UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Late effects among young adult cancer survivors Zhang, Yang 2013

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2014_spring_zhang_yang.pdf [ 613.78kB ]
Metadata
JSON: 24-1.0107198.json
JSON-LD: 24-1.0107198-ld.json
RDF/XML (Pretty): 24-1.0107198-rdf.xml
RDF/JSON: 24-1.0107198-rdf.json
Turtle: 24-1.0107198-turtle.txt
N-Triples: 24-1.0107198-rdf-ntriples.txt
Original Record: 24-1.0107198-source.json
Full Text
24-1.0107198-fulltext.txt
Citation
24-1.0107198.ris

Full Text

  LATE EFFECTS AMONG YOUNG ADULT CANCER SURVIVORS  by Yang Zhang  B.M., Tianjin University of Traditional Chinese Medicine, China, 1995 M.P.H., University of Hawaii, USA, 2004  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Population and Public Health)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)   December 2013  ? Yang Zhang, 2013 ii  Abstract Background and objective: Long term young adult cancer survivors (YACS) can face serious life-threatening complications in multiple organ systems, and these may become more clinically significant with aging. However, there have been limited attempts to understand the risk of late-effects in this group. Therefore, this thesis aimed:  to measure the overall and cause-specific risks of late effects among YACS, including late mortality, SMN and late morbidity leading to hospitalization; to identify the characteristics influence these risks; and to examine YACS? willingness in participating late effects studies in the future.   Methods: The first three studies used data collected from Childhood Adolescent and Young Adult Cancer Survivor (CAYACS) research program, an ongoing retrospective cohort study. In the fourth study, YACS were surveyed regarding their willingness to participate in late effects studies in the future. Cox proportional hazard regression and multivariate logistic regression models were used to examine the relationship between outcomes (late effects and willingness of participation) and key socio-demographic, clinical-related factors, including the primary cancer diagnosis.   Results: YACS showed increased risks of mortality, SMN and late morbidity leading to hospitalization compared with the general population. The diagnosis of the primary cancer had a significant impact on survivors? mortality and SMN. The highest risk of mortality was observed among central nervous system (CNS) tumor survivors, whereas the highest risk of SMN were seen in survivors of lymphoma. The risk of late morbidity were higher among survivors receiving all three treatment modalities, including chemotherapy, RT and surgery. The survey iii  study found that a large majority of the respondents were always willing to participate future genetic studies. Study methods, study sponsorship, and health concerns affected subjects? willingness to some degree. Ethnicity and income were independently associated with willingness to participate.  Conclusion: This research identifies the late effects among YACS and the feasibility of conducting late effects studies in the future. Evidence from this work points to the importance of careful monitoring for these late health problems which could reduce the overall late effects and to the need to develop effective clinical programs and guidelines to meet the needs of this population.   iv  Preface The research chapters of this thesis were constructed as scientific manuscripts that have either been submitted for publication (Chapter 6) or been published (Chapter 3-5) in peer reviewed journals. This statement is to certify that the work presented in this thesis is conducted, analyzed, written, and disseminated by Yang Zhang. All research described in this thesis was approved by the University of British Columbia - British Columbia Cancer Agency Research Ethics Board (UBC BCCA REB); certificate number H05-60113 for the studies in Chapter 3-5 and H09-03284 for the study in Chapter 6. The manuscripts have co-authors as listed below.  A portion of chapter3 has been published. Zhang Y, Goddard K, Spinelli JJ, Gotay C, McBride ML. Risk of Late Mortality and Second Malignant Neoplasms among 5-Year Survivors of Young Adult Cancer: A Report of the Childhood, Adolescent, and Young Adult Cancer Survivors Research Program. J Cancer Epidemiol. 2012;2012:103032. Yang Zhang conceived of the study along with Mary McBride, conducted all analyses, and wrote the manuscript. Co-authors were involved in reviewing and revising the manuscript.  A portion of chapter 4 has been published. Zhang Y, Goddard K, Spinelli JJ, Gotay C, McBride ML. Risk of Late Mortality and Second Malignant Neoplasms among 5-Year Survivors of Young Adult Cancer: A Report of the Childhood, Adolescent, and Young Adult Cancer Survivors Research Program. J Cancer Epidemiol. 2012;2012:103032. Yang Zhang conceived of the study along with Mary McBride, conducted all analyses, and wrote the manuscript. Co-authors were involved in reviewing and revising the manuscript.  v  A version of chapter 5 has been published. Zhang Y, Lorenzi MF, Goddard K, Spinelli JJ, Gotay C, McBride ML. Late morbidity leading to hospitalization among 5-year survivors of young adult cancer: A report of the childhood, adolescent, and young adult cancer survivors (CAYACS) research program. Int J Cancer. 2013 Aug 26. doi: 10.1002/ijc.28453. Yang Zhang adopted Lorenzi MF?s study design in childhood cohort from the same CAYACS research team and conducted the study and subsequently wrote the manuscript. All co-authors reviewed and revised the manuscript.  A version of chapter 6 has been submitted and is currently undergoing peer review for publication. Zhang Y, Gotay C, Spinelli JJ, Goddard K, McBride ML. Willingness of Young Adult Cancer Survivors to Participate in Genetic Studies. Yang Zhang designed and implemented the study, and authored the manuscript. All co-authors reviewed and revised the manuscript.  vi  Table of Contents Abstract .......................................................................................................................................... ii?Preface ........................................................................................................................................... iv?Table of Contents ......................................................................................................................... vi?List of Tables ................................................................................................................................ xi?List of Figures .............................................................................................................................. xii?List of Abbreviations ................................................................................................................. xiii?Acknowledgements .................................................................................................................... xvi?Dedication .................................................................................................................................. xvii?Chapter 1: Introduction ........................................................................................................... 1?1.1? An introduction to young adult cancer ........................................................................ 1?1.1.1? Epidemiology of young adult cancer ...................................................................... 1?1.1.2? Biological difference ............................................................................................... 1?1.1.3? Lack of improved outcomes in young adults with cancer ...................................... 2?1.2? Summary of late effects among young adult cancer survivors ................................... 4?1.2.1? Late mortality .......................................................................................................... 4?1.2.2? Second malignant neoplasm ................................................................................... 5?1.2.3? Other medical late effects ....................................................................................... 5?1.3? Willingness to participate in late effects research ....................................................... 9?1.4? Problem statement in current research and rationale ................................................ 10?1.4.1? Gaps in current late effects studies among YACS ................................................ 10?1.4.2? Late effects-specific measurement ........................................................................ 11?1.4.3? Understanding survivors? willingness to participate in late effects studies .......... 12?vii  1.5? Study objectives and thesis structure ........................................................................ 13?Chapter 2: Systematic review: Late effects among survivors of adolescent and young adult cancer ............................................................................................................................. 17?2.1? Introduction ............................................................................................................... 17?2.2? Objectives ................................................................................................................. 18?2.3? Methods..................................................................................................................... 19?2.4? Results ....................................................................................................................... 20?2.4.1? Major studies of adolescent and young adult cancer survivors ............................ 20?2.4.1.1? Childhood cancer survivor study (CCSS) ..................................................... 20?2.4.1.2? Nordic childhood cancer cohort (NCC) ........................................................ 21?2.4.2? Late effects among adolescent and young adult cancer survivors ........................ 25?2.4.2.1? Late mortality ................................................................................................ 25?2.4.2.2? Secondary malignant neoplasm .................................................................... 27?2.4.2.3? Late morbidity ............................................................................................... 31?2.5? Discussion ................................................................................................................. 32?2.6? Conclusion ................................................................................................................ 34?Chapter 3: Late mortality among young adult cancer 5-year survivors ........................... 44?3.1? Introduction ............................................................................................................... 44?3.2? Methods..................................................................................................................... 45?3.2.1? Study population ................................................................................................... 45?3.2.2? Data collection ...................................................................................................... 45?3.2.3? Study analysis ....................................................................................................... 47?3.2.4? Study approvals ..................................................................................................... 48?viii  3.3? Results ....................................................................................................................... 48?3.3.1? Descriptive analysis .............................................................................................. 48?3.3.2? All causes of death ................................................................................................ 49?3.3.3? Factors for the risks of death ................................................................................. 50?3.4? Discussion ................................................................................................................. 50?Chapter 4: Second malignant neoplasms (SMN) among young adult cancer 5-year survivors................................................................................................................................... 62?4.1? Introduction ............................................................................................................... 62?4.2? Methods..................................................................................................................... 63?4.2.1? Study population ................................................................................................... 63?4.2.2? Data collection ...................................................................................................... 63?4.2.3? Statistical analysis ................................................................................................. 64?4.2.4? Study approvals ..................................................................................................... 65?4.3? Results ....................................................................................................................... 66?4.3.1? Descriptive analysis .............................................................................................. 66?4.3.2? Risk of SMN ......................................................................................................... 66?4.3.3? Factors affecting SMN risk ................................................................................... 67?4.4? Discussion ................................................................................................................. 67?Chapter 5: Late morbidity leading to hospitalization among young adult cancer 5-year survivors................................................................................................................................... 76?5.1? Introduction ............................................................................................................... 76?5.2? Material and methods ................................................................................................ 77?5.2.1? Study population ................................................................................................... 77?ix  5.2.2? Late morbidity outcomes ...................................................................................... 78?5.2.3? Data collection and follow-up ............................................................................... 79?5.2.4? Statistical analysis ................................................................................................. 79?5.2.5? Ethical approvals ................................................................................................... 80?5.3? Results ....................................................................................................................... 80?5.4? Discussion ................................................................................................................. 82?Chapter 6: Young adult cancer survivors? willingness to participate in genetic studies . 93?6.1? Introduction ............................................................................................................... 93?6.2? Methods..................................................................................................................... 94?6.2.1? Cohort and data sources ........................................................................................ 94?6.2.2? Survey questionnaire ............................................................................................. 95?6.2.3? Study methods ....................................................................................................... 95?6.2.4? Data analysis ......................................................................................................... 96?6.3? Results ....................................................................................................................... 96?6.3.1? Participant characteristics ..................................................................................... 96?6.3.2? Willingness to participate in genetic late effects research .................................... 97?6.3.3? Factors affecting willingness to participate .......................................................... 98?6.4? Discussion ................................................................................................................. 98?Chapter 7: Conclusion .......................................................................................................... 112?7.1? Summary and discussion of the study findings ....................................................... 112?7.1.1? Late mortality and SMN among YACS .............................................................. 112?7.1.2? Late morbidity among YACS ............................................................................. 114?7.1.3? Willingness to participate in late effects studies among YACS ......................... 115?x  7.2? Strengths, limitations and methodological concerns .............................................. 117?7.2.1? Strengths ............................................................................................................. 117?7.2.2? Limitations .......................................................................................................... 119?7.3? Impact, contribution and implications .................................................................... 120?7.4? Future research ........................................................................................................ 123?7.5? Conclusion .............................................................................................................. 124?Bibliography ...............................................................................................................................126?Appendices ..................................................................................................................................136?Appendix A AYA Group Classification ......................................................................... 136? xi  List of Tables Table 2.1 Summary of studies in childhood/adolescent/young adult cancer survivors ................ 36?Table 3.1 Demographic and disease-related characteristics of young adult cancer survivors ...... 53?Table 3.2. Observed and expected deaths, standard mortality ratios, and absolute excess risks for death  ............................................................................................................................................. 55?Table 3.3. Observed and expected deaths, standard mortality ratios, and absolute excess risks for death by sex .................................................................................................................................. 57?Table 3.4. Hazard ratios for all cause mortality ............................................................................ 59?Table 4.1. Demographic and disease-related characteristics of young adult cancer survivors ..... 70?Table 4.2. Second malignant neoplasms, standard incidence ratios and absolute excess risks of young adult cancer by sex ............................................................................................................. 72?Table 4.3 Hazard ratios for SMN .................................................................................................. 74?Table 5.1. Sociodemographic characteristics of survivors and comparison group ....................... 85?Table 5.2. Factors affecting late morbidity leading to hospitalization risk among survivors ....... 86?Table 5.3. Rate ratios of late morbidity leading to hospitalization for young adult cancer survivors vs. comparison group .................................................................................................... 88?Table 5.4. Rate ratios of late morbidity leading to hospitalization for young adult cancer survivors vs. comparison group by treatment ............................................................................... 91?Table 6.1. Descriptive summary of demographic and disease-related factors ........................... 102?Table 6.2. Impact of study characteristics on comfort level by willingness ............................... 105?Table 6.3. Impact of demographic factors on willingness to participate .................................... 109? xii  List of Figures Figure 1.1 Common types of cancer afflicting AYA .................................................................... 15?Figure 1.2 Improvement in survival afflicting AYA .................................................................... 15?Figure 3.1 Cumulative mortality by sex ....................................................................................... 61?Figure 4.1 Cumulative incidence of SMN by sex ......................................................................... 75?Figure 5.1. Incidence rates of late morbidity leading to hospitalization  ...................................... 92?Figure 6.1. Impact of study characteristics on comfort level with participation (all subjects) ... 110?Figure 6.2. Young adult survey study flow diagram  ................................................................. 111? xiii  List of Abbreviations AYA: Adolescents and Young Adults YACS: Young adult cancer survivors RT: radiation therapy COD: cause of death SMN: second malignant neoplasm HL: Hodgkin lymphoma CV: cardiovascular CAD: coronary artery disease CVD: cardiovascular disease HR: hazard ratio CCSS: Childhood Cancer Survivors Study TSH: thyroid-stimulating hormone FSH: follicle-stimulating hormone LH: luteinizing hormone SN: subsequent neoplasm SMR: standardized mortality ratio SIR: standardized incidence ratio AER: absolute excess risk NCC: Nordic childhood cancer cohort Py: person-year CAYACS: Childhood, Adolescent, and Young Adult Cancer Survivors Research Program BC: British Columbia xiv  CCS: childhood cancer survivors BCCR: British Columbia Cancer Registry SEER: Surveillance, Epidemiology and End Results NMSC: Non-melanoma skin cancer SC: solid cancers  Dx: diagnosis Ca: cancer Avg: average  pop?n:	population	Rad:	radiation	Obs: number of observed deaths Exp: number of expected deaths	MSP: Medical Services Plan PHN: Personal Health Number ICD-9: International Classification of Diseases Version 9 ICDO: International Classification of Disease for Oncology VSA: Vital Statistics Agency CCR: Canadian Cancer Registry  SES: Socioeconomic status BCCA: BC Cancer Agency CNS: Central nervous system ALL: Acute lymphoblastic leukemia  RR: Rate ratio / Relative risk xv  CI: Confidence interval. SD: Standard deviation OR: Odds ratio CTCAE: Common Toxicity Criteria for Adverse Events BCCSS: British CCSS: Childhood Cancer Survivors Study NCC: Nordic childhood cancer   xvi  Acknowledgements This dissertation could not have been written without Mary McBride and John Spinelli who not only served as my supervisors but also gave me full support throughout my academic and research programs, and maximum degrees of freedom to do what I was interested in. They have taught me more than I could ever give them credit for here. I am grateful to my committee members, Drs. Karen Goddard and Carolyn Gotay, for serving on the committee and providing thoughtful guidance on my research, for the great suggestions to the dissertation, and for those long meetings we had.   For all the help at CAYACS research team and the discussions on any imaginable subjects, my sincere thanks to Nelson Ha, Laura Game, Sharon Relova, Shannon Vogels, Shebnum Devji, Rita Parmar and Miranda Tsonis, for their help and friendship. I am also grateful to my colleagues, Maria Larenzi and Dongdong Li, the best statisticians I've worked with. Thanks to my fellow students Olivia Tseng, Derrick  Lee, and Anar Dhalla, for the enlightening experiences we shared. Thanks also to Agnes Lai and Maria Andrews, for sharing their precious first-hand experience in conducting survey study and recruiting participants.  I would like to particularly thank the young adult cancer survivors who participated in the survey study. Without their contribution, the thesis would not have been possible.   My deepest gratitude to my family. Without their unconditional love and support, I could not complete such a long journey.  xvii  Dedication  I dedicate this dissertation to my parents Wenji Zhang and Xuecheng Zhang. ??????????????????? 1  Chapter 1: Introduction 1.1 An introduction to young adult cancer Young adults are considered to be in the transition phase from childhood to adulthood, both biologically and psychologically. Enormous changes take place in this transition period, so cancer in this age group adds an extraordinary challenge to the survivors? future growth and development.     