Mao, Public Health Agency of Canada
Mary McBride, British Columbia Cancer Agency
Nazeem Muhajarine, University of Saskatchewan
Donna Turner, CancerCare Manitoba
Hannah K. Weir, Centers for Disease Control and Prevention, USA

and with the editorial assistance of Shirley A. Huchcroft.

Citation: Material appearing in this report may be reproduced or copied without permission; however, the following citation to indicate the source must be used:
"Cancer Care Ontario: Cancer in Young Adults in Canada, Toronto, Canada, 2006." May 2006, ISBN 0-921325-10-X (print), ISBN 0-921325-11-8 (pdf)

On the Internet this report is available at http://www.cancercare.on.ca and http://www.cancer.ca

Additional copies may be requested from Divisions of the Canadian Cancer Society or by calling Cancer Information Service 1 888 939-3333 (see For Further Information).

La version française de cette publication est disponible sur demande.