1.1.1 Epidemiology of young adult cancer Cancer is the leading cause of death due to disease for adolescents and young adults (AYA) [1]. Approximately there are 10,000 young adults are diagnosed with cancer every year in Canada [2]. Cancer occurring between the age of 15 and 29 years is 2.7 times more common than it for those occurring from 0 to 14 years, and accounts for 2% of all invasive cancer. [3]. The overall 5-year survival rate for 15-29 years old patients increased 11% from 71% in 1975 to 87% in 2007. Females have a better survival rate than males [3]. Survival is significantly higher for the older age group (20-24 years) versus those diagnosed between 15 and 19 years. Among those diagnosed between 20 and 24 years, the worst outcomes are seen in leukemia, lymphoma, CNS tumors and sarcoma patients[3]. By ethnicity, the incidence of cancer in this age group is higher among non-Hispanic whites and lower among Asians. In the US, the survival rate was worse in African Americans, American Indians, and Alaska Natives [4].   1.1.2 Biological difference The incidence of different types of cancer changes dramatically from age 15 to age 29. In young children, embryonal, small round-cell tumors are common, such leukemia and lymphomas. In middle-aged people, epithelial malignancies are far more common and account for more than 85% of cancers [3]. The types of cancer seen in adolescents and young adults represents a transition between those of children and older adults (Figure 1.1). Hodgkin lymphoma, non-Hodgkin lymphoma, melanoma, testis cancer, female genital tract malignancies, thyroid cancer, soft-tissue sarcomas, leukemia, central nervous system tumor (CNS) and bone sarcomas account for 95% of the cancers in 15-30 years old [4].  2   Cancers in this age group also have a unique spectrum of biological and etiological properties, which differ from those in younger and older groups. Adolescents differ from young adults in that they are still physically growing. Therefore they may have more severe late effects because of damage to organ development. They may also be treated in the pediatric setting rather than the adult setting which may result in different treatment for the same disease. Different long term risks of mortality and morbidity, including second cancers, may arise as a result of these discrepancies. Therefore, it is reasonable to expect these patients will be at risk for different types of late effects.  1.1.3 Lack of improved outcomes in young adults with cancer The survival improvement among young adults has seen less  progress in the last several decades, while there has been a significant improvement for children and older adults diagnosed during the same time period. Compared with both younger and older age groups, the rate of improvement in 5-year survival among Adolescents and Young Adults (AYA) cases was lower from the mid 1970s to the late 1990s (Figure 1.2). The annual improvement in the 5-year survival rate in both of the childhood groups and patients aged >50 years has been more than 1.5% per year. For the 15-24 years group, the average improvement is less than 0.5% per year [5].   Several factors may account for the lesser improvement in outcomes for AYA with cancer, including: 1) delayed diagnosis of primary cancers; 2) inadequate treatment practices and settings; 3) poor understanding of the biology and etiology of cancers in this population; 4) inadequate amount  of patients and patients data; 5) lack of availability of clinical trials for this age group and poor participation; 6) unique psychosocial and supportive care needs; 7) inconsistent treatment and follow-up care guidelines; 8) limited emphasis on prevention and early detection; and 9) limited access to care and insurance coverage [5].   Some studies indicated that the time from the onset of the first cancer-specific symptoms to the first cancer treatment was longer in AYA patients than in childhood patients [6, 7]. Particularly, the interval from onset of symptom to the diagnosis was longer in AYA group [4, 8, 9]. The delay 3  in diagnosis and treatment may caused by the lack of awareness of cancer in young adults in both young adults patients themselves and health professionals, the complexity of the disease, and/ or the issues from the healthcare system. While individual and healthcare system factors mainly affect the interval from onset of symptoms to initial healthcare contact, healthcare systems factors alone influence the time from initial contact to assessment, and the disease-related factors are associated with the time from the assessment to the cancer treatment [10]. However, it is unclear that whether the delay among AYA group is related withage-specific types of cancer or other disease-related factors [4].   The choice of specialist and facility, pediatric or adult care, is another issue related to the diagnosis, treatment and follow-up care. While pediatric specialists and facilities provide superior outcomes for cancers most common in children, such as acute lymphoblastic leukemia (ALL), acute myeloid leukemia (AML), and non-Hodgkin lymphoma, adult oncologists and facilities provide better outcomes for non-pediatric cancers, such as colorectal carcinoma and breast cancer [11-13]. Currently, most of the AYA cancer patients are treated at adult cancer facilities. In Canada, only 30% of AYA cancer patients were managed in pediatric centers [14]. One solution for adult care specialists/ facilities is to adopt pediatric treatment protocols for pediatric types of cancer. However, this approach has faced several obstacles in practice, including the health insurance coverage, the treatment schedule and regimens, and the availability of supportive care [4]. Some researchers suggested that an ultimate approach would be developing unique AYA cancer care units with physicians and nurses specialized in AYA cancer patient management [4].   The lack of improved outcomes in young adult cancer survivors is a global problem. Some factors relating to the deficit are universal, including the lack of awareness about cancer in young adults, their families, and even physicians, and the low participation rates in clinical trials [4]. Other factors, such as health insurance coverage, are different by counties. Bleyer et al. reported the similar patterns of less progress in survival improvement among young adult cancer survivors when they found that survival data from Australia were comparable to US data, despite the fact that Australia has a universal healthcare system [4]. These results indicated that the health insurance coverage may not play a major role in the lack of improvement in outcomes. Furthermore, the longer time interval to cancer diagnosis in this age group was observed in 4  countries having and not having universal healthcare system [15], suggesting that health insurance coverage itself may not ensure the access to healthcare system for young adults.   Although the rate of improvement in young adult cancer survivors has been delayed, cancer diagnosis, treatment and supportive care still have advanced in the last three decades. Hence, the number of survivors of young adult cancer has been growing. Over 80% of the young adult cancer patients have survived more than 5 years; thus, there are more than 150,000 survivors of a cancer diagnosed in young adulthood since 1980 in Canada [2]. These increases have generated a cohort of long-term cancer survivors. The underlying cancer itself as well as the necessary treatment can cause serious health consequences, which may not be clinically evident until many years after the treatment [16]. In order to fully understand the risk of these long term health consequences occurring among survivors, a systematic review of late effects among young adult cancer survivors (YACS) was conducted in Chapter 2. A summary of the medical-related results are presented here for the introducing purpose. Meanwhile, because of the lack of studies focusing on late effects of YACS, most of the results presenting here are obtained from studies among childhood cancer survivors.  1.2 Summary of late effects among young adult cancer survivors 1.2.1 Late mortality The overall risk of mortality among survivors of adolescent and young adult cancer is increased by 8-9 times compared with the general population [17-22]. The overall cumulative mortality rate has been reported between 13% and 18% at 30 years from cancer diagnosis [18-20, 23].   The risk of death was found to be higher for female survivors, for those who were older at diagnosis, and for those treated with specific therapies (treatment-related factors) which included radiation therapy (RT) and certain types of chemotherapy agents [17, 18, 20, 22, 24]. The pattern of the cause of death (COD) depended on the length of survival and the type of primary cancer. Recurrence of the primary cancer was the major COD among survivors with less than 20 years of follow-up [20]. With the increase of follow-up time, the risks of death attributable to second malignant neoplasms (SMN) and other non-recurrent, non-external related causes, such as 5  ischemic heart diseases, increased. In addition, for long-term survivors over 15 years from diagnosis, there was a significant excess cardiac death risk for males [21].   1.2.2 Second malignant neoplasm The risk of having a SMN after primary cancer in adolescents or young adult survivors of childhood cancer was much higher than that of the general population. The cumulative incidence varied between 7.7% and 11.1% at 30 years after diagnosis [25, 26]. In general, factors which have been found to be associated with increased risk of SMN include being female, having Hodgkin lymphoma (HL) as the primary cancer, receiving therapy in an earlier treatment era and older age at diagnosis [25, 27, 28].  Because many YACS are relatively young when they develop a SMN as compared with survivors of adult cancer, some scientists believe that environmental and lifestyle factors have only a small role to play in terms of developing SMNs (including smoking, alcohol consumption, and dietary habits [4]). Other etiological factors, which include the type of primary cancer and treatment modality, especially previously receiving RT and alkylating agent chemotherapy, are more likely to be critical in the development of SMN [4].  1.2.3 Other medical late effects  1.2.3.1 Cardiovascular (CV) function Radiation therapy to the heart alone is associated with an increasing risk of late cardiac toxicity over time. Coronary artery disease (CAD) is the most common cardiac condition among survivors who received mediastinal RT.  This condition mainly affects the left main and left anterior descending arteries [29] and accounts for 40-50% of cardiac morbidity events [30]. The next most common cardiac condition is valvular disease, which has a late onset (>10 years after RT) and is related to high dose RT (>30 Gy) and young age at the time of treatment [30]. Other late cardiac complications include pericardial disease, conduction abnormalities and ventricular dysfunction. After receiving mediastinal RT, the overall 30-year cumulative incidence was 34.5% for any cardiovascular disease (CVD), 12.9% for MI, and 19.7% for valvular disorders [31].   6  The combination of RT and anthracyclines chemotherapy, such as doxorubicin, results in a higher risk of late cardiac diseases. Aleman and colleagues reported that anthracyclines significantly increased the risk of heart failure and valvular disorders related to mediastinal RT (HR= 2.81 and 2.10, respectively) [31]. Myrehaug et al. reported an approximate 2-fold increase of risk (HR=1.82) of cardiac morbidity with mediastinal RT without chemotherapy, and 3-fold increase risk (HR=2.77) after combined doxorubicin and mediastinal RT [32].   Traditional cardiovascular risk factors, including smoking, hypertension, diabetes, obesity, and high cholesterol, were reported to increase the risk of heart disease among survivors. Hull et al. reported that all traditional cardiac risk factors were significantly associated with developing CAD, and that all patients who developed CAD, had at least one cardiac risk factor present [33]. In another study, patients without these cardiovascular risk factors were found to be at no significant increased risk for serious cardiac disease after mediastinum RT with an intermediate total dose of between 30 and 40 Gy [34]. Myrehaug et al. reported that the presence of diabetes was significantly associated with adverse cardiac events (HR=4.4). Similarly, Aleman and colleagues reported that established cardiovascular risk factors, except hypertension, increased the incidence of CV diseases [31].   These findings emphasize the importance of not only screening for CV conditions, but also managing cardiovascular risk factors and maintaining a healthy life style, such as smoking prevention and cessation, body weight control and regular physical activity.   1.2.3.2 Endocrine function Both chemotherapy and RT are known to affect endocrine function. Chemotherapy can have a significant effect upon gonadal function by affecting steroid hormone secretion and reproductive potential; whereas RT impacts the function of the hypothalamic-pituitary axis, the thyroid, and the gonads [35].     7  Thyroid function It is well known that RT can induce thyroid-gland disease, including hypothyroidism, benign nodules [36-38] and papillary carcinoma of the thyroid. A meta-analysis of 4012 Hodgkin lymphoma survivors found that up to 40% of the childhood Hodgkin lymphoma survivors developed thyroid disorders with a mean radiation dose on the thyroid region >=35 Gy [39].   Hypothyroidism is the most common thyroid abnormality reported after radiation exposure. A higher dose of radiation, older age at diagnosis, and female sex were reported as independent risk factors associated with an increased risk of hypothyroidism [38]. The Childhood Cancer Survivor Study (CCSS) reported a relative risk (RR) of 17.1 (P<0.001) for hypothyroidism among childhood cancer survivors compared to sibling controls [38]. However, some researchers have noted that this number might be overestimated due to the possibility that survivors are more likely to be screened for thyroid function than the comparison sibling group [40]. Therefore, it might be likely that more subclinical hypothyroidism cases were diagnosed among survivors than the comparison group.   The incidence of hyperthyroidism is far less common than hypothyroidism. However, one study found that both the dose of neck irradiation and time since diagnosis were independent predictors of hyperthyroidism [38].   When survivors received the combination of chemotherapy and RT, the RR of hypothyroidism was 1.42 compared to survivors treated with RT only [37]. One study observed five of six survivors that had abnormal levels of thyroid-stimulating hormone (TSH), free T4, or used thyroid hormones had received combined modality [41].   Gonadal function The functions of both ovaries and testes are sensitive to the treatment of chemotherapy (such as alkylating agents) and RT (direct irradiation to gonad or brain impairing hypothalamic-pituitary axis). In some cases total gonadal failure may occur.   8  Male gonadal function Two types of sex-specific side effects need to be considered among male survivors: 1) the damage of the germinal epithelium, which is responsible for production of spermatozoa, controlled by follicle-stimulating hormone (FSH) and inhibin B; and 2) damage to the Leydig cells, which are responsible for testosterone production, controlled by luteinizing hormone (LH) [35].  The germinal epithelium is more sensitive to the effects of chemotherapy and RT than the Leydig cells. As a result, germinal epithelium dysfunction results in oligospermia or azoospermia far more commonly than Leydig cell dysfunction, which causes testosterone deficiency [42].   Alkylating agents cause prolonged epithelium dysfunction in both adult and children [43-48]. In childhood survivors, Heikens and colleagues reported germ cell damage was presented in all patients treated with alkylating agents (n=19), such as the MOPP regimen, and no recovery of spermatogenesis was found after 10 years of follow-up [45], suggesting that the impairment due to MOPP chemotherapy in gonadal function could be permanent. Higher alkylating agent doses, older age (>=50 years old) and later stage of disease (stage II in this study) also contributed to poor outcome [49].   Female gonadal function The risk of ovarian failure increases with age at cancer diagnosis and treatment [4]. Some studies reported that treatment before age 13 was not associated with higher risk, whereas patients treated between age of 13 and 19 had two-fold increase in risk of having premature ovarian failure [50, 51]. High-dose of alkylating agents may cause almost 100% ovarian failure[52], and abdominal irradiation is also associated with a high risk of ovarian failure [53].   Puberty  Initiation of puberty is the process of reactivation of the hypothalamic-pituitary-gonadal axis, and it is very sensitive to the influences of any of the components of the axis. Therefore, puberal development may be adversely affected by any lesions, hydrocephalus, and irradiation to the hypothalamus area, such as the treatment of acute lymphoblastic leukemia (ALL), brain tumor, hodgkin lymphoma and slide tumors outside the central nervous system [54]. Previous studies 9  have shown that cranial irradiation among prepubertal children may induce precocious puberty, especially among girls [55]. On the other hand, studies among ALL patients with normal, spontaneous puberty during treatment showed that the growth hormone-insulin-like growth factor-I (GH-IGF-I) was affected by the intensive treatment which may lead to the diminished final height [56].   Meanwhile, both chemotherapy and radiation have negative effects on germ cells. In females, sex-hormone production involves the presence of germ cells, whereas in males, endocrine and exocrine functions are separated. The late effects of endocrine function among female survivors are much more common than male survivors [54].  1.3 Willingness to participate in late effects research The lack of knowledge of late effects among YACS emphasizes the importance of recruiting cohorts of cancer survivors and conducting long-term effect studies. Many studies have tried to test the feasibility of inviting cancer survivors into future research studies using cancer registry databases from community, state or national levels [57-59]. However, most of the studies were conducted either among adult cancer survivors or among recently diagnosed young adult cancer patients [57, 60]. Information on willingness to participate among long-term survivors of young adult cancer is very limited.   The rate of willingness to participate varied by data source. Using hospital data, Gellar et al. investigated the feasibility of future research studies among adult cancer survivors. They found that 49% of the 6,031 survivors completed a one-page survey and 33% agreed to participate in future studies [57]. A large scale study among adult cancer survivors using data from state cancer registries conducted by American Cancer Society reported a participation rate of 35.1% [57].  The participation rate of cancer survivors was affected by many factors, including socio-demographic factors, such as residential area, income, education level, ethnicity, and employment, and disease-related factors, such as type of cancer [61]. Previous studies found that high SES factors were associated with increased willingness in participation, including urban residence, 10  higher income, and more education [57]. Interestingly, some studies suggested that the potential reason for the different participation rates across socio-demographic factors was caused by the unequal representation of certain subgroups in the cohort, such as subgroups of rural residents, low income, and certain types of cancer[62].   1.4 Problem statement in current research and rationale 1.4.1 Gaps in current late effects studies among YACS Study population The majority of late effects studies among cancer survivors have been primarily focused on survivors of childhood cancer or adult cancer. Little work has been done on understanding the late effects among young adult cancer survivors (YACS). A literature review by Woodward et al. in 2011 could not identify studies that primarily focused on survivors of young adult cancer [63]. To update our knowledge of late effects research among YACS, a literature review was conducted in chapter 2. Our review found that a number of studies of childhood cancer survivors included some AYA subjects. Much less is known about the effects among survivors of AYA cancer. The magnitude of the late effects among this population has not been fully explored.   Data sources  Data sources may also affect the reliability of studies. Studies conducted in a single institute or clinic may have limited generalizability to other populations because of the homogeneity of the participant group. Although a self-reported questionnaire can be a convenient method to collect information, it may lead to inaccurate estimates of the late effects risk, such as SMN. In contrast, medical examinations conducted by physicians are more likely to identify asymptomatic conditions, which might give a more accurate picture of the survivors? state of health. However, when the survivors? late effects from medical examinations are compared with the prevalence in the general population, an overestimate in the risk of the adverse health outcome might be inevitable. Administrative data and medical records of cancer survivors provide the most unbiased data regarding health outcomes. However, without comparing the prevalence of adverse 11  events in survivors to that of a healthy population, such as the general population, the risk of late effects among survivors cannot be fully estimated.   In reality, data from multiple sources are necessary, including administrative data, medical records and questionnaires. In this way, a broad range of outcomes can be captured, such as survivors? experience and their personal perspectives. The comparison of health outcomes between cancer survivors and a healthy population group will provide better opportunities to identify causal relationships.   Study design Much of the research on late effects used a cross-sectional study design. Although a cross-sectional study, such as survey study, can provide suggestive associations between certain possible risk factor(s) and a late effect, such data cannot be used to establish temporal relationships between the risk factors and the onset of late effects. Plus, long-term late effects information might be limited due to the absence of follow-up in the relevant period.   Since cancer survivors are a relatively small proportion of the population, a case-control study design of a specific late-effect is not a practical option as an extremely large number of patients would be required to contain even a small number of cancer survivors.   In contrast, a cohort study can provide information on long-term risk evaluation for several outcomes simultaneously. A prospective cohort study would be appealing, but would require extremely long accrual and follow-up periods to obtain sufficient numbers. Therefore, a retrospective cohort study with a matched general population group is the preferred design for studying late effects in cancer survivors.   1.4.2 Late effects-specific measurement The literature review presented above reveals that because of the cancer itself and its treatment, cancer survivors may suffer from a variety of significant adverse late and chronic complications, including serious life-threatening health problems which can affect almost every organ system 12  [29]. These late effects have been shown to become increasingly clinically significant with aging [64]. Most of the previous studies of late effects have used either traditional measures tailored to  general chronic diseases, such as incidence and mortality, or adverse event measures merely focusing on severity of the conditions. The fact that different studies of late effects in cancer survivors have used different methods and measurement of late effects leads to difficulty in comparing results across studies [65-67]. In addition, little is known about the magnitude of risks of late morbidity, and disease-related risk factors  among YACS.  1.4.3 Understanding survivors? willingness to participate in late effects studies It is increasingly recognized that genetic susceptibility is very likely to play a role in determining the risk of developing long-term late effects [68]. But the current understanding about how genetic factors may affect the development of late effects is very limited. Accruing sufficient numbers of participants is critical to ensure that the results of late effects studies are representative and reliable, and thus will provide benefit to both individuals and public health. Therefore, it is important to successfully advocate for long-term follow-up studies, including genetic studies, to YACS.  However, the recruitment of YACS to participate in these studies is likely to be challenging. So far, the majority of the research studies have focused on finding socio-demographic and other clinical-related predictors associated with willingness among participants, and have yielded conflicting results. There has not been a study that assessed the willingness of YACS to participate in late effects studies, much less those that involve genetic factors. It is even more important to understand the factors that affect participants? willingness to participate in studies. These factors may include things such as study design, recruitment method and study sponsors. It is well recognized that willingness to participate is a complex behavioral issue, influenced by many factors and that rates may vary in different populations. A better understanding of YACS? willingness to participate in late effects studies can provide important knowledge to improve the participation rate, and further improve our understanding of late effects among the YACS. A survey study with a questionnaire offers a tool to quantify YACS? willingness to participate in a late effects study. 13   1.5 Study objectives and thesis structure  It is clear from the literature that the number of young adult cancer survivors is increasing, and there have been limited attempts to understand the risk of late-effects in this group. The primary objectives of this study were: 1) to measure the overall and cause-specific risks of late mortality and SMN among YACS and compare them with the general population; 2) to assess the risks of late morbidity that could be utilized in future monitoring measures among YACS; 3) to identify the characteristics that influence these risks; and 4 ) to quantify YACS? willingness to participate late effects studies in the future.   This dissertation consists of seven chapters: the introductory chapter, a systematic review chapter, four chapters that address the results of our studies related to the objectives mentioned above, and a concluding chapter. The primary objectives of each chapter are described below.   Chapter two is a systematic review of late effects among young adult cancer survivors. Specific research questions that are answered in this chapter include: 1) identifying the current major study cohorts of young adult cancer survivors; 2) the variation in the estimation of young adult cancer survivors? risks of late mortality, second malignant neoplasm and late morbidity; and 3) the variation in the demographic, SES, and clinical factors associated with late effects outcomes among survivors of young adult cancer.  Chapter three to six address the study objectives in order using data from administrative database and medical records. Chapter three presents the results of a retrospective cohort study of late mortality among YACS. The specific objectives of this study were: 1) to assess the long-term risks of overall and cause-specific mortality among survivors of YACS, compared to the risk in the British Columbia general population; 2) to evaluate the impact of demographic and clinical factors affecting these risks.   In order to assess the long-term risk of second malignant neoplasms for YACS, Chapter four presents the results of a population-based study. The specific research objectives were: 1) to 14  assess the long-term risk of overall and diagnosis-specific SMN in a cohort of YACS in British Columbia, compared with the general population; 2) to estimate the effect of demographic and disease-related factors on risk.   Chapter five presents the results of a retrospective study using linked registry, clinical, and long-term outcome data in cancer survivors compared to a general population control group. The specific research questions of this study included: 1) to examine the risk of late morbidity leading to hospitalization among for YAVS comparing with the general population; 2) to estimate the effects of demographic and disease-related factors on late morbidity.  Chapter six presents the results of a survey study about YACS? willingness to participate in studies examining genetic factors related to late effects. Specific objectives of this study include: 1) to examine the extent to which young adult cancer survivors would be willing to participate; 2) to identify the impact of study characteristics on comfort level with participation; 3) to explore factors associated with survivors? comfort with participation. Figure 6.2 demonstrates detailed study procedures and participant accrual supplementary to the content of chapter six.  The final chapter demonstrates a summary of the major results and discusses the strengths, contributions and limitations of this research. Potential implications of the findings and some future research directions are discussed as well. 15  Figures Figure 1.1 Common types of cancer afflicting AYA  *Source: Closing the Gap: Research and Care Imperatives for Adolescents and Young Adults with Cancer; Report of the Adolescent and Young Adult Oncology  Progress Review Group; National Cancer Advisory Board Meeting; Sep 6, 2006  Figure 1.2 Improvement in survival afflicting AYA  16  *Source: Closing the Gap: Research and Care Imperatives for Adolescents and Young Adults with Cancer; Report of the Adolescent and Young Adult Oncology  Progress Review Group; National Cancer Advisory Board Meeting; Sep 6, 2006 17  Chapter 2: Systematic review: Late effects among survivors of adolescent and young adult cancer 2.1 Introduction In the last four decades, although effective treatment has significantly improved the survival rates of young adult cancer survivors, long-term survivors of young adult cancer are still at risk for late effects, including late mortality, second malignant neoplasm (SMN), and other late morbidities that affect almost all organ systems [29, 40, 65, 69]. It has been recognized that the recurrence of primary cancer is the major cause of death in YACS up to 20 years after diagnosis [17], whereas the cumulative incidence of adverse cardiac late morbidity continues to increase up to 30 years after diagnosis, especially for those who received the highest cardiac radiation doses and the highest cumulative doses of anthracyclines [70].  Many aspects of cancer in young adults along with its treatments can potentially compromise survivors? prognosis, even many years after the clinical ?cure?. These risks vary according to the age of diagnosis, sex, type of original cancer, type of treatment exposed, and other demographic, socio-economic and clinical-related characteristics. For example, older age at diagnosis was associated with an increased risk of subsequent neoplasm [27]. Being female, combined with the specific exposure subgroup, was associated with a higher risk of certain types of late effects, such as cardiac dysfunction, obesity, steroid-induced osteonecrosis, and primary hypothyroidism [71].   Because of their relatively young age of cancer diagnosis, young adult cancer survivors have considerably more potential years of life than older cancer survivors; therefore, long-term late effects have large impacts physically, psychologically and financially on both the survivor him- or herself and the society.   The age definition of young adult is always a concern when we define the age group. However, there is no universally agreed definition of precisely what constitutes the young adult age range. Different studies use different age ranges, such as 15-24 years, 15-39 years, or 20-44 years [72]. These variations create difficulties in comparing study results and in identifying the risk of late effects. In the US, pediatric oncologists care for patients up to 21 years old; whereas in Europe, 18  the age range of a pediatric oncology practice is restricted to under 14 or 19 years old. The incidence of different cancers changes with age. The common non-epithelial cancers of early adulthood begin to appear during the teenage years, and the incidence of these tumors continues to increase up to the late 20s in females and early 30s in males [2]. Therefore, for the purpose of the literature review in this thesis, the age range considered will be 15-29 years of age. The lower limit of this age range excludes most childhood cancer patients, and the upper limit includes most of the non-epithelial cancer patients.  Most of the late effects of young adult cancer survivors (older than 15 years) were assessed in the studies of childhood cancer. Since the treatment and the type of disease of childhood and young adult cancer patients can be similar, the study results of childhood cancer survivors may be similar to those found for young adult cancer survivors, and the research methods of childhood cancer studies can be applied to studies of young adult cancer survivors as well. However, little research has focused on the late effects of cancer on young adult survivors. Woodward et al. published a review summarizing the effects of young adult cancer between 1999 and 2009 [63]; they could not identify any studies focusing on the YACS group similar to the study cohorts for childhood cancer, such as those of the CCSS.   In order to describe the published evidence on the late effects of cancer in adolescent and young adult cancer survivors so far, and to emphasize the importance of this research field, we have conducted a literature review.  2.2 Objectives The objectives of this review were: 1) to identify large cohort studies focussed on the late effects of cancer in young adult cancer survivors; 2) to better understand the impact of the cancer diagnosis on young adults and the associated treatments on survivors? health status; and 3) to examine demographic, socio-economic, and clinical factors associated with late effects outcomes in young adult cancer survivors.   19  2.3 Methods A literature review for adolescent and young adult cancer survivors was conducted using the electronic database Pubmed. Key word searching and/or subject searching were performed. The following keywords were used: cancer, survivor, adolescent, young adult, late effects, health conditions, health status, chronic disease, health outcomes, late mortality, second malignancy, second malignant neoplasm, subsequent neoplasm, and secondary cancer. Articles were restricted to: 1) human studies; 2) studies of adolescents/young adults diagnosed with cancer between the ages of 15 to 29 who survived at least 3 years post-diagnosis; 3) survivors diagnosed in 1970 and afterwards; 4) cancer survivor sample sizes of at least 500 possessing sufficient study power to be examined for specific outcomes; 5) studies published between 1990 and 2011; and 6) cohort or case-control studies. Furthermore, only studies written in English were consulted. Only one study in the review was exempted from the 3 years post-diagnosis criteria, because of the large number of young adult cancer survivors (N= 2066 for age at diagnosis 15-24 years) [28]. Studies focusing on specific types of cancer, such as melanoma or Hodgkin lymphoma, were not included. Articles restricted to quality of life or health status which did not affect survivors? life expectancy were also excluded.   The goal was to capture population-based studies of a wider range of late effects of cancer, including late mortality, second malignant neoplasm, and severe and life-threatening health issues among adolescent/young adult survivors.  It is well known that cancer treatment has improved dramatically since the 1970s. Before 1970, children diagnosed with cancer primarily received surgery and radiation, whereas those diagnosed after 1970 were treated with multi-modal therapies combined with chemotherapy [23]. As a result, overall mortality rates have decreased from 11-18% prior to 1970 [73, 74] to 9-13% after 1970 [75, 76]. Therefore, studies focusing on survivors diagnosed and treated before 1970 were excluded from the review.   20  2.4 Results A limited number of investigations of cancer?s late effects have been listed in Table 2.1. Some cohorts have extended their original cohorts and published updated results in different papers. Therefore, only the most updated results are listed in this review.  2.4.1 Major studies of adolescent and young adult cancer survivors 2.4.1.1 Childhood cancer survivor study (CCSS) The CCSS is a multi-institutional retrospective cohort study conducted in the USA at 26 participating centers. The original cohort recruited 20,720 5-year survivors with cancer having been diagnosed between 0 and 21 years of age from 1970 to 1986. Recently, the cohort was expanded to include survivors diagnosed between 1986 and 1999, including 26,093 eligible participants, and eventually recruited 20,729 survivors [77]. It employed a self-report questionnaire to obtain demographic characteristics, health practices, and medical conditions. Information on the treatment of the primary cancer was obtained from medical records. Information on subjects? vital status and cause of death was obtained from either the National Death Index (for deaths occurring after 1979), or from death certificates (for deaths before 1979) [17]. Subsequent neoplasm (SN) information was initially obtained and ascertained from self-or proxy reports in questionnaires and/or death certificates, and then confirmed from pathology reports or medical records [27]. Chronic health outcomes were assigned and grouped into a severity score, and graded 1 through 4 (ranging from mild to life-threatening or disabling) [16].  By the end of 2002, 2,821 deaths occurred with overall cumulative mortalities of 18.1% at 30 years post diagnosis [17, 18]. The overall SMR was 8.4 (95%CI, 8.0-8.7). Among subjects with longer follow-up time, mortality due to primary cancer, including recurrence, decreased, while mortality due to SMN (SMR, 15.2; 95% CI, 13.9 to 16.6), cardiac disease (SMR, 7.0; 95% CI, 5.9 to 8.2) and pulmonary disease (SMR, 8.8; 95% CI, 6.8 to 11.2) increased. RT, alkylating agents and epipodophyllotoxins increased the risk of death due to SMN. Cardiac RT and anthracyline increased the risk of death due to cardiac disease [18].   21  1,402 patients who developed 2,703 neoplasms were identified among 14,359 5-year survivors. The cumulative incidence of SN at 30 years was 7.9% (95% CI = 7.2% to 8.5%) excluding non-melanoma skin cancer. The overall SIR was 6.0 (95% CI = :5.5 to 6.4). Hodgkin lymphoma (SIR = 8.7, 95% CI = 7.7 to 9.8) and Ewing sarcoma (SIR = 8.5, 95% CI = 6.2 to 11.7) had the highest risk of an SMN. Being female, older age at diagnosis, having treatment in earlier years, a primary diagnosis of Hodgkin lymphoma, and having RT increased the risk of SNs [27].   The health status of 10,397 cancer survivors was compared to that of 3034 nearest-age siblings. Over 62% of survivors reported at least one chronic condition, and 27.5% had a severe or life-threatening condition (grade 3 or 4). The cumulative incidence of a chronic health condition reached 73.4% (95% CI, 69.0 - 77.9) at 30 years after the cancer diagnosis, and 42.4% (95% CI, 33.7 to 51.2) for severe, disabling, or life-threatening conditions or deaths due to chronic conditions. Compared to that of siblings, the risk was 3 times higher for a chronic condition, and 8 times for a severe of life-threatening condition [16]. The major limitation of this study is the self-report data collection method. Without external verification, some health conditions collected from questionnaire, such as hypertension and osteoporosis, may be underreported. In addition, mental health outcomes were not collected in this study.   2.4.1.2 Nordic childhood cancer cohort (NCC)  The NCC is a population-based cohort with survivors diagnosed with cancer before 20 years of age. Information on both the cause of death and the incidence of SMNs was obtained from country-wide cancer registers in the five Nordic countries (Sweden, Denmark, Finland, Norway, and Iceland ).   In a recent updated report published in 2006, there were 2324 deaths among 21,984 5-year survivors diagnosed between 1960 and 1999. The overall SMR was 8.3, and the AER was 6.2 per 1000 person-year (py) [19]. The cumulative mortality was 9.7% at 20 years, and 14.1% at 30 years after diagnosis. The pattern of the cause of death (COD) varied according to the type of primary cancer and the survival time, namely: primary cancer was the major COD among survivors having shorter follow-up, whereas SMN and non-cancer causes were the main COD among survivors with longer follow-up [19].  22   Olsen et al. reported 1180 SMN cases, 28% of which were brain tumors, among 47,679 cancer survivors diagnosed from 1943 to 2005 [78]. The SIR was 3.3 (95% CI, 3.1 -3.5). The cumulative risk for the 1975- 2005 (time of diagnosis) sub-cohort was 13.3% by the age of 50, excluding non-melanoma skin cancer [78]. 28% of SMNs were reported as brain tumors. This result was higher than the results from other studies. This may be due to the inclusion of meningiomas cases. Meanwhile, SMN cases occurring within the first 5-year survival time  were also included in the analysis. As a result, the risk of SMN may have been overestimated when compared with the risk among the general population.  Another limitation of these studies was the lack of treatment information. Therefore, these studies could not link specific cancer types and the dose of their treatments, both through radiation and chemotherapy, with specific causes of death and sites of SMNs.   2.4.1.3 Netherlands study (Childhood cancer registry of EKZ/AMC) All the participants from the Netherlands study were five-year survivors diagnosed between the ages of 0 and 18 years and treated in a single research institution (EKZ/AMC) between 1966 and 1996. Baseline and follow-up data were abstracted from the medical records. For patients lost to follow-up, their GPs or physicians in other hospitals were contacted by mailed questionnaires to trace the patients? recent medical status. For patients still lost to follow-up, municipal registries covering the population of the Netherlands were approached to request the patients? vital status and recent medical status [20, 25, 69]. The rates of medical follow-up were remarkably high, ranging from 92% [25] to 96.9% [20] among the different studies conducted in this cohort. Only 1.5% of survivors were lost to follow-up (in the CCSS: 5.1% were lost to follow-up and 16% gave no response to questionnaires) [69]. However, since the cohort was recruited from a single institution, the homogeneity of the study population limits the generalizability of the results.   In the report by Cardous-Ubbink MC et al., there were 120 deaths among 1,378 patients (median age at diagnosis 5.9 years) with a cumulative risk at 25-years of 11% and at 30-years of 13% [20]. The SMR was 17.2 for all causes of death, 17.0 for second cancer, 2.05 for other causes, and 5.64 for cardiovascular diseases.  23   Among the 1368 survivors, 62 SMN cases were identified, including meningiomas. The cumulative risk at 20-years was 4.4% and 30-years at 11.1% [25]. The SIR was 11.2 for SMNs including meningiomas, 9.45 for SMNs excluding meningiomas, and 12.1 for solid SMNs only. The overall SIR in this study was higher than that of other studies; for example, the CCSS reported a SIR of 6.0 after excluding nonmelanoma skin cancer [27], and the Nordic study reported an SIR of 3.3 [78].   Geenen et al. conducted a retrospective cohort study using physician evaluations to report adverse health events among 1362 survivors [69]. 74.5% (N=1015) of survivors had 1 or more adverse events, 24.6% (N=316) had 5 or more adverse events, and 35.9% (N=473) of survivors had at least 1 severe or life-threatening or disabling adverse event. A major limitation of this study was the lack of a comparison group in the general population, therefore, the risk of adverse outcomes compared with the general population could not be estimated.    2.4.1.4 French ? British Study A cohort including 4,400 3-year survivors of cancer diagnosed before 17 years of age between 1942-1986 and treated in eight centers in France and the UK was identified. Baseline and follow-up data, including treatment information, were recorded from clinical records. Leukemia cases, non-melanoma skin cancer cases and retinoblastoma cases from UK were excluded.  In the report by de Vathaire et al., there were 113 SMN cases (excluding leukemia) identified for a median age at diagnosis of 6.0 years and a mean follow-up of 15 years [26]. The cumulative incidence of solid SMN was 4.9% at 25 years and 7.7% at 30 years [26]. The SIR was 9.2 for solid SMN, and the AER was 1.88 per 1,000 py. The results showed that the risk for thyroid cancer increased with the dose of RT, but not with chemotherapy. The RR of solid SMN decreased with follow-up time and with attained age, after the RT alone and after the RT plus chemotherapy. Multivariate analysis was not performed, therefore, the risks of SMN in this study were not adjusted by other social-demographic and clinical factors.   24  2.4.1.5 Slovenia study Jazbec et al. reported that among the 1577 childhood cancer survivors diagnosed before 16 years of age and treated in a Slovenia national referral center for all pediatric cancer patients from 1961 to 2000, 48 patients developed SMN (mean time of follow-up: 13.2 years), including both non-melanoma skin cancer and meningiomas [79]. The cumulative incidence of SMN was 7.4% at 20 years post-diagnosis and 12.6% at 25 years. The SIR of 8.71 was higher than the results of the CCSS and the Nordic cohort (CCSS, 6.0; NCC, 3.3), but lower than the SIR of the 11.2 from Netherlands cohort, which may reflect the discrepancy in the definition of SMN among the studies.   2.4.1.6 North England study Utilizing the Northern Region Young Person?s Malignant Disease Registry, the North England study identified 39 SMN cases among 2066 survivors of young adult cancer diagnosed between 15 and 24 years of age, including those diagnosed with basal cell carcinomas and meningiomas [28]. The median age of the primary cancer diagnosis was 20.9 years, and 21.3 years for the SMN. The SIR was 3.1 for the young adult group, and the highest SIRs were found among Hodgkin lymphoma and brain tumor subjects (SIR=5.8 and 4.3 respectively ).  2.4.1.7 Roswell Park Cancer Institute's long-term follow-up project (LTFUP) Utilizing the data from Roswell Park Cancer Institute's Long-Term Follow-Up Project, Lawless et al. reported 38 deaths among 565 15+ year survivors of cancer diagnosed before age of 20 years. The mortality risk was increased among the long-term survivors (SMR= 2.84 for male, 3.71 for female.). SMN was the leading COD in both genders. Cardiac death was elevated among male survivors, but not among female survivors. However, because of the small number of subjects in the cohort and the fact that univariate analysis was applied to all patients diagnosed from the same institute, the effects of other mortality risk factors were not examined in this study.  2.4.1.8 Childhood, adolescent, and young adult cancer survivors program (CAYACS) Using population-based administrative data, the CAYACS study identified 5-year survivors diagnosed before the age of 25 from the BC Cancer Registry (BCCR). Baseline data and SMN were obtained from BCCR, while COD was determined from Vital Status [80].  25   Among 2,354 survivors diagnosed before age 20 from 1970 to 1995, MacArthur et al. reported 181 deceased cases, including 139 cancer-related deaths [22]. SMR was nine times higher than that of the general population, and was increased for those with recurrences within the first 5 years of diagnosis and with a primary diagnosis of ALL and CNS. Mortality risk was also increased due to a cancer-related cause of death, as well as due to circulatory and respiratory diseases.  The CAYACS group identified 55 SMN cases excluding NMSC and benign tumors among 2,322 childhood cancer survivors [81]. The cumulative incidence was 3% at 20 years and 5.1% at 25 years. Being female and diagnosed before the age of 10 were associated with increased risks of SMN.  In the report by Lorenzi et al., hospitalization-related late morbidity was compared between 1374 survivors and a comparison group with 10 times the size of the survivor group [65]. 41% of the survivors and 17% of the comparison group had at least one type of hospitalization-related late morbidity, corresponding to Grades 3, 4, and 5 in Common Toxicity Criteria for Adverse Events (CTCAE). 19.1% of the survivors and 6.3% of the comparison had multiple types of morbidity. Survivors had a 4-times increased risk of late morbidity leading to hospitalization (RR=4.1, 95% CI, 3.7 - 4.5). This study excluded non-hospitalization conditions in the analysis; therefore, outcomes that do not lead to hospitalization, such as fatigue and obesity, were not measured in this study.  2.4.2 Late effects among adolescent and young adult cancer survivors 2.4.2.1 Late mortality In general, survivors of young adult cancer showed an increased risk of late mortality compared to the age-, sex- and calendar year specific rates in the general population [17-22]. The overall standard mortality ratio (SMR) was reported to be similar from the three major studies: 8.4 from the CCSS data (institutional-based; SMR=7.9 for 15-20 years age at diagnosis) [23], 8.3 from the Nordic data (population-based) [19], and 9.1 from the CAYACS data (population-based) [22], 26  but higher in the Netherlands study (SMR=17.2 ) (single institute) [20]. The overall cumulative mortality rate was reported between 13% and 18.1% at 30 years from the cancer diagnosis [18-20, 23].   The pattern of cause of death (COD) varied by the years of survival and the type of primary cancer. The recurrence of primary cancer was the major COD among survivors with less than 20 years of follow-up [20]. With the increase of follow-up time, the risks of SMR attributable to SMN and other non-recurrent, non-external related causes increased. Lawless et al. reported that SMN was the leading COD among survivors with a 15+ year follow-up, responsible for 39% (15/38) of death cases [21]. The CCSS found that almost half of the deaths were due to non-recurrent, non-externally related causes at 30 years from diagnosis [18]. Garwicz and colleagues found that SMN as a COD increased from 4% in the interval of 5-10 years after diagnosis to 33%  more than 30 years after diagnosis and non-cancer causes from 14% to 51% during the same period [19].   The risks of death due to recurrence were higher among survivors with medulloblastonma, certain types of CNS tumours, such as ependymoma, and Ewing sarcoma, as well as survivors diagnosed in an earlier treatment year or diagnosed at and older age  (15-20 years old in CCSS). Patients with an original diagnosis of medulloblastoma were at the highest risk of death due to SMNs, mainly caused by the craniospinal radiation received during the treatment. Meanwhile, patients with Hodgkin lymphoma and Ewing sarcoma were also at a higher risk of death due to SMNs [18]. Because of routine treatments using both anthracycline chemotherapy and chest/pulmonary RT, survivors of renal tumours and Hodgkin disease had the highest risk of cardiac-related death [18].   Other risk factors for mortality were also evaluated in most of the studies, including sex, year of diagnosis, age at diagnosis, and treatment. The risks of death due to SMN (CCSS: RR, 1.3; 95% CI, 1.1-1.8) and other deaths (CCSS: RR, 1.9; 95%CI, 1.5-2.5) [17, 18, 20, 22] were found to be higher for females (CCSS: SMR=13.2) than for males (CCSS: SMR=6.7). Also, for long-term survivors at 15+ years from the diagnosis, a significant excessive cardiac death risk was seen among males (SMR =4.14) [21].  27   The year of diagnosis may also play a role in the risk of late mortality. Using population-based data from the Surveillance, Epidemiology and End Results (SEER) in the US, Armstrong et al. reported a decrease in all cause cumulative mortality among childhood cancer survivors diagnosed in 1974-1980 to those diagnosed in 1995-2000 [24]. The Nordic study reported a lower mortality within 5-9 years after diagnosis among survivors from 1990 to 1999, compared to those from previous periods, but very little change in the 10+ years after diagnosis [19]. Overall, these findings indicate that the improvement in primary cancer treatment may have effectively delayed, but not prevented, death from primary cancer.  RT, alkylator or high dose levels of epipodophyllotoxin exposure were also related to a high risk of death due to SMN [18]. Cardiac RT and cumulative exposure to anthracycline increased the risk of death due to cardiac disease [18]. The CCSS reported that survivors with a cumulative anthracycline dose of 401 mg/m2 or higher had a 3.1 times higher risk of cardiac mortality [17]. A study by Tukenova et al. showed a similar result from RR of 4.4 for anthracycline doses of 360 mg/m2 [82]. These findings suggest that the treatment originally applied to cure the primary cancer may have long-term late effects on mortality even after the risk of recurrence of the primary cancer stabilizes.  The SMR for death from SMN was measured in several studies, but the results were not consistent. In the CCSS, the SMR was 15.2 for death from SMN [17]. The Netherlands study reported a similar SMR of 17.0 [20]. However, Moller et al. found a SMR for SMN mortality of 4.9 in Nordic countries [83]. The wide range in SMRs was likely related to the difference in study designs (i.e. hospital-based vs. population-based studies), the distribution of primary cancer diagnoses, the year of diagnosis, and the intensity of treatment on the primary cancer.  2.4.2.2 Secondary malignant neoplasm There are several considerations that need to be kept in mind before interpreting the second malignant neoplasm results. The definitions of secondary malignant neoplasm (SMN) were inconsistent among the studies [25-28, 78, 79, 81].  Non-melanoma skin cancer (NMSC) was 28  included in some studies, as well as benign meningiomas. A few studies included not only SMN but also all other subsequent neoplasms (SN) in the analysis, including both non-malignant meningioma and NMSC. The confirmation of SMN was either from a cancer registry or from the medical records. These discrepancies in definition may bring about some uncertainty in comparability between the studies. In addition, in some studies such as the those of the CCSS, the information of the SMN was initially obtained from self- or proxy report questionnaires: therefore, any under-reporting of the SMN diagnosis may lead to an inaccurate estimation of the SMN.   Overall, the risk among adolescents or young adults of having an SMN after a primary cancer diagnosis is much higher than that of general population. The cumulative incidence observed varied between 7.7% and 11.1% at 30 years after the diagnosis [25, 26]. Olsen et al. reported the accumulative risk of SMNs of 13.3% at the age of 50 for survivors diagnosed in 1975-2005, after the exclusion of non-melanoma skin cancer (NMSC) [78].   After excluding NMSC and non-malignant meningiomas, the CCSS showed that the SIR was 6.0 [27]. This result was slightly higher than that of the SIR(SIR=5.0) from the CAYACS study [81]. However, it was lower than that of the SIR of 9.45 from the Dutch study [25], possibly due to the  differences in SMN definitions and treatment protocols over time. In the Dutch study, basal cell carcinoma was not excluded from the analysis of SMN, nor was a third primary cancer, which may lead to an overestimation in the risk of SMN compared with the studies only including second primary cancer. Olsen JH et al. reported an SIR of 3.3 using data from population-based cancer registries in the Nordic cohort. However, the author also mentioned that the rules for coding SMN were not well-defined in the first two to three decades of the study period among these cancer registries, affecting 9% of the total SMN cases [78]. In the study of de Vathaire et al, the SIR of solid SMN (SIR=9.2) showed a substantial risk in non-leukemia 3-year survivors [26]. Considering the fact that the mean age at the end of the follow-up was younger (less than 21 years of age) than in the other studies, this result may reflect only the risk of SMNs in the childhood and young adult period. Taken together, the wide range in SMN risk estimates reported from various studies may reflect the difference in SMN definitions, as well as the differences in study designs, e.g. hospital-based versus population-based studies, and the distribution of types of 29  primary cancers, the year of diagnosis and age at diagnosis.   In general, the survivor?s sex, the type of primary cancer, the treatment era, and the age at diagnosis were significantly associated with the risk of SMN. Female survivors had an increased risk of SMN, with RRs of 1.35-1.6 independent of study design and study population [25, 27, 28]. After adjusting all of the other risk factors, the CCSS found that survivors of Hodgkin lymphoma had the highest risk of SNs (RR=1.5, 95% CI, 1.1-1.9) [27]. Furthermore, studies showed consistent evidence for an increased risk of SMN for those diagnosed during the earlier treatment era and an older age at diagnosis in multivariate analyses, even though some of the evidence is not statistically significant. CCSS data found that the SN risk was significantly lower among survivors treated in the later era (1980-1986 vs. 1970-1974) (RR=0.6, 95% CI, 0.5-0.7), and higher among survivors diagnosed at an older age (15+ years of age vs. 0-4 years of age) (RR=1.3, 95% CI, 1.1-1.6) [27]. The Dutch study reported no statistical significance by treatment era (after October, 1984 vs. before October, 1984) or by age [25].   A number of studies identified that chemotherapy including the use of alkylating agents, such as MOPP and COPP (mechlorethamine, cyclophosphamide, oncovorin, procarbazine, and prednisone) regimens, was significantly associated with a higher risk of secondary leukemia, especially acute myeloid leukemia [84, 85]. A meta-analysis reported the risk of leukemia was 37-fold (measured by standardized incidence ratio, SIR) [35]. However, Robison et al. observed that this treatment-related risk of leukemia decreased after 15 years of diagnosis among Hodgkin lymphoma survivors [86].  Meanwhile, RT is the treatment modality that is consistently associated with an increased risk of SMN, including a 2.7-fold (95% CI, 2.2-3.3) increase in relative risk (RR) reported from the CCSS, a two-fold increase in Hazard Ratio (HR) from the Dutch study, and a 2.6-fold (95%CI, 1.3-5.2) increase reported among survivors of young adult cancer (diagnosed at 15 to 19 years of age) in the Nordic study [25, 27, 87]. Although some studies have suggested chemotherapy also plays a role in the excess risk [88-91], the most common solid cancer (SC) reported after RT tend to increase after 15 years of diagnosis [92, 93].   30  Breast cancer is the most common secondary SC in females, especially after the diagnosis of Hodgkin lymphoma, accounting for about 40% of SCs among female survivors, but it has also been reported in males. For Hodgkin lymphomas survivors, the risk of breast cancer increased in diagnosed in young adult (aged 20-29 years), while the relative risk was lower than those treated in childhood and adolescence, mainly due to the increased background rates at older ages when comparing the survivor group with the general population [94].   The risk of breast cancer is also increased with an increased dose among women who received RT. In previous studies, Mantal RT (35-45 Gy to axillary, mediastinal, and neck nodes ) was associated with a 2 to 20 times increase relative risk (RR) of breast cancer, depending on the age when receiving treatment [30, 92, 95-97]. In a matched case-control study among women diagnosed before age 30 years, Travis et al. reported an 8-fold increased risk (95% CI, 2.6-26.4) of breast cancer with a dose of  >40 Gy, compared with the lower dose (<4Gy) and no alkylating agents [98]. In a study of childhood cancers treated between 1970 and 1986, Inskip and colleagues found a linear increase in breast cancer risk with RT dose, which reached 11-fold for a local breast dose of 40 Gy compared with an absence of radiation [99]. These results suggest that reducing the RT dose to a lower level, such as less than 40 Gy, can decrease the risk of secondary breast cancer.   RT is also strongly associated with secondary thyroid cancer, especially among female Hodgkin lymphoma survivors. A study from the CCSS observed an increased risk of subsequent thyroid cancer with radiation doses up to 20-29 Gy (OR,9.8; 95%CI, 3.2-34.8), although the risk decreased at doses greater than 30 Gy [100]. Both the increase and decrease of the risks were more marked in survivors diagnosed with Hodgkin lymphoma before that age of 10 years than those diagnosed at an older age [100]. In a recent study, O?Brien et al. reported 5 cases identified among 112 Hodgkin lymphoma survivors treated with low dose RT and MOPP/ABVD (adriamycin, bleomycin, vinblastine, dacabazine) after a median of 20 year follow-up [101]. These findings are consistent with the cell-killing effect of radiation at high doses, suggesting that high doses of RT kill thyroid tissue cells or that the cells that survived lost the capacity for proliferation. In contrast, after a median of 15 years of follow-up, no thyroid cancer has been identified from the CCSS among Hodgkin lymphoma survivors not treated with RT [41].  31   2.4.2.3 Late morbidity The results of the CCSS strongly indicate that survivors had a higher prevalence of self-reported chronic conditions. 62.3% of survivors had at least one chronic condition (vs. 36.8% of siblings), while 27.5% had a severe or life-threatening condition (Grade 3 or 4) (vs. 5.2 % of siblings). Thus, compared with that of the siblings, the risk among survivors increased 3 times for a chronic condition, and 8 times for a severe or life-threatening condition [16]. The CAYACS study had a slightly higher prevalence and lower relative risk. The discrepancy was mainly caused by the differences in outcome definition, type of diagnosis, length of follow-up period, type of outcome measured and the study method used in the recruitment of participants [65]. 41% of survivors versus 17% of the population sample had at least one type of hospitalization-related late morbidity. The survivors had a four-fold increase in risk for at least one type of hospitalization-related late morbidity.   The incidence of late morbidity in the Dutch study was higher than those of the report from the CCSS (74.5% vs. 62.3% of survivors had one or more adverse events, 35.9% vs. 27.5% of at least one severe, life-threatening or disabling adverse event). The potential explanation for the higher rates was that the physician evaluation method in the Dutch study was more likely to diagnose the asymptomatic conditions, comparing with the self-report method used in the CCSS.  The risk of chronic health conditions was consistently higher for SMN and endocrine disorders [16]. The CCSS reported a 14.8-fold increased risk for SMN (endocrine disorders results not shown), and the CAYACS study found an almost 22 times increased risk in SMN and an 8 times risk in endocrine, nutritional and metabolic disorders [16, 65]. Similarly, the Dutch study found that 11.9% of severe or life-threatening events were SMN, 9.4% were metabolic events and 5.3% were endocrine disorders [69].   Overall, all cancer diagnosis were observed to be significantly more likely to have either any of the conditions or severe conditions. Combinations of therapy were related to elevated risks of late morbidity. In the CCSS, exposure to a combined treatment modality was associated with at least a 32  10-fold increase in the risks of having a Grade 3 or higher morbidity [69]. A seven-fold increased relative risk (RR) of having hospital-related late morbidity was observed among survivors with combined RT, chemotherapy and surgery in the CAYACS study [65]. An increase in the cumulative dose of the alkylating agent itself or in combination with RT was associated with an increased RR, but in contrast, an increase in the cumulative dose of anthracycline appeared to have little effect on the risk of having chronic health conditions [65].   Some factors modified the risk of late morbidity. Female survivors had poor late morbidity outcomes compared with those of males,  were 1.5 times as likely to have severe conditions in the CCSS [16], and had an RR of 1.1 (95%CI, 1.03-1.18) in the Dutch study [69]. The CCSS discovered an independent modification effect of age at diagnosis after adjusting for socio-demographic factors; namely, survivors who had been diagnosed at an older age were more likely to have any of the conditions or severe conditions [16]. However, this association was not confirmed by the CAYACS data. Contrary to these results, the CAYACS study showed that a decreased risk of late morbidity was independently associated with a longer follow-up time and a later year of diagnosis [65].   2.5 Discussion We identified eight original cohorts where long-term late effects were assessed among the survivors of young adult cancer around the world. We found that cancer diagnosed in the young adult period increased the risk of late mortality, SMN and late morbidity among long-term survivors. The results suggest that cancer and its treatment have a significant impact on young adult survivors? health status from various aspects. and that this impact tends to persist for many years after the successful completion of treatment and the clinical ?cure? of the disease.  This review confirms that so far there is no large cohort study specifically focusing on the survivors of young adult cancer. While large cohort studies, such as the CCSS, were designed to detect the late effects of cancer and its treatment, the majority of the participants in these studies were childhood cancer survivors, and only a minority were the survivors of young adult cancer, namely, those older than 15 years of age at diagnosis. Although some findings from childhood 33  cancer survivors are generalizable to survivors of young adult cancer, the current literature offers little information to help understand the unique experiences held among young adult survivors, such as the risk of developing breast cancer among female survivors of Hodgkin lymphoma.   It should be noted that, since the incidence  of some health conditions discussed in these studies, such as SMN, are very low in the general population, when these rates are compared with the rates among cancer survivors, even small numbers of events in cancer survivors can lead to a significant increase in relative risk (RR). In addition, it may not be possible to separate health issues related to treatment from those attributable to primary cancer, especially when considering the fact that disease severity was not measured in these analyses. The effect of a certain type of treatment may partly reflect the persistent predisposition of the primary cancer itself as well as residual damage from the disease and/or its treatment.   In the CCSS, the information about subsequent neoplasms and chronic conditions was initially obtained from a self-report questionnaire. However, previous studies have shown that the sensitivity (the proportion of study participants with a registry-documented cancer who self-reported the cancer) ranged from 0.79 for an exact match of the cancer site with the year of diagnosis to 0.93 for a match of any reported cancer but a variance in the actual cancer site [102]. For self-reported diagnoses leading to hospitalization, the true positive rate (the probability for a self-reported hospitalization diagnosis to be confirmed by hospital records) varied by condition, from almost 100% for breast cancer and 84% for ischemic heart disease to 54% for ulcers and 32% for colon polyps [103]. It indicates that for some conditions, the self-report method may lead to an underestimation of the disease risk and, therefore, medical record verification is required in follow-up studies.  Some studies used administrative data to assess both population health and specific health conditions for both cancer survivors and the general population. The use of such reliable data sources is a significant strength, especially when it comes to describing the population as a whole. However, administrative data offers little information to help understand the effects of the social and psychological impacts, such as anxiety, of a cancer diagnosis on cancer survivors. Thus, it cannot provide results in self-perceived function and well-being in daily life settings. 34   The loss to follow-up is always a concern in a cohort study. It is possible that people who remained healthy were either more likely to move out of the area, or less likely to use the healthcare system, thus leading to a higher proportion of loss to follow-up. Therefore, the completion of follow-up information is critical to minimize the potential effect of bias. In the studies included in this literature review, the proportion of survivors completing follow-up ranged from 68% in the CCSS to 92% in the Netherlands study, 94-96% in CAYACS study (lost follow-up: 4% of the survivors and 6% of the comparison), and to 99.4% (30 cases being lost to follow-up from among 5077 subjects) in the North England study [25, 27, 28]. However, the exact impact from the loss to follow-up on the risk of late effects was not reported.  Furthermore, there are several considerations need to be kept in mind in interpreting the results from the literature. The information of the COD mainly relied on diagnoses from death certificates. The accuracy and validity of the classification of CODs have been questioned by previous studies [104, 105]. For example, in survivors, the listed CODs for certain diseases, such as cardiovascular disease, may be under- reported, or misclassified as cancer related CODs. This may lead to an underestimation of the mortality risk of these others diseases. Overestimation of the risk of late effects is also a concern when the follow-up exams for survivors are more complete than that for the general population or the comparison group, and surveillance bias may be introduced into the results.   2.6 Conclusion Our review of the literature shows that young adult cancer survivors may have an increased risk of late effects, including late mortality, second malignant neoplasms, and late morbidity. However, most studies have focused on survivors of cancer diagnosed in childhood, or on only one or two late effects, or specific diagnosis groups. Due to the heterogeneity of the study designs, it was difficult to assimilate the results across all of the studies. Our review suggests that an assessment of late effects among young adult cancer survivors is an interesting area to be studied. A critical step in the future would be to design an applicable, reliable, and valid follow-up program to measure the risk and to  identify high risk groups among survivors of young adult 35  cancer. Particular attention should be given to address the methodological issues when conducting a late effects study among these survivors.  36  Tables Table 2.1 Summary of studies in childhood/adolescent/young adult cancer survivors Reference Study population Date of Dx Data collection Outcome  Relevant Findings Comments Childhood Cancer Survivor Study (CCSS), USA Mertens AC [17]; Armstrong GT [18, 78] ?N=20,483 ?5-yr survivors ?Dx age: <21 years ? EOF: Dec 31, 2002 1970-1986 1)multi-institutional study (26 centers ) 2)medical record for primary dx treatment 3)VS and COD: 1979-2002, determined using the National Death Index; 1975-1978, death certificate Mortality ? N of death: 2,821 (13.8%), 57.5% due to recurrent; 337,334 py; AER: 7.36% per 1,000py ? Overall cumulative mortality (CM), 18.1% at 30 years; CM from non-recurrence, non-external causes, 3% at 20 years; 7% at 30 years, half due to SMN ? SMR: Overall, 8.4; death due to SMN, 15.2; death due to cardiac, 7.0; pulmonary, 8.8; other medical, 2.6. ?RT (RR,2.9), alkylating(RR,2.2), and epipodophyllotoxins (RR,2.3) increase the risk of death due to subsequent malignancy ?Cardiac RT (RR,3.3) and high dose of anthracycline (RR,3.1) are associated with late cardiac death ? Rad (Y/N), no dosage; no surgery info; Chemo, grouped by type of agents, dose calculated ? participants without death certificate, such as Canadian residence, exclude from this study, but included in other studies ? potential misclassification in COD Friedman DL [27] ?N=14,359 ?5-yr survivors ?Dx age: <21 yr  1970-1986 1)-2) 3) SMN info initially obtained and ascertained from self-or proxy report in questionnaires and/or death certificate, confirmed by pathology report or medical record.  SN ? N of SN survivors, 1,402; N of SNs, 2,703; mean age of follow-up, 8.3years;  ? Cumulative incidence, 20.5% at 30 years for all SNs; 7.9% for SN, excluding NMSC ? Overall SIR=6.0, AER=2.6 per 1000years; SIR for HL=8.7, SIR for Ewing sarcoma=8.5 ? Female sex, older age at dx, earlier treatment era, HL, and RT increased the risk of SMN ? Behaviors risk factors of SN were not included ? SN info initially obtained from self-report?potential underestimate of the SMN risk ? SIR calculation excluded the NMSC and non-malignant meningiomas, but includes both SMNs and SNs 37  Table 2.1 Cont?d Reference Study population Date of Dx Data collection Outcome  Relevant Findings Comments Oeffinger KC [16], Driiler L [40] ? N=10,397, sibling=3034 ?5-yr survivors ?Dx age: <21 yr  1970-1986 1)-2) 3) a severity score (grade 1-4, ranging from mild to life-threatening or disabling) assigned to each condition 4) 14.6% lost to follow-up; among the remaining, 81.2% completed baseline questionnaire 5) sibling participation (eligible N=4782, 80.4% participate) 6) self-report questionnaire on demographic characteristics, health practices, medical conditions Chronic Health Conditions ? N of survivor:10,397, sibling, 3034.  ? 62.3% of survivors reported at least one chronic condition; 27.5% had a severe or life-threatening condition (grade 3 or 4).  ? The cumulative incidence of a chronic health condition reached 73.4% (95% CI, 69.0 - 77.9) at 30 years after the cancer diagnosis, 42.4% (95% CI, 33.7 to 51.2) for severe, disabling, or life-threatening conditions or death due to a chronic condition.  ? Compared with sibling, the risk increased 3 times for a chronic condition, and 8 times for a severe of life-threatening condition  ? self-reported questionnaire method -- report bias ? info from patient-based questionnaire cannot separate conditions attributable to treatment from disease related Kurt BA [67] ? N=10366 ? 5-yr survivors ?Dx age: <21 yr  1970-1986 1)-2) 3) self-reported hospitalization info, including date, reason, associated procedures/surgery 4)socio-demographic variables were collected from questionnaire 5)expected hospitalization rate were obtained for the years of 1992-2005 from NHDS summaries Hospitalization rate ? Avg age=28.6 (sd: 7.7, range: 13-51); avg survival time=20.9 (sd: 4.6, range: 13-32); ? SIR=1.6 ? Female, an existing chronic condition, SMN, and RT increased the risk of hospitalization ? participation rate 80% in 2000, 71% in 2005 ? expected hospitalization rate was adjusted for calendar year only (no adjust in sex and age), due to the limitation in reported cause-specific NHDS hospitalization rates ? Univariate analysis in GLR           38  Table 2.1 Cont?d Reference Study population Date of Dx Data collection Outcome  Relevant Findings Comments Nordic childhood cancer (NCC) cohort --5 Nordic countries-- Sweden, Denmark, Finland, Norway, Iceland Garwicz S [19] ?N=21,984  ?5-yr survivors ?Dx age: 0-19 years 1960-1999 1) country-wide population-based cancer registries 2) Information of cause of death: from country-wide cancer registers in the five Nordic countries 3) COD: 1) death due to first cancer; 2) death due to SMN; 3) Non-cancer death Mortality ? N of death, 2,324 ? overall SMR, 8.3; AER, 6.2 per 1,000py (SMR, CCSS: 8.4, BCCSS: 10.7, SEER: 8.9) ? Cumulative mortality, 9.7% at 20 years; 14.1% at 30 years after diagnosis.  ? pattern of COD varied by primary cancer and time of survival (shorter follow-up , COD primary cancer; longer follow-up, COD SMN and non-cancer causes) ? COD was determined by the most probable cause of death relevant from the clinical point of view by assessing all recorded causes of death (both underlying and contributing), done by oncologists ? Lack of treatment info Olsen JH [78] ?N= 47,[25, 26]697 ?not 5-yr survivors ?Dx age: 0-19 yr 1943-2005 1)Country-wide population-based cancer registries 2)SMN: incidence and SMN information was determined based on data from national registries of the Nordic countries SMN ? N of SMN, 1180 cases; 28% brain tumor ? SIR=3.3;  ? cumulative incidence for diagnosed in 1943- 1959, 18%, 34%, and 48% by the age of 50, 60, 70, and 80, excluding non-melanoma skin cancer ? cumulative incidence for diagnosed in 1975- 2005, 13.3% by the age of 50, excluding non-melanoma skin cancer ? large proportion of patients dx before 1970 ? 28% SMNs were brain tumor, possibly including asymptomatic cases of meningiomas, leading to overestimated the risk ? not 5-year survivors ? lack of well-defined rules of coding SMN in the first two- three decades among these cancer registries, account for 9% of SMN cases dx before 1980. ? Lack of treatment info           39  Table 2.1 Cont?d Reference Study population Date of Dx Data collection Outcome  Relevant Findings Comments Netherlands cohort (Childhood cancer registry of EKZ/AMC) Cardous-Ubbink MC [20] ?N=1378  ?5-yr survivors ?Dx age: 0-18 years  1966-1996 (be treated)  1)Childhood cancer registry of EKZ/AMC-- a single hospital in the Netherlands 2)baseline and follow-up data abstracted from medical records 3)ppl lost to follow-up: a) mailed questionnaire to GP and physicians in other hospital ; b) requested info to municipal registries   Mortality ?N of death=120, SMR=17.2, AER=7 per 1000py; SMR=17.0 for SMN, SMR=2.05 for other causes, including 5.64 for cardio diseases? cumulative risk at 25-yr: 11%, at 30-years: 13% ? With combined modality + >=1 recurrence, SMR=92.3, AER=37.0 per 1000 py ? >=20 years of follow-up, SMR=4-5, AER of other COD vs. primary Ca.=2.3/0.3 per 1,000 py ? >=25 years of survival, SMR of other COD=5.4 ?all survivors were treated in the same institute   ? Treatment modality Yes/no, no dosage info ? Couldn?t analyze subgroups due to the small number ? Small number of Long-term survivors with intense treatment ? 96.9% medical follow-up Cardous-Ubbink MC [25] ?N=1368 ?5-yr survivors ?Dx age: 0-18 years  1966-1996 (be treated)  1) - 3);  4) The Childhood Cancer and PLEK registry of EKZ/AMC provided SMN data SMN ?N of SMN=62; including meningiomas,  SIR=11.2, AER=3.2 per 1000py; excl. meningiomas, SIR=9.45, AER=2.51; SIR=12.1 for solid SMN, AER=2.74;  ? Cumulative risk at 20-years: 4.4%; at 30-years: 11.1% ? SIR=40 for meningiomas, SIR=9 for basal cell carcinomas ? With RT, HR=1.84 for all SMN, HR=3.39 for solid SMN ?3rd ca was retain in the analysis of SMN, SMN within first 5-yr excluded ?SMN includes meningiomas and basal cell  ?SMN size is relatively small, not able to analyze RT field and doses or type of chemo agents ? 92.0% medical follow-up Geenen MM [69] ?N=1362 ?5-yr survivors ?Dx age: 0-18 years  1966-1996 (be treated) 1) - 3);  4) pt visited to hospital to complete the medical status Adverse Health outcome ?74.5% (N=1015) of survivors had 1+ adverse events, 24.6% (N=316) had 5+ adverse events. ?35.9% (N=473) of survivors had at least 1 severe or life-threatening or disabling adverse  ?RT was associated with the highest risk of developing a high or severe burden of disease, while chemotherapy only was associated with significantly lower risk than surgery.  ?A high or severe burden of adverse events was most often observed in survivors of bone tumors (64%) and least often in survivors of leukemia or Wilms tumor (12% each). ?health outcome: did not compare the prevalence of adverse events in survivors with that in a healthy pop'n  ? 1.5% lost to follow-up, 94.3% evaluated by a physician, including 79% evaluated in the institution  ? Homogenous group, limit the generalizability of the findings to other pop?n. ? a single  institution 40  Table 2.1 Cont?d Reference Study population Date of Dx Data collection Outcome  Relevant Findings Comments French-British Study de Vathaire F [26] ?N=4400 ?3-yr survivors ?Dx age: 0-17 years  Before 1986 (1942-1986) 1) Treated in 8 centers in France and the UK 2) Baseline and follow-up data, clinical data of SMN, RT data and chemo data were recorded from clinical records. 3) No primary and subsequent leukemia, no retinoblastoma for the UK, no non-melanoma skin cancer SMN ? N of solid SMN=113, mean follow-up=15 years, 88% completed follow-up (12% lost follow-up) ? Cumulative incidence of solid SMN was 4.9% at 25 years, 7.7% at 30 years ? SIR=9.2 for solid SMN, AER=1.88 per 1,000 py ? The risk of thyroid ca increased with the dose of radiation, but not with chemo ? RR of SMN decreased with follow-up time and attained age, after RT alone and after RT + chemo ? No primary and subsequent leukemia, no retinoblastoma for the UK centers, no non-melanoma skin cancer ? No multivariate analysis ? Mean age at the end of follow-up was young (21years)?SMNs in childhood and young adult period ? Expected rate from Danish Cancer Registry  Reference Study population Date of Dx Data collection Outcome  Relevant Findings Comments *Slovenia Study Jazbec J [79] ?N=1577 ?not 5-yr survivors ?Dx age: <16 years  1961-2000 1)population-based study cohort2)baseline, SMN and follow-up data were collected from national cancer registry, medical records 3)SN dx and COD were reviewed  SMN ? N of SMN=48; mean follow-up time: 13.2years ? Acute leukemia, CNS and thyroid were most common SMN ? Cumulative incidence of SMN: 7.4% at 20 years, 12.6% at 25 years ? SIR=8.71 ? Overall survival rate after SMN was 65% at 10 years after dx           ? Treatment exposure: no dosage ? Non-melanoma skin ca and meningiomas were included as SMN ? Univariate analysis only ? SMN within the first 5-yr survival was included 41  Table 2.1 Cont?d Reference Study population Date of Dx Data collection Outcome  Relevant Findings Comments North of England Study Hammal DM [28] ? N=4072 (0-24 years); N=2066 (15-24 years) ?6-month survivors ?Dx age: <25 years  1968-1999 1) survivors were identified from Northern Region Young Person?s Malignant Disease Registry 2) EOF: June 30, 2011 SMN ? Total N of SMN=68, including 8 NMSC, 4 meningioma; median follow-up, 6.5 years. ? Young adult(15-24years): N of SMN, 39; median age at 1stca dx: 20.9years, 2nd ca dx: 21.3years ? incidence rate of SMN: overall, 1.7 per 1000py; young adult group, 1.9 per 1000py ? SIR: overall, 4.4; young adult: 3.1 ? Young adult SIR: HL, 5.8; Brain: 4.3; sarcomas: 2.2; Leuk/lymph:2.5; carcinomas: 1.8; Other, 1.7. only SIRs of HL and Brain were significant  ? Basal cell carcinomas and meningiomas were included as SMN ? Univariate analysis ? Only 30 cases lost to follow up among 5077 subjects ? No treatment info  Reference Study population Date of Dx Data collection Outcome  Relevant Findings Comments Roswell Park Cancer Institute's Long-Term Follow-Up Project (LTFUP) Lawless SCW [21] ? N=565  ? 15+ year survivors ? dx age: 0-19 1960-1989 1)patients diagnosed in one institute 2)COD determined by medical records, using underlying COD 3) institute database provided info of baseline data and follow-up data, including primary dx, SMN dx, and treatment Mortality ? SMN was the leading COD among male and female long-term survivors (15/38 deaths, 39%) ? Excess overall mortality was noted among both males (SMR = 284) and females (SMR = 371). ? Mortality risk increased in both gender, for COD in primary cancer, SMN, and cardiac deaths among males, in most age groups.  ? Among long-term survivors without relapse, overall mortality in both genders did not differ significantly from the general population.      ? One institute data ? long term survivors (15+ years) have increased risk of mortality ? cancer recurrence and SMN are important factors of prognosis. ? Numbers were small ? Univariate analysis only 42  Table 2.1 Cont?d Reference Study population Date of Dx Data collection Outcome  Relevant Findings Comments Childhood, adolescent, young adult cancer survivors program(CAYACS), Canada MacArthur AC [22] ? N=2354 ?5-yr survivors ?Dx age: <20 years ? EOF: Dec 31, 2000 1970-1995 1) Survivors identified from BC Cancer Registry (BCCR) 2) Baseline data obtained from BCCR 3) COD determined from Vital Status, reviewed and validated Mortality ? N of death=181(24,491py), including 139 cancer-related deaths; AER=6.6 per 1000py; mean age at dx: 9.4; mean time of follow-up: 15 years ? SMR=9.1, higher from pts with recurrence, ALL or CNS.  ? Being male and dx before 1980 were associated with increased AER ? Mortality risk increased due to COD of circulatory and respiratory diseases.    ? COD: 5% of fatalities coded as non-cancer deaths were attributable to primary cancer ? Data linkage method to avoid loss to follow-up ? Small number of certain groups  MacArthur AC [81] ? N=2322 ?5-yr survivors ?Dx age: <20 years ? EOF: Dec 31, 2000  1970-1995 1)-3) 4) SMN, from BCCR, excluding NMSC and benign tumors SMN ? N of SMN, 55 (26701 pys); AER, 1.7 per 1000yr ? Cumulative incidence, 3% at 20 years, 5.1% at 25 years  ? SIR=5.0 ? Being female and dx age <10years old were associated with increased risks of SMN                ?  43  Table 2.1 Cont?d Reference Study population Date of Dx Data collection Outcome  Relevant Findings Comments ? Childhood, adolescent, young adult cancer survivors program(CAYACS), Canada Lorenzi M [65] ? N of survivor=1374 ? N of comparison=13740 ?5-yr survivors ?Dx age: <20 years ? EOF: Dec 31, 2000 (from 1986)  1981-1995 1)-3) 4) late morbidity definition?corresponds to Grade 3, 4, 5 in CTCA 5)  Hospital-related morbidity ? 41% of survivors vs. 17% of the population sample had at least one type of hospitalization-related late morbidity ? adjusted RR 4.1 (95% CI 3.7-4.5) ? highest risk: survivors of leukemia (RR 4.8, 95% CI 4.0-5.8), CNS tumors (RR 4.8, 95% CI 4.0-5.8), bone and soft tissue sarcomas (RR 4.9, 95% CI 3.8-6.2), and kidney cancer (RR 4.9, 95% CI 3.4-7.0) ? Highest adjusted RR: neoplasms (including secondary cancers) (RR 21.7, 95% CI 16.3-28.7). ? Morbidity was elevated for all treatment, and highest for the combination for all three (RR 7.1, 95% CI 5.5-9.0).  ? Over time, morbidity for late effects other than neoplasms became more prevalent. ? Shorter follow-up time  than CCSS and Dutch study ? Included death cases in hospitalization ? Exclude non-hospitalization condition ? No speciality info ? 13.4% not linked: out be BC (4% of survivors vs. 6% of comparison), inaccurate data in linkage, non-registration  in healthcare,  Bradley N[66] ? N of survivor=1157 ? N of comparison=11570 ?5-yr survivors ?Dx age: <20 years ? EOF: Mar 31, 2000 (from Jan 1998)  1970-1992 1)-3) 4) hospitalization data included hospital discharge for every visit, not emergency/ outpatient visit 5) gender, attained age determined by baseline data 6) SES, regional health authority, and rural/urban area determined by postal code at Jan 1, 1998   ? 240 (21%) of survivors vs. 614 (5.3%) of the population sample were admitted to hospital at least once  ? adjusted OR=4.36 (95% CI 3.68-5.16) ? Hospitalized survivors had a higher average number of admissions (2.0 versus 1.5 admissions, respectively) and longer mean DIH (10.9 versus 7.8d, respectively) than controls.  ? Female gender and older age increased the risk of hospitalization, as did the presence of a relapse or second cancer by 5 years post-diagnosis. ? 64% linkage rate ? Out-immigration rate: 2% of the BC population of any age leaves BC  ? Preventable hospitalization issue   44  Chapter 3: Late mortality among young adult cancer 5-year survivors 3.1 Introduction Cancer incidence between 15 and 29 years of age (the AYA group) is 2.7 times more common than for those from 0 to 14 years, and accounts for 2% of all invasive cancer [3]. In the past four decades, the 5-year survival rate for cancer survivors has been significantly improved, primarily due to the improvement and advances in cancer diagnostic and treatment. However, the survival improvement for the AYA group has lagged behind that for childhood cancers. The improvement in 5-year survival diagnosed from the mid 1970s to the late 1990s among AYA cases was lower than the rates of improvement in both younger and older groups. The significant improvement in survival rate among childhood cancer has not been equally observed among AYA cancer patients.   Previous studies have shown that survivors of childhood and adolescent cancer have increased mortality risks, compared with the general population [17, 20, 22, 76, 83, 106]. Primary cancer, including both recurrence and progression of the original cancer, was identified as the leading cause of death among childhood cancer survivors, representing 61% to 75% of the deaths, depending on the era of treatment and the type of cancer [18, 23, 107]. It has also been noted that the pattern of late mortality has changed over time since diagnosis. While mortality risk from recurrence decreases rapidly after 15 years from diagnosis, the risk of death for SMN and non-cancer-related diseases, such as circulatory disease, increases [18, 107]. However, little has been reported specifically about mortality for young adult cancer. For example, although the literature regarding mortality among childhood cancer survivors has found that female sex is associated with increased SMR for second cancer and other disease-related causes[17], we do not know if this also is true for survivors who were diagnosed as young adults.  The Childhood, Adolescent and Young Adult Cancer Survivors (CAYACS) Research Program utilizes record linkage methodology and data from population-based registries, administrative databases and medical charts, to examine issues among survivors of cancer diagnosed under age 25 years in British Columbia (BC), Canada[80]. Since a review of the available literature indicated that there was little published information on long-term mortality risks for survivors of young adult cancer, we conducted a population-based retrospective study to assess the long-term  45  risks of overall and cause-specific mortality among survivors of young adult cancer in the CAYACS cohort, as compared to these risks in the British Columbia population. We also evaluated the demographic and clinical factors affecting these risks.  3.2 Methods 3.2.1 Study population CAYACS is a population-based retrospective survivor cohort study, established for survivors of cancer diagnosed under 25 years of age in the province of British Columbia (BC), Canada. The study design and methodology has been described in detail elsewhere [80]. In brief, the cancer survivor cohort was identified from the BC Cancer Registry (BCCR). The Registry receives notice of all cancer diagnoses occurring in Canada to BC residents, via multiple sources of ascertainment [80]. The ascertainment of cancer incident cases from BCCR is estimated to be 95% or better.   In this study, all participants were survivors with a first diagnosis of young adult cancer (age 20-24 years), surviving at least five years from diagnosis, with a diagnosis included in the Adolescent and Young Adult Cancer Classification (AYA) [108], excluding non-melanoma skin cancer. All survivors were diagnosed between January 1, 1970 and December 31, 1995, with follow-up to the end of 2007 through annual provincial health insurance program records, or until emigration out of the province, or death.   3.2.2 Data collection Death information Death records of all deaths in Canada to BC residents are maintained at the Ministry of Health Vital Statistics Agency (VSA), and routinely reported to the BC Cancer Registry. Information documenting death, including date and causes of death coded using the International Classification of Disease version 9 (ICD-9)[109], was obtained through BC Cancer Registry and was reviewed by oncologists from BC Cancer Agency. Only the primary cause of death was used in this study. Survivors without death records were considered as alive at the end of the study period.  46   Specific causes of death were combined into three major groups: 1) death due to recurrence or progression of the original cancer (ICD 140-239); 2) death due to a second original cancer (ICD 140-239); and 3) death due to a non-cancer-related cause, including disease-related causes and external causes, such as suicide and poisoning (ICD 001-139, 240-999). These three major groups were further divided into 26 minor groups, such as circulatory disease, digestive disease, and external cause of death.   The sex-, age-, cause-, and year-specific mortality rates for the BC population were obtained from the VSA.   Modifying variables The BC Cancer Registry provides baseline demographic, diagnostic, and follow-up information for young adult cancer survivors, including date of birth, gender, date of primary cancer diagnosis, and morphological and histological codes of diagnosis based on International Classification of Disease for Oncology (ICDO), 3rd edition. [110]. All the primary cancer ICDO diagnosis codes were further converted into a ten-category classification system relevant to the major diagnoses among adolescents and young adults (AYA), based on morphologic and histological diagnosis codes: leukemia, lymphoma, central nervous system tumor (CNS), bone tumor (osseous & chondromatous neoplasm), soft tissue sarcoma, germ cell tumor, melanoma, carcinoma, miscellaneous specified neoplasm, and unspecified malignant neoplasm (ref for the AYA Classification) (Appendix 1).   Treatment information on the primary cancer was directly abstracted from medical records in tertiary cancer centers in BC, including an indicator of chemotherapy, radiation therapy (RT), and surgery for original cancer treatment. As all RT facilities in the province are located in a tertiary cancer center, the patients? RT records were complete, but information regarding chemotherapy and treatment-related surgery was missing in approximately 26% of cases, since these cases were not referred to any of these centers [80].    47  3.2.3 Study analysis The overall and cause-specific mortality rates were measured using the standardized mortality ratio (SMR), absolute excess risk (AER) of death, and cumulative survival. The primary explanatory variable is the primary cancer diagnosis classified by the AYA classification system. Other potential modifiers of risk of death were examined, including gender, calendar period of diagnosis by 10-year time period (1970-1979, 1980-1989, and 1990-1995), length of follow-up as the time from diagnosis by the following categories 5-19 years and 20 years and more, and treatment modality, including having chemotherapy, radiation, and surgery.   The frequency and proportion of deaths in the cases were calculated by AYA diagnosis groups, gender, diagnosis period, and time since diagnosis. Relative mortality was measured according to the overall and cause-specific standardized mortality ratio (SMR). The SMR was defined as the ratio of observed number of deaths among young adult cancer survivors relative to the expected number of deaths in the BC population, standardized by 5-year age group, gender, and 5-year calendar period. The expected number of deaths was calculated by multiplying the number of person-years at risk in each age, gender and calendar period stratum by the corresponding mortality rate among BC general population and then summing over all strata. The absolute excess risk (AER) of the survivor group was also estimated. The AER of death was calculated as the observed number of deaths minus the expected number of death, divided by the person-years at risk. The AER was reported as per 1,000 cancer survivors per year.  Cumulative survival was estimated as a function of time since 5 years after the initial cancer diagnosis. In order to estimate the effect of demographic and disease-related factors on the risk of death, the Cox proportional hazard regression was used, adjusted for potential confounding variables.   In order to address missing chemotherapy and surgery status on cases who were not referred to any of the tertiary cancer centers, multiple imputation was utilized under the assumption of missing at random [111]. The imputation model contained information on type of original cancer (seven categories in the mortality analysis), diagnosis period (10-year group), having relapse<5years post-diagnosis (Yes/No), and having radiation (Yes/No), and created 20 datasets  48  without missing chemotherapy or surgery value. The analysis was repeated 20 times and the results were combined and summarized to generate valid statistical inferences. All calculations were performed using the statistical package SAS version 9.2 (SAS Institute Inc, Cary, North Carolina, USA).  3.2.4 Study approvals  The study received ethical approval from the University of British Columbia /BC Cancer Agency Clinical Research Ethics Board. Approval for access, use and linkage of data was obtained from the BC Cancer Registry and the BC Cancer Agency Health Records Departments. To protect confidentiality, all counts less than 5 were masked in tables and text.  3.3 Results 3.3.1 Descriptive analysis  The demographic and treatment related features of the cohort are described in Table 3.1. A total of 1248 individuals were diagnosed with cancer between 20 and 25 years old in BC from 1970 to 1995, and further reported to BC Cancer Registry. Of these, 47.2% (n=589) were males, and 31.6% (n=394) were diagnosed before 1980. The median age at the end of follow-up was 45.0 (SD ? 7.9), and the median duration of follow-up was 17.5 years with 5-year post diagnosis (SD?7.8; range 0.1-33.0), with 21711 person-year of follow-up from 5-year survival to December 31, 2007 or death, depending on which occurred at first. Common types of cancer included carcinoma (n=341, 27.3%), including cancers in the thyroid (n=117) and cervix and uterus (n=92), lymphoma (n=307, 24.6%), melanoma (n=222, 17.8%), and germ cell tumor (n=178, 14.3%). 27.4% of the survivors (n=342) received chemotherapy during their treatment, 29.6% (n=369) received radiation, and over 50% (n=628) had treatment-related surgery.   During the time of follow-up, a total of 138 (11.1% of the cohort) deaths occurred more than 5-year after diagnosis. Of these, 117 deaths (84.8%) had cause of death information. The most common types of cancer among deceased survivors were lymphoma (33.3%), carcinoma (15.2%), and central nervous system tumor (CNS) (15.2%). In comparison to survivors being alive by the end of follow-up, deceased survivors had a significantly shorter follow-up time (6.7 vs. 18.3  49  years, p<0.001), and were more likely to be male, and have been diagnosed in an earlier time period. 32.6% deceased patients received chemotherapy, 54.3% reported had radiation, and 59.4% had treatment-related surgery.  3.3.2 All causes of death Overall, mortality for 5-year young adult cancer survivors was 6 times higher than the BC general population (SMR =5.9, 95% CI, 4.9-6.9) (Table 3.2). Survivors of all original diagnoses measured had elevated mortality compared to the BC population rate. Overall, survivors with Central nervous system (CNS) (SMR=23.6; 95% CI, 15.1-35.1) and leukemia (SMR=19.4; 95% CI, 8.9-36.8) were observed having the highest SMRs. The smallest excess risks were seen in subjects originally diagnosed with germ cell tumors or carcinomas and having a longer time of follow-up. Overall, females (SMR=6.3; 95% CI, 4.8-8.3) had a higher risk of death compared to males (SMR=5.6; 95% CI, 4.5-6.9).   Among the 117 deceased cases with known cause of death, death due to original cancer was the leading cause, with 62 (53.0%) cases attributed to the recurrence or progression of the original cancer diagnosis, representing a crude rate of 2.85 per 1,000 person-years (not shown in table).An additional 23 (19.7%) deaths were attributed to a different cancer, and 32(27.4%) deaths were reported as due to non-cancer related causes, including external causes, such as accident and suicide. Cancer survivors were 5.6 times more likely to die of circulatory diseases (n=10; SMR=5.6, 95% CI, 2.68-10.29), such as ischemic heart disease, and 3.3 times more likely to die of infective and parasitic diseases (n=5; SMR=3.3, 95% CI, 1.07-7.7), but there was not extra risk in death from external causes (n=7; SMR=0.62, 95% CI, 0.25-1.27).  Of the 98 deaths followed less than 20 years (87 cases with known causes of death), the leading cause of death was the original cancer (56/87 deaths, 64.4%, not shown in table), followed by non-cancer death (19/87 deaths, 21.8%). For survivors who had survived at least 20 years (30 cases with known cause of death), deaths due to a cause other than cancer (13 of 30 deaths, 43.3%, not shown in table), and SMNs (10 of 30 deaths, 33.3%), were the two most common causes of death.   50  Male survivors showed lower risk of death due to original cancer, but higher risk of death due to SMN and non-cancer causes than females (Table 3.3). However, females had a lower SMR for both death due to SMN and non-cancer related categories. Male survivors also showed a higher risk of cumulative mortality over time than females (Figure 3.1).                                                                                                                                                                                                       3.3.3 Factors for the risks of death Table 3.4 shows the unadjusted and adjusted associations between characteristics of survivors of young adult cancer, and the likelihood of death during the follow-up period. After adjusting for other factors, survivors of CNS had a statistically significant increased risk of dying (Hazard Ratio [HR] adj = 3.4, 95%CI, 2.1-5.7) compared to lymphoma survivors, whereas germ cell tumour and carcinoma survivors had decreased risks (HR adj = 0.4, 95%CI, 0.2-0.7; HR adj = 0.5, 95%CI, 0.3-0.8, respectively). Male survivors were 1.7 times more likely to die than their female counterparts (HRadj =1.7, 95%CI, 1.2-2.4).  Survivors who survived at least 5 years but who had relapsed in the first 5 years following their diagnosis had an almost 3-fold increased risk of death (HRadj =2.9, 95%CI, 1.9-4.5), compared to those who had not. RT was also associated with a 2-fold increased risk (HRadj =2.0, 95%CI, 1.3-3.1), compared to those who had not received RT.   3.4 Discussion  This study identified elevated risks of mortality among survivors of young adult cancer compared with the general population. As well, this study provides information on the effects of demographic and clinical factors on  risk of mortality for this group. Exposure to RT was directly associated with an increased risk of death for survivors. Recurrence of primary cancer and SMN were both highly associated with increased risk of mortality. These results underline the importance of applying effective surveillance and interventions for young adult cancer survivors, especially for the subgroups with identified risk factors.  Our analysis indicated a six-fold increase in mortality among 5-year survivors of young adult cancer. This estimate is lower than the SMRs identified from studies in childhood cancer survivors, which ranged from 7.5 to 10.8[18, 22, 76, 83, 107, 112]. These studies all focused on  51  survivors of younger age than our study, with diagnoses 0-15 years old in the British study[107], 0-20 years old in the Nordic study[83], and 0-21 years old in the Childhood Cancer Survivor Study (CCSS) study[18, 76]. This difference in SMR is likely caused by the lower proportion of some types of cancer diagnosed in young adulthood compared to childhood, such as CNS tumours, which have a poor prognosis when diagnosed at younger ages [18]. It is also possible that cancer therapy delivered at an early age when the young person is experiencing a singular amount of growth and development may have a profound impact later on affecting late mortality.  Studies of childhood cancer found that recurrence or progression of original cancer was the leading cause of late death for survivors [18, 22, 76, 83, 107]. Several studies raised a concern that the cause of death was potentially related to the length of follow-up [21, 83, 107]. A study among 15+ year survivors of childhood and adolescent cancers reported second malignancies to be the leading cause of death, [21]. Our study identified the original cancer to be the leading cause of death for survivors with less than 20 years follow-up, and non-cancer related death to be the leading cause of death amongst survivors with more than 20 years of follow-up, followed by second malignancies, representing 43.3% and 33.3% of deaths respectively. Taken together, these results suggest that the longer time the patient has survived, the less likely the survivor is to die from the original cancer.  Gender-specific differences in absolute risk of death and standardized mortality rate have been observed and discussed in other cancer survivor studies [18, 22, 76]. Within our study, males survivors had a lower overall standardized risk of mortality than females by age, gender, and year of diangosis, which is similar to previous studies conducted in pediatric cancer survivors [18, 22, 76, 107, 112]. However, after adjustment by demographic and treatment-related factors, the hazard ratio of death for males is higher than for females. These differences may primarily arise due to the high absolute risk of mortality and the small numbers, especially for the cases with non-cancer-related cause of death (N=24 for male vs. N=8 for female).  In this study, cause of death from death certificates was initially used for the analysis. Death information for the survivors, including date and causes of death coding, was obtained from death certificates through the Ministry of Health Vital Statistics Agency (VSA), and subsequently   52  reviewed by oncologists. If the VSA cause of death was found not to match the oncologist-determined cause of the patient?s death, a clinically-determined cause of death code was added to the record. Otherwise, the VSA cause of death code was kept. A comparison of the causes of death according to the death certificate with the oncologist-validated causes of death in the same cohort has showed that 5% of fatalities coded as non-cancer deaths on the death registration were attributed to the first malignancy by the oncologist [22]. If we assume that the sensitivity and specificity are stable across administrative datasets at least at the levels of diseases, by using the modified clinical codes for cause of death for the survivor group, differential misclassification will be introduced into the study, which might lead to an underestimation of SMR for some of the cancer-related causes of death. On the other hand, if the death registration codes were used, nondifferential misclassification would occur in the study; the presence of nondifferential disease misclassification does not impact the exposure-disease association estimation [113].   Although the availability of administrative datasets and tumour treatment information on the majority of the subjects and the long follow-up time with complete treatment information were strengths of the study, our study also shares the limitations encountered by other researchers in this field. Because of the relatively small numbers of deceased  cases (N=138 deceased cases) observed, we did not have sufficient study power to assess treatment-related dose-response for mortality risk. Another issue was that, while vital status information was collected for death occurring out of BC, it was not available for individuals who were not in Canada at the time of death.   In conclusion, the data from our long-term survivors? follow-up cohort indicate that young adult cancer survivors experience an elevated risk of mortality compared with the general population. Although the increased risk of mortality in young adult cancer survivors is generally observed across all demographic and disease-related groups, the risk is highest in survivor groups characterized by having a relapse or second cancer within 5 years of diagnosis, having a primary diagnosis of CNS, and having radiation therapy. The increased risk in cause-specific SMR suggests that future efforts should concentrate on developing better methods of monitoring, treating and preventing the recurrent or second malignancies, as well as other late morbidities, especially diseases related to blood and circulatory systems.  53  Tables Table 3.1 Demographic and disease-related characteristics of young adult cancer survivors   Entire Cohort N(%) Alive N(%) Dead N(%) Overall 1248 1110(88.9) 138(11.1) Age at end of follow-up median in years, standard deviation (SD) 45.0(7.9) 45.7(7.2) 34.3(8.5) Follow-up time from 5-yr survival (median in years, SD) 17.5(7.8) 18.3(7.1) 6.7(8.5) Sex                Male 589(47.2) 505(45.5) 84(60.9)          Female 659(52.8) 605(54.4) 54(39.1) Type of Primary Cancera               Lymphoma 307(24.6) 261(23.5) 46(33.3)         Central nervous system tumor 59(4.7) 35(3.2) 24(17.4)         Soft tissue sarcomas 66(5.3) 55(5) 11(8)         Germ Cell Tumor 178(14.3) 169(15.2) 9(6.5)         Melanoma 222(17.8) 206(18.6) 16(11.6)         Carcinoma (except of skin) 341(27.3) 320(28.8) 21(15.2)         Others* 75(6.0) 64(5.7) 11(7.9) Birth year (10-yr) 1945-1954 271(21.7) 221(19.9) 50(36.2) 1955-1964 528(42.3) 467(42.1) 61(44.2) 1965-1975 449(36) 422(38) 27(19.6) Diagnosis period (10-yr)       1970-1979 394(31.6) 323(29.1) 71(51.4) 1980-1989 520(41.7) 477(43) 43(31.2) 1990-1995 334(26.8) 310(27.9) 24(17.4) Attained Age (by 10-year)       25-39 366(29.3) 276(24.9) 90(65.2) 40-49 522(41.8) 488(44) 34(24.6) 50+ 360(28.8) 346(31.2) 14(10.1) Follow-up time (years) 5-19 years 521(41.7) 423(38.1) 98(71) 20+ years 727(58.3) 687(61.9) 40(29)  54  Table 3.1 cont?d   Entire Cohort N(%) Alive N(%) Dead N(%) Relapse < 5 years post-diagnosis       Yes 82(6.6) 55(5) 27(19.6) No 1166(93.4) 1055(95) 111(80.4) SMN >= 5 years post-diagnosis    Yes 62(5.0) 39(3.5) 23(16.7) No 1186(95.0) 1071(96.5) 115(83.3) Having Chemotherapy       Yes 342(27.4) 297(26.8) 45(32.6) No 552(44.2) 473(42.6) 79(57.2) Missing 354(28.4) 340(30.6) 14(10.1) Having Radiation Yes 369(29.6) 294(26.5) 75(54.3) No 879(70.4) 816(73.5) 63(45.7) Having Surgery       Yes 628(50.3) 546(49.2) 82(59.4) No 270(21.6) 228(20.5) 42(30.4) Missing 350(28) 336(30.3) 14(10.1)  *Other types of original cancers include leukemia, lone tumor, and miscellaneous-specified tumor.   55   Table 3.2. Observed and expected deaths, standard mortality ratios, and absolute excess risks for death f   Entire Cohort N(%) Alive N(%) Obsaf Expb SMRc 95% CId AERe Overall 1248 1110(88.9) 138 23.5 5.9 4.9-6.9* 5.3 Sex         Male 589(47.2) 505(45.5) 84 15 5.6 4.5-6.9* 7.1         Female 659(52.8) 605(54.4) 54 8.5 6.3 4.8-8.3* 3.8 Type of Primary Cancer                      Lymphoma 307(24.6) 261(23.5) 46 5.7 8 5.9-10.7* 7.8        Central nervous system tumor 59(4.7) 35(3.2) 24 1 23.6 15.1-35.1* 27.8        Soft tissue sarcomas 66(5.3) 55(5) 11 1.2 8.9 4.4-15.8* 8.9        Germ Cell Tumor 178(14.3) 169(15.2) 9 4.3 2.1 1.0-4.0* 1.6        Melanoma 222(17.8) 206(18.6) 16 3.5 4.6 2.6-7.4* 3.4        Carcinoma (except of skin) 341(27.3) 320(28.8) 21 6.3 3.3 2.1-5.1* 2.2        Othersg 39(3.1) 37(3.3) 11 1.4 7.7 3.8-13.8* 7.8 Follow-up time (years)                      5-19 years 521(41.7) 422(38) 98 3.5 27.8 22.5-33.8* 19.4            Death due to second cancer --- --- 12 0.3 35.9 18.6-62.8* 2.4            Noncancer deaths --- --- 19 3.2 5.9 3.6-9.3* 3.2        20+ years 727(58.3) 688(62) 40 20 2 1.4-2.7* 1.2            Death due to second cancer --- --- 11 4.1 2.7 1.4-4.8* 0.4            Noncancer deaths --- --- 13 15.9 0.8 0.4-1.4 -0.2 Diagnosis period               1970-1979 394(31.6) 323(29.1) 71 12.7 5.6 4.4-7* 6.0  1980-1989 520(41.7) 477(43) 43 8.7 4.9 3.6-6.7* 3.8 1990-1995 334(26.8) 310(27.9) 24 2.1 11.4 7.3-17* 7.0   56  Table 3.2. Cont?d   Entire Cohort N(%) Alive N(%) Obsaf Expb SMRc 95% CId AERe Having Chemotherapy               Yes 342(27.4) 297(26.8) 45 5.6 8 5.8-10.7* 7.6 No 552(44.2) 473(42.6) 79 10.8 7.3 5.8-9.2* 7.1 Missing 354(28.4) 340(30.6) 14 7.2 2 1.1-3.3* 1.0  Having Radiation               Yes 369(29.6) 294(26.5) 75 7.9 9.4 7.4-11.8* 10.2 No 879(70.4) 816(73.5) 63 15.6 4 3.1-5.2* 3.1 Having Surgery               Yes 628(50.3) 546(49.2) 82 11.6 7.1 5.6-8.8* 6.7 No 270(21.6) 228(20.5) 42 4.8 8.7 6.3-11.8* 8.5 Missing 350(28) 336(30.3) 14 7.1 2 1.1-3.3* 1.0  Major Causes of death (COD)     Death due to cancer --- --- 85 4.4 19.3 15.4-23.9* 3.7         Death due to second cancer --- --- 23 4.4 5.2 3.3-7.8* 0.9     Noncancer deaths --- --- 32 19.1 1.7 1.1-2.4* 0.6     Missing --- --- 21 N/A N/A N/A N/A  *p<0.05 a. Obs: Number of observed deaths b. Exp: Number of expected deaths c. SMR: Standardized mortality ratio d. 95%CI: 95% confidence interval e. AER: Absolute excess risk, per 1,000 years at risk f. As per Ministry of Health requirements, any numbers less than five cannot be specified.   g. Other types of original cancers include leukemia, lone tumor, and miscellaneous-specified tumor.  57   Table 3.3. Observed and expected deaths, standard mortality ratios, and absolute excess risks for death by sex f   Entire Cohort N(%) Male (N=589)  Female (N=659) Obsaf SMRc 95% CId AERe  Obsaf SMRc 95% CId AEReOverall 1248 84 5.6 4.5-6.9* 7.1 54  6.3 4.8-8.3* 3.8 Type of Primary Cancer                            Lymphoma 307(24.6) 27 6.7 4.4-9.7* 8.7 19  11.1 6.7-17.4* 6.9         Central nervous system tumor 59(4.7) 18 23.0 13.1-36.3* 35.3 6  7.3 0.2-40.9 4         Soft tissue sarcomas 66(5.3) 9 10.7 4.9-20.2* 13.9 <5 0 N/A N/A         Germ Cell Tumor 178(14.3) 9 2.2 1.0-4.2* 1.8 0  0 N/A N/A         Melanoma 222(17.8) 7 3.5 1.4-7.3* 3.8 9  5.9 2.7-11.2* 3.2         Carcinoma (except of skin) 341(27.3) 8 3.5 1.5-6.9* 4.2 13  3.2 1.7-5.5* 1.7         Othersg 75(6.0) 6 5.7 2.1-12.4* 7.4  5  9.2 4.5-16.1* 4.8 Follow-up time (years)           5-19 years 521(41.7) 62 25.3 19.4-32.4* 26.1 36  33.4 23.4-46.2* 13.5             Death due to second cancer --- <5 24.4 6.6-62.3* 1.7 8  47.1 20.3-92.9* 3             Noncancer deaths --- 14  6.1 3.3-10.3* 5.1 5  5.5 1.8-12.9* 1.6         20+ years 727(58.3) 22  1.8 1.1-2.7* 1.3 18  2.4 1.4-3.8* 1.1             Death due to second cancer --- 6  4.0 1.5-8.7* 0.6 5  2 0.6-4.6 0.3             Noncancer deaths --- 10  0.9 0.4-1.7 -0.1 <5 0.6 0.1-1.8 -0.2 Diagnosis period                    1970-1979 394(31.6) 47 6.4 4.7-8.4* 10.3 24  4.5 2.9-6.7* 3.2 1980-1989 520(41.7) 21 3.4 2.1-5.2* 3.3 22  8.7 5.5-13.2* 4.4 1990-1995 334(26.8) 16 11.0 6.3-17.9* 9.9  8  12.3 5.3-24.3* 4.4 Having Chemotherapy   Yes 342(27.4) 28 6.4 4.2-9.2* 7.7 17  13.9 8.1-22.2* 7.5 No 552(44.2) 47 6.7 4.9-8.9* 9 32  8.6 5.9-12.2* 5.5 Unknown 354(28.4) 9 2.5 1.2-4.8* 2.4 5  1.4 0.5-3.3 0.3  58  Table 3.3 Cont?d  Entire Cohort N(%) Male (N=589) Female (N=659) Obsaf SMRc 95% CId AERe Obsaf SMRc 95% CId AEReHaving Radiation                    Yes 369(29.6) 46 8.0 5.9-10.7* 11.4 29  13.1 8.8-18.9* 8.9 No 879(70.4) 38 4.1 2.9-5.6* 4.6  25  4 2.6-5.9* 2.1 Major Causes of death (COD)                        Death due to cancer --- 46  27.6 20.2-36.9* 4.5 39  14.3 10.1-19.5* 3         Death due to second cancer --- 10  6.0 2.9-11.1* 0.9 13  4.8 2.5-8.1* 0.9     Noncancer deaths --- 24  1.8 1.2-2.7* 1.1 8  1.4 0.6-2.7 0.2     Missing --- 14  N/A N/A N/A  7  N/A N/A N/A  *p<0.05 a. Obs: Number of observed deaths b. Exp: Number of expected deaths c. SMR: Standardized mortality ratio d. 95%CI: 95% confidence interval e. AER: Absolute excess risk, per 1,000 years at risk f. As per Ministry of Health requirements, any numbers less than five cannot be specified.   g. Other AYA major groups include leukemia, bone tumor, and miscellaneous specified tumor.      59  Table 3.4. Hazard ratios for all cause mortality   All Causes of Death (N=138) Male  Female   Univariate  HR (95%CI)Multivariate HR (95%CI)Multivariate HR (95%CI)  Multivariate HR (95%CI)Sexa         Female 1 1 --- ---         Male 1.9(1.4-2.7) 1.7(1.2-2.4) --- --- Type of Primary Cancerb         Lymphoma 1 1 1 1         Central nervous system tumors 3.1(1.9-5.2) 3.4(2.1-5.7) 4.5(2.5-8.3)  N/A     Soft tissue sarcomas 1.1(0.6-2.2) 1.2(0.6-2.3) 1.8(0.8-3.8) N/A Germ Cell Tumor 0.3(0.2-0.7) 0.4(0.2-0.7) 0.5(0.2-0.9) N/A         Melanoma 0.5(0.3-0.9) 0.7(0.4-1.2) 0.7(0.3-1.7) 0.6(0.3-1.3)         Carcinoma (except of skin) 0.4(0.2-0.6) 0.5(0.3-0.8) 0.7(0.3-1.5)  0.4(0.2-0.8)*         Others 1(0.5-1.9) 1.1(0.6-2.1) 0.9(0.4-2.3) 1.0(0.5-2.1) Diagnosis periodc 1970-1979 1 1 1 1 1980-1989 0.7(0.4-1.0) 0.6(0.4-0.9) 0.4(0.2-0.7) 1.2(0.6-2.4) 1990-1995 0.9(0.5-1.5) 0.9(0.5-1.4) 0.8(0.4-1.4) 1.1(0.5-2.8) Relapse < 5 years post-diagnosisd      No 1 1 1 1 Yes 4.1(2.7-6.3) 2.9(1.9-4.5) 2.8(1.6-4.8) 2.8(1.3-6.1)* Having Chemotherapye No 1 1 1 1 Yes 1.1(0.8-1.6) 0.9(0.6-1.5) 1.1(0.6-2.0) 1.2(0.6-2.3) Having Radiatione No 1 1 1 1 Yes 2.7(2-3.8) 2.0(1.3-3.1) 1.7(0.9-2.8) 3.4(1.7-6.7)*  *p<0.05 a. Multivariate analyses adjusted for type of primary cancer, diagnosis period and relapse<5 years post-diagnosis, b. Multivariate analyses for the cohort adjusted for sex, diagnosis period and relapse<5 years post-diagnosis; multivariate analyses for both female and male adjusted for diagnosis period and relapse<5 years post-diagnosis. c. Multivariate analyses for the cohort adjusted for sex, type of primary cancer and relapse<5 years post-diagnosis; multivariate analyses for both male and female adjusted for type of primary cancer and relapse<5 years post-diagnosis.  60  d. Multivariate analyses for the cohort adjusted for sex, type of primary cancer and diagnosis period; multivariate analyses for the male and female adjusted for type of primary cancer and diagnosis period e. Multivariate analyses for the cohort adjusted for sex, type of primary cancer, diagnosis period, and relapse<5 years post-diagnosis; multivariate analyses for male and female adjusted for type of primary cancer, diagnosis period, and relapse<5 years post-diagnosis  61   Figures Figure 3.1 Cumulative mortality by sex 0 5 10 15 20 25 300.000.050.100.150.200.250.300.35Time from diagnosis, yearsCumulative mortalityFemaleMale    62  Chapter 4: Second malignant neoplasms (SMN) among young adult cancer 5-year survivors          4.1 Introduction Advances in therapy for young adult cancers have led to an increase in the relative 5-year survival rate from 71% in 1975 to 87% in 2007 [114]. This has created a growing population of young adult cancer survivors. In Canada, presently more than 150,000 Canadians are survivors of a young adult cancer diagnosed since 1980 [115]. Long-term survivors of young adult cancer may face serious health risks including an increased risk of a second malignant neoplasm (SMN), excess late mortality, and functional impairment of multiple organ systems [4, 63].   Long-term survivors of childhood and adolescent cancer are at risk for the development of a SMN. This risk is likely to be the consequence of multiple factors, including environment factors and genetic susceptibility related to the original diagnosis, the treatment used, immunosuppression, and hormonal factors [4]. Previous studies have reported that the relative risk of developing a SMN varies between 2.4 and 6.1 in this group [116], depending on characteristics of the study population, the types of SMN considered, the year of diagnosis, and the length of follow-up time included in the analyses. However, there is little information on SMN risks specifically among young adult cancer survivors.  Female childhood cancer survivors appear to be at increased risk for the development of a SMN as compared to their male counterparts [25, 81, 93, 95]. The greatest number of SMNs are breast and thyroid cancers, especially among survivors of Hodgkin lymphoma [98, 100, 117, 118]. Other studies of Hodgkin lymphoma survivors reported that female survivors were at increased risk for lung cancer [88] and bone tumors [119].  In a study of acute lymphoblastic leukemia (ALL) survivors, Bhatia and colleagues reported an increased risk of solid tumors (RR=2.9, 95%CI=1.5-5.8) in females [120]. However, most studies did not specify the risks of other SMNs for females, after excluding breast cancer and thyroid cancer.    63  Previous studies among survivors of childhood cancer have reported that older age at treatment is related to decreased risk of certain types of SMN in middle age. However, it?s largely unknown whether these decreased risks are also present among young adult cancer survivors after long-term survival when the background incidence in the general population start to elevate.   In recent years, there have been many studies of SMNs among childhood cancer survivors diagnosed before 20 years old; population-based studies of SMN risk for young adult older than 20 years old are rare. In this study, we assessed the long-term risk of overall and diagnosis-specific second malignancy among a cohort of young adult cancer survivors in British Columbia, Canada within the Childhood, Adolescent, and Young Adult Cancer Survivors Research Program (CAYACS) cohort, compared their risks with those of the general population, and assessed the effect of demographic and clinical-related factors on risk.   4.2 Methods 4.2.1 Study population As stated previously, CAYACS is a population-based retrospective survivor cohort study, established for survivors of cancer diagnosed under 25 years in the province of British Columbia (BC), Canada. The study design and methodology has been described in detail elsewhere [80]. In brief, cancer survivor cohort was identified from the BC Cancer Registry at the BC Cancer Agency (BCCA). In this study, all participants were 5-yr survivors with a first diagnosis of young adult cancer (age 20-24 years) included in the Adolescent and Young Adult Cancer Classification (AYA), excluding non-melanoma skin cancer, and were residents in BC at time of diagnosis. All the eligible survivors were diagnosed between January 1, 1970 and December 31, 1995, with follow-up to the end of 2007. SMN rates were compared with rates for  the total BC population. In total, 1,248 cancer survivors were identified.   4.2.2 Data collection Second malignant neoplasm (SMN) SMNs were ascertained from the BCCR. As defined in this study, a SMN was 1) a neoplasm that occurred between 5 years after the index diagnosis and December 31, 2007, 2) a diagnosis  64  included in the International Classification of Disease for Oncology version 3 (ICDO-3) [121] with a behavior code of 3 or higher, indicating a malignant tumor, and 3) in a new anatomic location or in the same location with a new histology code. Benign tumors, local recurrences and metastases from the original cancer were not considered as SMNs. Only the first subsequent neoplasm occurring more than 5 years after the original diagnosis was counted as a SMN in this study; thus survivors who developed a second malignant neoplasm within the first 5 years after the original diagnosis were removed from the analysis (n=9). Death information was reported from the Ministry of Health Vital Statistics Agency to the BCCR and accessed for the study to determine loss to follow-up due to death.  Modifying variables Demographic characteristics, diagnostic information on the original cancer, and follow-up information were collected from the BCCR. Medical records from tertiary cancer centers in BC were abstracted to obtain clinical information, including whether or not the survivors received surgery, chemotherapy or radiation therapy (RT) for the original cancer, and the location of the radiation therapy. Approximately 74% of subjects were referred to a tertiary cancer center[80]. As all RT facilities in the province are located in a tertiary cancer center, the survivors? RT records were considered complete, but information regarding chemotherapy and surgery was missing for those cases not referred to any of these centers, as we did not have access to information for cases treated outside this system.  4.2.3 Statistical analysis Frequency distributions and differences in demographic and clinical characteristics across outcome groups were calculated.   The expected numbers of second cancers, standardized incidence ratios (SIR), and absolute excess risks (AER) of second cancers among the young adult survivor cohort were estimated by type of original diagnosis, diagnosis period, length of follow-up, type of treatment, and major type of SMN. Person-years at risk for SMN were calculated from 5 years post-original cancer diagnosis to diagnosis of the first SMN, death, or the end of the follow-up (December 31, 2007).  65  The expected number of SMNs was computed by multiplying the sex-, age group-, cause-, and period-specific incidence rate in BC by the person-years at risk in each category in the study group. The SIR was calculated as the number of observed SMNs divided by the expected number. The AER of SMN was calculated as the observed number of SMNs minus the expected number, divided by the person-years at risk, and reported as per 1,000 cancer survivors per year.   The cumulative incidence function for SMN was estimated for each sex. The Cox proportional hazards regression model was used to estimate the effects of different survivor- (sex), disease- (e.g., type and period of diagnosis of the original cancer, having early relapse) and treatment-related characteristics (chemotherapy, RT, and surgery) on the risks of SMN. In addition, the method developed by Fine and Gray was also applied in the analysis of SMN to estimate subdistribution hazard ratios (HRs) by considering death as a competing risk [122]. Since these two methods gave very similar results for effects of SMN risk factors, only the results of subdistribution hazard ratios were presented.   For missing chemotherapy and surgery status on cases who were not referred to any of the tertiary cancer centers, multiple imputation was utilized under the assumption of missing at random [111]. The imputation model contained information on type of original cancer (seven categories in the mortality analysis, five categories in the SMN analysis), diagnosis period (10-year group), having relapse<5years post-diagnosis (Yes/No), and having radiation (Yes/No), and created 20 datasets without missing chemotherapy or surgery value. The analysis was repeated 20 times and the results were combined and summarized to generate valid statistical inferences. All calculations were performed using the statistical package SAS version 9.2 (SAS Institute Inc, Cary, North Carolina, USA).  4.2.4 Study approvals  The study received ethical approval from the University of British Columbia /BC Cancer Agency Clinical Research Ethics Board. Approval for access, use and linkage of data was obtained from the BC Cancer Registry and the BC Cancer Agency Health Records Departments. To protect confidentiality, all counts less than 5 were masked in tables and text.  66   4.3 Results 4.3.1 Descriptive analysis  In total, 1248 survivors were identified, including 589 (47.2%) males and 659 (52.8%) females. The survivors were followed to the end of 2007, more than 20 years on average after their original cancer diagnosis (mean=22.1 years for the SMN study). Among the survivors in the cohort, 27% (n=341) were diagnosed with carcinomas, including cancers in the thyroid (n=117) and cervix and uterus (n=92); 25% had lymphoma (n=307); and 18% had melanoma (n=222) (Table 4.1). About a third (31.6%) were diagnosed before 1980. In total, 369 (29.6%) young adult cancer survivors received RT as a component of their treatment, and 342 (out of 894 with known chemotherapy status) (38.3%) survivors received chemotherapy.   Of the 1248 survivors of young adult cancer in the study, 62 developed a SMN during the follow-up period (Table 4.1). The average latency period between cohort entry (the date after 5-year of diagnosis) and diagnosis of the SMN was 15.0 years with a cumulative incidence of 5.0% (95%CI, 3.5%- 6.6%) after study entry (5-year after diagnosis). Of the 62 cases who had a SMN, 37.1% (n=23) were deceased by the end of follow-up. The two most common SMNs were breast carcinoma (n=18) among female survivors, which was most often after a lymphoma diagnosis (n=10), and cancer of digestive system among males (n=9).   4.3.2 Risk of SMN Young adult cancer survivors had a 3-fold higher rate of SMN (SIR=3.0, 95% CI, 2.3-3.8) compared to the cancer incidence rate of the BC population (Table 4.2). Lymphoma survivors were 7 times more likely to experience a SMN (SIR=7.0, 95%CI, 4.7-9.9). The SIR decreased with follow-up time, from 11.6 (95% CI, 7.9-16.6) in the period 5-19 years from diagnosis, to 1.7 (95% CI, 1.2-2.5) for the period 20+ years from diagnosis. The SIR increased from 2.5 to 4.0 for survivors diagnosed in the 1970s compared to those diagnosed in the early 1990s. There was no significant difference in cumulative incidence of SMNs in either sex over time (Figure 4.1).   67  RT increased the risk of SMN compared to survivors not receiving RT (SIR, 5.7 vs. SIR, 1.8 for patients without radiation exposure). Of the 35 SMNs in survivors who received RT, 19 (54.3%) developed a SMN within the previous radiation field (as determined by the radiation oncologist), most following RT for Hodgkin lymphoma (16 of 19). Twenty-six of 31 SMN cases with a original diagnosis of Hodgkin lymphoma had received RT. Twenty-seven survivors developed an SMN in the absence of radiation exposure, including 15% (4 of 27) treated with chemotherapy, 52% (14 of 27) without chemotherapy, and 9 cases with missing chemotherapy data.   4.3.3 Factors affecting SMN risk Compared with survivors of lymphoma, survivors of both melanoma and carcinoma had decreased risks for the development of a SMN (HRadj=0.3, 95%CI=0.1-0.6; HRadj =0.2, 95%CI=0.1-0.5, respectively) (Table 4.3). The other independent predictor for SMN, in addition to type of original diagnosis, was exposure to RT (HRadj=2.0, 95%CI=1.1-3.7).   4.4 Discussion  To our knowledge, this is the first study to evaluate the risk of SMN, and the effects of disease-related characteristics, in a cohort of survivors of young adult cancer. We found increased risks of SMN in the survivors compared with the general population. Survivors with an original diagnosis of lymphoma had the highest risk for the development of a SMN. Exposure to RT was directly associated with an increased risk of SMN. Use of linked population-based registries, health records, and vital statistics data in this study, allows for essentially complete ascertainment of cases and SMN outcomes over many years, and assessment of the impact of demographic and clinical factors on risk.  In this study, we found that female survivors of young adult cancer were at increased risk for the development of breast cancer, particularly those previously treated for Hodgkin lymphoma. Several previous studies conducted in childhood cancer survivors have reported that thoracic RT is associated with an increased risk of breast cancer, and it is estimated that at least 12% to 20% of women exposed to moderate to high dose thoracic RT would go on to develop breast cancer [25, 81, 93, 123]. Age at treatment did not appear to be significantly associated with the risk of  68  subsequent breast cancer in most studies [95, 98, 99]. One study reported that the risk of breast cancer was not significant in girls treated between age 5-9 years old, but the risk became significantly increased in girls treated after age 10 [124]. The possible explanations for these findings include that proliferating and developing breast tissue, rather than prepubertal breast tissues, may be more sensitive to the effect of radiation; familial cancer syndromes may have important effect in breast cancer risk; and the length of follow-up may not be sufficient for developing breast cancers [125]. Our findings are consistent with the recommendation for early breast cancer screening in women exposed to moderate to high dose thoracic RT as part of a diagnosis of cancer in childhood and young adulthood, starting at 25 years of age or 8 years after radiation [126-128].  Our finding of a 3.56 times risk of a SMN in male survivors of young adult cancer is consistent with results of other previous studies in childhood cancer survivors, such as 3.8 times risk SMN in male from childhood study in CAYACS [81, 123]. Our study also found that survivors with a first primary cancer of lymphoma and germ cell tumor are at an elevated risk of developing a SMN. However, the non-significant risks of SMN among survivors of melanoma and carcinoma may reflect the less carcinogenic effects of surgery, the main therapy for these conditions, or the effects mainly caused by the small sample size of these subgroups and the follow-up time not being long enough for developing farther SMNs.  It has been clearly demonstrated that chemotherapy and RT are associated with increased risks of SMN [25, 93, 129, 130]. In our study, we found 6-fold excess risks of SMN associated with chemotherapy and RT. 30% of SMNs developed in the RT exposure field, whereas 14 cases developed SMNs without RT and chemotherapy, suggesting that genetic or other factors may also play an important role in the occurrence of these SMNs. The earliest diagnosis time perod is the 1970?s in our study, which has resulted in shorter follow-up time from diagnosis than other studies with diagnoses starting from the 1940s. However, with the dramatic changes in cancer treatment in past 30 years, adverse late effects observed among survivors treated before the introduction of modern cancer treatment may not be applicable to survivors treated in recent decades. Therefore, it is more important to emphasize the risks of mortality and SMN among recent diagnosed survivors.  69   Although the availability of a population-based cohort, administrative datasets and tumour treatment information on the majority of subjects and the long follow-up time  were strengths of the study, our study was limited by relatively small numbers of cases, which ruled out the possibility of assessing treatment in more detail. Even after more than 20 years follow-up, the majority of young adult cancer survivors have just reached middle age when the incidences of many chronic diseases start to increase . It is possible that, with longer follow-up, the risks of SMN may increase together with other long-term late health effects. This study emphasizes the importance of screening for late effects, and the dose-response for SMN risk. Other limitations included the missing data on treatment modality (approximately 28% of cases with missing chemotherapy and surgery), and the unknown information on SMN, due to events that may have occurred outside of Canada.  In conclusion, our cohort study shows that young adult cancer survivors experience elevated risks of SMN over many years compared with the general population. Although the increased risks of SMN were generally observed across all demographic and treatment-related groups, the risk was highest in survivor groups characterized by a more recent year of diagnosis, an original diagnosis of lymphoma, and receipt of radiation therapy. These elevated risks suggest that future efforts should focus on the development of screening programs for early detection of late effects, and strategies to help prevent these significant long term health problems among survivors of young adult cancer.   70  Tables Table 4.1. Demographic and disease-related characteristics of young adult cancer survivors    Total N (%) Not having SMN N (%) Having SMN N (%) Overall 1248 1186 (95.0) 62 (5.0) Age at diagnosis (mean in years, SD) 22.7 (1.4) 22.7 (1.4) 22.3 (1.5) Age at end of follow-up (mean in years, SD) 44.8 (7.9) 44.9 (7.9) 42.3 (7.9) Follow-up time 5-yr post-diagnosis (mean in years, SD) 17.1 (7.8)  17.2 (7.8) 15.0 (7.7) Sex       Male 589(47.2) 564 (47.6) 25 (40.3) Female 659(52.8) 622 (52.4) 37 (59.7) Young Adult Cancer Diagnosis               Lymphoma 307(24.6) 276(23.3) 31 (50)         Germ Cell Tumor 178(14.3) 171(14.4) 7 (11.3)         Melanoma  222(17.8) 216(18.2) 6 (9.7)         Carcinomas 341(27.3) 329(27.7) 12 (19.4)         Others* 200 (16.0) 194 (16.4) 6 (9.7) Birth year (10-yr)     1945-1954 271(21.7) 247 (20.8) 24 (38.7) 1955-1964 528(42.3) 496 (41.8) 32 (51.6) 1965-1975 449(36) 443 (37.4) 6 (9.7) Diagnosis period (10-yr)       1970-1979 394(31.6) 362(30.5) 32 (51.6) 1980-1989 520(41.7) 496(41.8) 24 (38.7) 1990-1995 334(26.8) 328(27.7) 6 (9.7) Attained Age (by 10-year) 25-39 379(30.4) 354 (29.9) 25 (40.3) 40-49 523(41.9) 500 (42.1) 23 (37.1) 50+ 346(27.7) 332 (28.0) 14 (22.6) Follow-up time (years)       5-19 years 520 (41.7) 490 (41.3) 30 (48.9) 20+ years 728(58.3) 696 (58.7) 32 (51.6) Relapse < 5 years post-diagnosis    Yes 82 (6.6) 76 (6.4) 6 (9.7) No 1166 (93.4) 1110 (93.6) 56 (90.3)  71                 *Other types of original cancers include leukemia, central nervous system tumor, bone tumor, soft tissue sarcoma, and miscellaneous-specified tumor. Table 4.1 cont?d   Total N (%) Not having SMN N (%) Having SMN N (%) Vital Status       Alive 1110(88.9) 1071(90.3) 39(62.9) Deceased 138(11.1) 115(9.7) 23(37.1) Having Chemotherapy       Yes 342 (27.4) 318 (26.8) 24 (38.7) No 552 (44.2) 523 (44.1) 29 (46.8) Missing 354 (28.4) 345 (29.1) 9 (14.5) Having Radiation Yes 369 (29.6) 334 (28.2) 35 (56.5) No 879 (70.4) 852 (71.8) 27 (43.6) Having Treatment-related Surgery    Yes 628 (50.3) 593 (50.0) 35(56.5) No 270 (21.6) 251 (21.2) 19(30.6) Missing 350 (28.0) 342 (28.8) 8(12.9)  72   Table 4.2. Second malignant neoplasms, standard incidence ratios and absolute excess risks of young adult cancer by sexf   Entire Cohort (N=1248) Male (N=589)  Female (N=659)   Obsa Expb SMRc 95% CId AERe Obsa SIRb 95% CIc AERd  Obsa SIRb 95% CIc AERd Overall 62 21 3.0  2.3-3.8* 1.9 25 3.6 2.3-5.3* 1.9 37 2.7 1.9-3.7* 2.0  Type of Original Cancer                           Lymphoma 31 4.4 7.0  4.7-9.9* 5.3 12 6.5 3.4-11.4* 3.9 19 7.3 4.4-11.4* 6.7         Germ cell tumours 7 2.3 3.1 1.2-6.3* 1.6 7 3.7 1.5-7.6* 1.9 0 0 N/A N/A         Melanoma  6 3.4 1.7 0.6-3.8 0.7 0 0 N/A N/A 6 2.4 0.9-5.3 1.5         Carcinomas 12 7.9 1.5 0.8-2.7 0.6 <5 2.7 0.6-7.9 1.4 9 1.3 0.6-2.5 0.4         Otherse 6 2.9 2.1 0.8-4.5 1.0 <5 2.4 0.5-7.1 1.0  <5 1.5 0.3-4.3 0.6 Follow-up time (years)        5-<20 30 2.6 11.6 7.9-16.6* 5.5 12 11.2 5.8-19.6* 4.7 18 12 7.1-18.9* 6.2 >=20 32 18.4 1.7 1.2-2.5* 0.8 13 2.2 1.2-3.7* 0.9 19 1.5 0.9-2.4 0.7 Diagnosis period                    1970-1979 32 12.7 2.5 1.7-3.5* 2.0 16 4.4 2.5-7.2* 3.3 16 1.8 1.0-2.9* 1.2 1980-1989 24 6.7 3.6 2.3-5.3* 1.9 7 2.5 1.0-5.2* 0.9 17 4.3 2.5-6.9* 2.9 1990-1995 6 1.5 4.0  1.5-8.7* 1.5 <5 3.1 0.4-11.3 0.9  <5 4.7 1.3-11.9* 1.9 Having Chemotherapy Yes 24 3.8 6.3 4.1-9.4* 3.9 10 5.1 2.4-9.3* 2.6 14 7.7 4.2-12.9* 5.9 No 29 9.4 3.1 2.1-4.4* 2.1 11 3.3 1.6-5.9* 1.7 18 3.0 1.8-4.7* 2.4 Missing 9 7.8 1.2 0.5-2.2 0.2 <5 2.4 0.7-6.0 1.1 5 0.8 0.3-1.9 -0.2 Having Radiation                    Yes 35 6.2 5.7 4-7.9* 4.5 14 5.2 2.9-8.8* 3.2 21 6.0 3.7-9.2* 6.0  No 27 14.8 1.8 1.2-2.7* 0.8 11 2.5 1.3-4.5* 1.1  16 1.5 0.9-2.5 0.6 Having Surgery        Yes 35 9.4 3.7 2.6-5.2* 2.5 12 3.1 1.6-5.4* 1.5 23 4.2 2.6-6.2* 3.5 No 19 3.8 5.0  3.0-7.8* 3.5 10 6.7 3.2-12.3* 4.0 9 3.9 1.8-7.3* 3.1 Missing 8 7.7 1.0  0.5-2.0 0.1 3 1.8 0.4-5.2 0.6   5 0.8 0.3-1.9 -0.2  73  Table 4.2 Cont?d   Entire Cohort (N=1248) Male (N=589)  Female (N=659)   Obsa Expb SMRc 95% CId AERe Obsa Expb SMRc 95% CId AERe  Obsa Expb SMRc Major types of SMN            Ca: digestive system 10 2 4.9 2.4-9.1* 0.4 9 9.7 4.4-18.3* 0.8 <5 0.9 0-5.1 0.0      Ca: respiratory system 7 1.1 6.2 2.5-12.8* 0.3 <5 7.0 1.4-20.5* 0.3 <5 5.7 1.6-14.6* 0.3     Ca: breast 18 5.1 3.6 2.1-5.6* 0.6 N/A N/A N/A N/A 18 3.6 2.1-5.6* 1.1     Ca: endocrine glands 6 1.2 5.0  1.8-10.8* 0.2 <5 9.2 1.1-33.2* 0.2 <5 4.0 1.1-10.3* 0.3      Others 21 11.5 1.8 1.1-2.8* 0.1 11 2.0 1.0-3.6* 0.1  10 1.6 0.8-3.0 0.0   *p<0.05 a. Obs: Number of observed deaths b. SMR: Standardized mortality ratio c. 95%CI: 95% confidence interval d. AER: Absolute excess risk, per 1,000 years at risk e. Other type of original cancer group includes leukemia, central nervous system tumor, bone tumor,soft tissue sarcoma, and      miscellaneous specified tumor. f. As per Ministry of Health requirements, any numbers less than five cannot be specified.    74   Table 4.3 Hazard ratios for SMN    SMN(N=62)   Unadjusted  HR (95%CI) Adjusted HR (95%CI) Sexa         Female 1 1         Male 0.9(0.5-1.5) 0.7(0.4-1.2) Type of Primary Cancerb             Lymphoma 1 1         Germ cell tumours 0.4(0.2-0.9)* 0.5(0.2-1.1)         Melanoma  0.3(0.1-0.7)* 0.3(0.1-0.6)*         Carcinomas 0.3(0.1-0.5)* 0.2(0.1-0.5)*         Otherse 0.3(0.1-0.7)* 0.3(0.1-0.7)* Diagnosis periodc     1970-1979 1 1 1980-1989 1.5(0.8-2.7) 1.5(0.8-2.8) 1990-1995 1.8(0.6-5.1) 1.7(0.6-5.0) Relapse < 5 years post-diagnosisd No 1 1 Yes 1.9(0.8-4.3) 1.2(0.5-2.9) Having Chemotherapye No 1 1 Yes 1.9(1.1-3.3)* 1.3(0.7-2.5) Having Radiatione     No 1 1 Yes 2.9(1.8-4.9)* 2.0(1.1-3.7)*  *p<0.05 a. Multivariate analyses adjusted for type of primary cancer, diagnosis period and relapse<5 years post-diagnosis b. Multivariate analyses adjusted for sex, diagnosis period and relapse<5 years post-diagnosis c. Multivariate analyses adjusted for sex, type of primary cancer and relapse<5 years post-diagnosis d. Multivariate analyses adjusted for sex, type of primary cancer and diagnosis period e. Multivariate analyses adjusted for sex, type of primary cancer, diagnosis period, and relapse<5 years post-diagnosis   75  Figures Figure 4.1 Cumulative incidence of SMN by sex     76  Chapter 5: Late morbidity leading to hospitalization among young adult cancer 5-year survivors 5.1 Introduction Although improvements in treatment have dramatically increased the survival rate of cancer diagnosed in young people since the 1970s, cancer survivors may suffer from a variety of significant adverse late and chronic complications [29].  These include serious life-threatening health problems which can affect multiple organ systems. These complications are associated with the cancer itself and previous therapy which generally includes chemotherapy, radiation therapy (RT) and surgery.  Late effects have been shown to become increasingly clinically significant with aging [64].  Life-time follow-up with regular monitoring is important for survivors of childhood and young adult cancer and is likely to lead to prevention and earlier diagnosis of long term health complications [131]. Research on these complications among long-term survivors is likely to improve our understanding of the etiology of these late effects.  Most late morbidity studies involving survivors of cancer in young people focus on childhood cancer survivors (CCS). The risk of late effects in this group is very significant. Oeffinger and colleagues using self-reported questionnaire data, found that 62.3% of CCS diagnosed before 21 years of age between 1970 and 1986  had at least one chronic condition and 27.5% had severe or life-threatening conditions[16]. Geenen MM et al. reported that 74.5%  of CCS diagnosed with cancer prior to 17 years of age between 1966 and 1996 had one or more adverse events, and 36.8% had at least one severe or life-threatening or disabling adverse event, using medical assessment of adverse events from a single institute in the Netherlands [69]. In British Columbia, over 40% of CCS had at least one type of late morbidity leading to hospitalization, after up to 25 years of follow-up [65].  Young adult cancer survivors (YACS) have a unique cancer spectrum as carcinomas are much commoner in this age group. Although certain cancer types can occur in all age groups from childhood to adulthood, such as lymphoma and acute leukemia, cancers in the young adult cohort tend to have different biologic and immunophenotypic characteristics [132]. Therefore, risks and type of late effects in CCS and YACS are likely to be different. Little is  77  known about the overall burden of late morbidity affecting YACS. The magnitude of risks of late morbidity, and disease-related risk factors, have not been fully explored among YACS.   This study was conducted using linked registry, clinical, and long-term data for survivors of cancer diagnosed between 20 and 24 years of age from the Childhood, Adolescent, and Young Adult Cancer Survivors (CAYACS) Research Program [80].  The CAYACS Program is an ongoing, population-based research resource for survivors of cancer diagnosed before 25 years old in the province of British Columbia (BC), Canada. It consists of a retrospective survivor cohort and a randomly-selected comparison group, frequency-matched by sex and birth year, from population-based registries, linked with medical records and provincial government health administrative databases. Since all ?medically necessary? healthcare services are government-funded, this linkage allows us to obtain information on all ?medically necessary? services for the study groups. We estimated  the risk of late morbidity leading to hospitalization among YACS, who had survived at least 5 years, compared to the general population; and we examined the effects of demographic and disease-related factors on late morbidity.  5.2 Material and methods 5.2.1 Study population Survivor cohort: The survivor cohort was identified from the population-based British Columbia Cancer Registry (BCCR). BCCR ascertains all newly diagnosed cancer cases among residents of BC. The ascertainment of cancer incident cases by BCCR is estimated to be 95% or better. The survivors in this study were restricted to individuals primarily diagnosed between 20 and 24 years old from 1981 to 1999 with a first primary cancer included in the Classification of Cancers in Adolescents and Young Adults (AYA) [108], resident in BC at the time of diagnosis, who had survived at least five years after diagnosis, and who linked to the Client registry of the Medical Services Plan (MSP), the provincial health insurance plan (using a person-specific Personal Health Number (PHN), available since 1986).   We defined valid linkage for survivors as a minimum of one year of MSP registry linkage during the follow-up period between January 1, 1986 and December 31, 2006 from five years post- 78  diagnosis (start of follow-up). In total, 1252 young adult survivors were identified, and 902 were successfully linked to the MSP (including 448 males and 454 females)  yielding a linkage rate of 72.0%. Meanwhile, because cases with relapse and SMN were more likely to be treated with radiation for certain conditions, relapse/SMN developed within the first five years post-diagnosis are likely to overlap with treatment modality. Therefore, cases who relapsed (N=55) or experienced an SMN (N=10) were removed from the analysis, resulting in a survivor cohort with 839 cases.  Comparison group: A comparison group 10 times the size of the survivor group  (8390 individuals resident in BC 1986 or later) was randomly selected from the MSP Client registry records. The population sample was frequency-matched by sex and birth year to resemble the survivor cohort.  5.2.2 Late morbidity outcomes Late morbidity was defined as any chronic or late-occurring health condition, whether or not it was a recognized late effects of the disease or treatment [65]. In this study, morbidity leading to hospital admission was measured. Person-specific hospitalization records containing morbidity information was linked to  records of survivors and comparison subjects using PHN. These records capture data on hospital admissions of all inpatient care from all health care facilities in BC, including hospitalization dates and discharge diagnosis. Emergency visits and outpatient care visits were unavailable. Each discharge record has a recorded diagnosis ?most responsible? for the hospitalization, and up to 15 additional diagnoses, coded according to the International Classification of Diseases version 9 (ICD-9 [133]). In this study, only the ?most responsible? hospital discharge diagnosis was identified, and classified according to ICD-9 chapter (chapters correspond generally to organ systems).  The first hospitalization attributable to a chapter-specific type of problem was counted as the indicator of that type of late morbidity. Consequently, multiple hospitalizations with the same type of condition for each individual were counted only once in the analysis.    79  5.2.3 Data collection and follow-up The BCCR provided demographic and diagnostic information for young adult cancer survivors, including date of birth, sex, date of primary cancer diagnosis, and morphological and histological codes of this diagnosis based on International Classification of Disease for Oncology (ICDO), 3rd edition [121]. All the primary cancer diagnoses were further converted into a ten-category AYA classification based on morphological and histological diagnosis codes [108]. Cancer-related treatment and relapse information was manually abstracted by research staff from medical records, and included chemotherapy, radiotherapy (RT) and surgery. Second cancer information was ascertained from BCCR. The MSP Client Registry provided the demographic information for the comparison group.   Both the survivor and the comparison groups were followed from January 1, 1986 to December 31, 2006, loss to follow-up, or date of death, depending on which occurred first. Loss to follow-up was identified by inactive status in the MSP. Possible reasons include moving out of BC or death out of country. Date and cause of death for survivors were obtained from the BCCR via routine linkage with the BC Vital Statistics Agency (VSA). Death information for the comparison group was obtained from a separate linkage to the VSA.  Socioeconomic status (SES), urban/rural residential classification, and region of residence at start of follow-up, were determined using residential postal code data. SES was classified according to the average neighborhood-adjusted income quintiles using census data linking residential postal codes to Statistics Canada geographic areas [134] . Urban/rural residential classification was generated through conversion of postal code into six groups based on census data of population size and density provided by Statistics Canada [135]. Region of residence was classified according to regional health authority.   5.2.4 Statistical analysis The frequency and proportion of types of late morbidity leading to hospitalization, overall and by demographic and clinical characteristics (for survivors), were calculated for both the survivor and comparison groups, and compared using chi-square tests. In order to estimate the relative risk for  80  the late morbidity leading to hospitalization overall and in each type of late morbidity, the Poisson regression model was applied, adjusting for demographic and clinical factors. Person-years at risk were calculated as the time from the later of 5 years after the original cancer diagnosis or the entry to the MSP to death, loss to follow-up, or the end of follow-up period (December 31, 2006). Person-time was divided into 5-year age categories to adjust for changes in the number of individuals contributing at each age group.  Person-time was used as an offset in the Poisson regression. For individuals experiencing more than one type of late morbidity, only the first event in each ICD 9 chapter was counted in the analysis. The statistical software package SAS was used for data management and analysis (SAS Institute Inc, Cary, NC, USA).   5.2.5 Ethical approvals The data file for analysis does not contain any personal identifying information. Ethical approval for this study was obtained from the BC Cancer Agency (BCCA)/University of British Columbia clinical Research Ethics Board. Data access approvals were obtained from the BCCR, BCCA Health Records for access to medical records, and the BC Ministry of Health for administrative health files.  5.3 Results Sociodemographic characteristics of the study groups are shown in Table 5.1. Among the survivors, mean age at the end of follow-up was 38.2 years (SD?5.8; range, 26.4-50.4), and mean time from diagnosis was 11.2 years (SD?5.3; range, 5.0 -26.0). Since start of follow-up was at 5 years post-diagnosis, the mean follow-up time in this study was 6.2 years. Over 50 % (50.9%) were female and 6.2% were deceased at the end of 2006. The mean age of the comparison group was 37.6years (SD?6.3; range, 25.0-51.0) at the end of follow-up, mean follow-up time was 6.2 years (SD?5.2; range, 0.1 -21.0), and 1.0% were deceased at the end of the study.   Among the 839 eligible survivors, the most common primary cancer diagnoses were lymphoma (n=183, 21.8%), carcinoma (n=161, 19.2%), melanoma (n=167, 19.9%) and germ cell tumor (n=158, 18.8%) (Table 5.2). Almost half (n=415, 49.4%) of the survivors were diagnosed with their primary cancer before 1990, and about one-third (n=255, 30.4%) had surgery as their only  81  treatment. Another third (n=384, 33.8%) received chemotherapy as part of treatment, and 29.4% (n=247) received RT.   After excluding hospitalizations due to pregnancy-related conditions and congenital anomalies, half (n=415, 49.5%) of the survivors had at least one type of late morbidity leading to hospitalization during follow-up (Table 5.2). The rates of having at least one type of hospital-related condition ranged from 67.7% for central nervous system (CNS) tumor and bone tumor survivors to 39.9% for germ cell survivors. One quarter of the survivors (n=203, 24.2%) had two or more types of late morbidity, and 102 (12.2%) had three or more types of late morbidity (results not shown). Among CNS tumor survivors, 18 (37.5%) had at least two types of hospital-related morbidity. On the other hand, 37.9% (n=3183) of the comparators had at least one condition leading to hospitalization, and only 16.1% (n=1350) had two or more types of hospital-related morbidity. Female survivors had 40% more risk of having hospital-related conditions than males (Rate Ratio (RR)=1.40, 95%CI=1.10-1.78).   Figure 5.1 shows the overall incidence rates of late morbidity, excluding pregnancy-related hospitalizations and congenital anomalies conditions, by sex and age, for both survivors and comparators during the follow-up period. For both sexes, younger survivors had a higher incidence of hospital-related morbidity compared to the population sample than older survivors. For those aged 25-29 years old, the incidence of hospital-related morbidity for survivors was 1.57 times higher than the corresponding rate for the comparison group ( 95% CI, 1.34-1.83). The RR decreased to 1.52 (95% CI, 1.18-2.09) for the 35-39 year age group and 1.45 (95% CI, 0.96-2.18) for the 40-60 year age group. Meanwhile, both female survivors and comparators had a higher risk of a hospital-related condition than males before 30-34 years old, and a lower risk beyond this age.  The 415 cases of late morbidity reported among survivors, represent an incidence of 81.1 per 1,000 person-years (95%CI, 73.5-89.4),  and the 3182 reported hospitalizations among the population sample gave an incidence of 54.3 per 1,000 person-years  (95%CI, 52.4-56.2) (Table 5.3). Neoplasms, digestive system diseases and genitourinary system diseases had the highest incidence (13.6 per 1,000 person-years, 14.1 per 1,000 person-years and 13.5 per 1,000 person- 82  years respectively) among survivors. Of the 107 survivors who were subsequently hospitalized for cancer, 26 were hospitalized for the same cancer type; 81 had a different cancer. After adjusting for attained age, sex, SES, region of residence, and urban/rural residential status, the risk of all hospital-related morbidity among survivors was 1.55 times higher than the comparison group (95%CI 1.40-1.72). The highest risks for morbidity leading to hospitalization were found for neoplasms (RR, 3.85; 95%CI=3.09-4.80) and blood diseases (RR, 3.96; 95%CI=1.83-8.60).   Compared to the population sample, the risk of having hospital-related morbidity was significantly higher for survivors who had received any combination of treatment modality (Table 5.4). The rate ratio was significantly higher among survivors with three treatment modalities compared to the comparison group  (RR, 2.96; 95%CI,1.88-4.65). This increased rate ratio was primarly seen for survivors of carcinoma  (RR, 3.86; 95%CI,1.73-8.62).  5.4 Discussion  This study is the first to examine the overall burden of long-term late morbidity leading to hospitalization among YACS. We found that over half of 5-yr YACS developed at least one type of  late morbidity leading to hospitalization. Compared with the general population, YACS have increased risks of having hospital-related morbidity, particularly for blood diseases and neoplasms. The risks were higher among survivors receiving all three treatment modalities. These findings emphasize the importance of follow-up surveillance for late effects among young adult cancer survivors.  The incidence of late morbidity leading to hospitalization found in our study was higher than the risk of severe or life-threatening conditions (CTCAE grade 3 or 4 [136])  reported from  the Childhood Cancer Survivor Study (CCCS) [16, 69, 137-139], which examined risks among survivors diagnosed from birth to 21 years old using questionnaires. The survivors in these studies were also  diagnosed in earlier eras (1970-1986) [137-139], when treatments may have been significantly different, although a recent study on childhood cancer included survivors diagnosed in a later year (1966-1996) [69]. In two recent studies, the estimated risks of having chronic conditions or adverse events were higher than our findings (62.3% and 74.5%  83  respectively), whereas the risks of severe conditions were lower (27.5% and 36.8% respectively) [16, 69]. The differences among studies may be caused by different definitions of outcomes, the inclusion of different groups of diagnoses and the proportion of survivors with different diagnoses that would be expected to have different risks of late effects, as well as the differences in follow-up periods.  Previous studies of CCS have showed that female sex was associated with a higher risk of long-term adverse outcomes, including intellectual function, cardiac dysfunction, obesity, abnormal pubertal timing, steroid-induced osteonecrosis, and primary hypothyroidism [16, 69, 71, 140, 141]. Oeffinger KC et al. reported 1.5 times higher risk (95%CI, 1.3-1.6) for female survivors of grade 3 or higher health conditions [16]. Hudson and colleagues found that being female was associated with reporting at least 1 adverse general health and mental health (OR=1.4, 95%CI, 1.3-1.6) for all level of impairment [140]. In our study, we have reported a similar difference of 1.4 times increased risk (95%CI, 1.1-1.8) of late morbidity leading to hospitalization for female YACS. However, the difference we observed was mainly caused by the second cancer and diseases in genitourinary system among female survivors (results not shown), suggesting that cancer treatment received in young adult period may have a different impact to the survivors comparing with the treatment delivered at an earlier age.   Our findings are consistent with previous reports that combination therapy with multiple treatment modalities, including chemotherapy, RT and surgery, is associated with an increased risk of significant morbidity [16, 69]. In our study, we found a 3 times higher risk of hospital-related morbidity associated with the use of chemotherapy, RT and surgery for all YACS, and 2 times higher risks for YACS who had received treatment with RT and surgery.   Some limitations of this study are related to the record linkage methodology and the administrative data sources. Our study focused on the  late effects leading to hospitalization among young adult cancer survivors; as a result, conditions not leading to hospitalization were  not considered in the analysis, as well as the repeated events within the same type of late morbidity. Because only active subjects from MSP were included in the study, information about subjects who had moved out of the province, was not available which may lead to overestimates  84  in morbidity for survivors.  However, linkage rates (72% for this study; 79% for the total original cohort [80]) were high compared to participation rates for studies using conventional identification, recruitment, tracking, and data collection methods (eg recruitment and data collection rate=61.4%  in the CCSS [77]). Furthermore, loss to follow-up (4% for survivors vs. 6% for the comparison group [80]) was extremely low. It is possible that a small number of survivors may be included in the selection of the comparison group, however, because of the small prevalence of survivors in the population, this should not compromise our ability to estimate the risk of late morbidity in this study.  Advances in diagnosis, treatment, and supportive care have improved the 5-year survival of young adult cancer patients. However, these survivors still face the high risks for a wide range of late morbidities. In our study, more than half of the survivors developed problems leading to hospitalization over a maximum of 21 years follow-up. In addition, it is possible that some exposures may have cumulative effects that are expressed over long periods of time and other late-onset effects may occur in the future. It is also possible that changes in therapy over recent years may be associated with a different spectrum of late effects in young adult cancer survivors, which have not been captured in our current study cohort. Therefore, it is imperative that survivors of young adult cancer have ongoing long term follow up to screen for potential late toxicity related to their previous disease and therapy. Future studies should be conducted on screening and early detection of these late health problems, in the hope that the overall late effects risk can be reduced.  In this way effective clinical programs and guidelines can be developed to meet the needs of  young adult cancer survivors.   85  Tables	Table 5.1. Sociodemographic characteristics of survivors and comparison group   Survivors  Comparison Group  N %  N % Overall  839 100  8390 100 Sex Male 412 49.1  4120 49.1  Female 427 50.9  4270 50.9        Urban/rural residential status Urban 723 86.2  7046 84.0 Rural 97 11.6  1032 12.3 Unknown 19 2.3  312 3.7        Region of residence Vancouver Coastal 211 25.1  2566 30.6 Interior 111 13.2  1052 12.5 Fraser 275 32.8  2468 29.4 Island 143 17.0  1206 14.4 Northern 66 7.9  618 7.4 Unknown 33 3.9  480 5.7        SES1 5 (Highest) 143 17.0  1295 15.4  4 137 16.3  1356 16.2  3 168 20.0  1398 16.7  2 162 19.3  1745 20.8  1 176 21.0  1949 23.2  Unknown 53 6.3  647 7.7        Attained Age (2006) 25-34 years 270 32.2  3074 36.6 35-44 years 390 46.5  3970 47.3 45-60 years 127 15.1  1262 15 Deceased 52 6.2  84 1  1	Socioeconomic	Status	quintile		 86  Table 5.2. Factors affecting late morbidity leading to hospitalization risk among survivors    Total Survivors2,3 w/Morbidity  RR1,2,3 95% CI    N %    Overall  839 415     Sex Male 412 232 56.3  1.00   Female 427 183 42.9  1.40 1.10-1.78         Urban/Rural Residential Status Urban 723 351 48.5   1.00  Rural 97 56 57.7   1.21 0.88-1.67 Unknown 19 8 42.1   0.88 0.31-2.48         Region of Residence Vancouver Coastal 211 93 44.1   1.00  Interior 111 58 52.3   0.94 0.66-1.34 Fraser 275 138 50.2   0.74 0.55-0.99 Island 143 74 51.7   0.69 0.49-0.97 Northern 66 36 54.5   0.83 0.55-1.26 Unknown 33 16 48.5   0.64 0.28-1.50         SES4 5 (Highest) 143 70 49.0   1.00   4 137 70 51.1   1.34 0.94-1.91  3 168 84 50.0   1.22 0.86-1.72  2 162 71 43.8   1.45 1.01-2.09  1 176 93 52.8   1.37 0.96-1.95  Unknown 53 27 50.9   1.12 0.62-2.02         Type of Original Cancer Lymphoma 183 91 49.7   1.00  Leukemia 19 9 47.4   1.46 0.69-3.08 CNS 48 32 66.7   1.70 1.02-2.83 Bone 15 10 66.7   1.01 0.48-2.15 Soft Tissue Sarcoma 43 22 51.2   0.89 0.48-1.64 Germ Cell 158 63 39.9   0.83 0.52-1.33 Melanoma 167 79 47.3   0.92 0.57-1.49 Carcinoma 161 87 54.0   1.01 0.65-1.57 Other 15 9 60.0   1.55 0.73-3.29 Unknown 30 13 43.3   0.99 0.53-1.87         Calendar Period of Diagnosis 1981-1985 210 150 71.4   1.00  1986-1990 205 124 60.5   1.56 1.19-2.05 1991-1995 243 100 41.2   2.02 1.51-2.70  1996-1999 181 41 22.7  4.54 3.06-6.75                          87          Table 5.2 Cont?d   Total Survivors w/Morbidity  RR1,2 95% CI    N %    Treatment Modality Chemo Only 63 29 46.0   1.00  Radiation Only 23 13 56.5   0.57 0.33-1.00 Surgery Only 255 132 51.8   0.38 0.20-0.73 Chemo and Surgery 90 41 45.6   0.90 0.57-1.40 Chemo and Radiation 105 48 45.7   0.71 0.43-1.19 Radiation and Surgery 93 55 59.1   0.66 0.44-0.98 Chemo, Radiation and Surgery 26 19 73.1   0.82 0.47-1.41 Unknown 184 78 42.4   0.73 0.54-1.00 1 Rate ratio was adjusted for all other variables in the table, as well as age 2 Pregnancy-related hospitalization events, including normal delivery, were excluded from the analysis.3Congenital anomalies events were excluded from the analysis. 4 Socioeconomic Status quintile       88  Table 5.3. Rate ratios of late morbidity leading to hospitalization for young adult cancer survivors vs. comparison group    Incidence Rate Ratio1,2,3   N Rate per 1000 pys 95% CI Raw 95% CI Adjusted 95% CI          Overall2,3 Survivors 415 81.1 73.5-89.4 1.50 1.35-1.66 1.55 1.40-1.72  Controls 3182 54.3 52.4-56.2              Infection Survivors 15 1.8 0.9-3.2 1.94 1.12-3.36 2.11 1.21-3.66  Controls 87 0.9 0.7-1.2              Neoplasm Survivors 107 13.6 10.8-17.0 3.98 3.19-4.95 3.85 3.09-4.80  Controls 319 3.4 3.0-3.9              Endocrine Survivors 9 1.1 0.4-2.3 1.71 0.85-3.46 1.64 0.81-3.32  Controls 59 0.6 0.5-0.8              Blood Survivors 9 1.1 0.5-2.3 4.23 1.97-9.10 3.96 1.83-8.60  Controls 24 0.3 0.2-0.4              Mental Survivors 29 3.5 2.2-5.4 1.04 0.71-1.52 1.08 0.74-1.58  Controls 315 3.4 3.0-3.9              Nervous System Survivors 30 3.6 2.3-5.5 1.46 1.00-2.14 1.43 0.98-2.10  Controls 229 2.5 2.1-2.9              Circulatory System Survivors 34 4.1 2.7-6.1 1.95 1.36-2.81 1.91 1.32-2.76 Controls 198 2.1 1.8-2.5                        89           Table 5.3 Cont?d   Incidence Rate Ratio1,2,3   N Rate per 1000 pys 95% CI Raw 95% CI Adjusted 95% CI Respiratory System Survivors 46 5.6 3.9-7.9 1.69 1.24-2.30 1.70 1.25-2.33 Controls 309 3.3 2.9-3.8              Digestive System Survivors 109 14.1 11.2-17.6 1.23 1.01-1.50 1.23 1.01-1.50 Controls 1004 11.5 10.6-12.3              Genitourinary System Survivors 105 13.5 10.7-17.0 1.36 1.11-1.67 1.35 1.10-1.65 Controls 875 9.9 9.2-10.7              Skin Survivors 14 1.7 0.8-3.1 1.59 0.91-2.78 1.55 0.89-2.73 Controls 99 1.0 0.8-1.3              Musculoskeletal System Survivors 65 8.1 6.0-10.8 1.35 1.04-1.75 1.34 1.04-1.74 Controls 547 6.0 5.4-6.6              Symptoms, Signs and Ill-Defined Conditions Survivors 58 7.1 5.1-9.6 1.77 1.34-2.34 1.78 1.34-2.35 Controls 371 4.0 3.5-4.5              Injury and Poisoning Survivors 72 8.9 6.7-11.8 1.17 0.92-1.49 1.25 0.98-1.60 Controls 685 7.6 7.0-8.3              Supplementary Classification Survivors 111 14.1 11.2-17.5 1.70 1.39-2.08 1.63 1.33-1.99 Controls 742 8.3 7.6-9.0        90  1Rate	ratio	was	adjusted	for	age,	sex,	health	authority,	residential	status,	and	SES	.	2Pregnancy?related	hospitalization	events,	including	normal	delivery,	were	excluded	from	the	analysis.	3 Congenital anomalies events were excluded from the analysis.	 91  Table 5.4. Rate ratios of late morbidity leading to hospitalization for young adult cancer survivors vs. comparison group by treatment  Rate Ratio (95% CI)1, 2,3  All Survivors Lymphoma Germ Cell Melanoma Carcinoma Surgery Only 1.35 ( 0.94-1.95 ) 1.4 ( 0.89-2.2 ) N/A4 0 0 Chemo Only  1.27 ( 0.74-2.2 ) 1.14 ( 0.28-4.56 ) N/A4 0 3.03 ( 1.26-7.29 )Radiation Only  1.69 ( 1.42-2.02 ) 1.26 ( 0.6-2.64 ) 1.55 ( 1.02-2.34 ) 1.36 (1.04-1.79) 1.83 ( 1.29-2.6 ) Chemo and Surgery  1.29 ( 0.94-1.75 ) 1.36 ( 0.99-1.87 ) 1.08 ( 0.72-1.61 ) 0 13.38 ( 4.3-41.6 )Chemo and Radiation  1.46 ( 1.1-1.94 ) 1.51 ( 0.91-2.51 ) 0 0 0 Radiation and Surgery  1.74 ( 1.33-2.28 ) 2.51 ( 1.25-5.03 ) 1.09 (0.60-1.97) 0 1.63 ( 0.81-3.27 )Chemo, Radiation and Surgery  2.96 ( 1.88-4.65 ) 1.40 ( 0.89-2.2 ) 0 0 3.86 ( 1.73-8.62 )1Rate	ratio	was	adjusted	for	sex,	health	authority,	residential	status,	and	SES	.	2Pregnancy?related	hospitalization	events,	including	normal	delivery,	were	excluded	from	the	analysis.	3	Congenital anomalies events were excluded from the analysis.	4N/A:	Not	applicable.		Cannot	be	estimated	due	to	the	small	number   	 92  Figure Figure 5.1. Incidence rates of late morbidity leading to hospitalization *   *	Pregnancy?related	hospitalization	events,	including	normal	delivery,	were	excluded	from	the	analysis.  93  Chapter 6: Young adult cancer survivors? willingness to participate in genetic studies 6.1 Introduction Because of significant improvements in treatment, more than 80% of adolescents and young adults diagnosed with cancer between the ages of 15 and 29 survive at least five years [142]. However, many of these survivors will experience significant adverse late and chronic health problems involving multiple organ systems [4, 16, 33, 137-140, 143]. It is increasingly recognized that genetic susceptibility is very likely to play a role in determining the risk of developing long-term late effects [68].  There is currently limited understanding about how genetic factors may affect the development of specific late effects, and the participation of cancer survivors in research is urgently needed. However, recruitment of young adult cancer survivors (YACS) to participate in these studies is likely to be challenging and it is important to understand the factors that determine their willingness to participate.   A review of the available literature indicates that previous studies about interest in participation in medical research have yielded conflicting results. Murphy et al. observed that more than 70% of subjects were willing to participate in a genetic study [144], whereas a survey conducted by Katz and colleagues demonstrated a low rate (around 28%) of willingness to participate in a hypothetic research among racial minorities[145]. A British Columbia-based survey on willingness to participate in general health research found a willingness to participate rate of over 85% [146]. Two studies among the general population identified factors relating to willingness to participate in psychiatric genetic research, specifically (1) perceived risks, ranging from discomfort to actual harm; (2) personal benefits, such as increasing knowledge and receiving better treatment; (3) altruism, including benefit to the society and improvement in medical knowledge (a common motivation in all studies); and (4) attitudes to medical research, including doctors? support and perceived value of medical research goals [147-150] . Additionally, research has shown that socio-demographic factors such as age, sex, ethnicity, education level, marital status, having children, income, health status and family history are associated with willingness to participate [144, 151-156]. For example,  previous studies have identified that older or less educated individuals are less willing to participate in genetic studies, regardless of ethnicity  [144, 151-156]. Overall, these results emphasize that willingness to participate is a complex behavioral  94  issue, influenced by many factors and that rates may vary in different populations  However, we are not aware of any studies that assess the willingness of YACS  to participate in late effects studies, particularly those that involve genetic testing, and what factors are correlated with their willingness.   Information regarding attitudes and behavioral intentions of YACS about participating in late effects studies is needed to more effectively conduct such studies in the future, identify motivations for participation, and design studies accordingly.  The aim of this study was to determine the extent to which YACS would be willing to participate in studies assessing genetic factors in late effects, and to identify factors associated with their comfort with participation.  6.2 Methods 6.2.1 Cohort and data sources This study used a self-administered survey design. Participants were randomly selected from the population-based British Columbia Cancer Registry (BCCR) database at the BC Cancer Agency (BCCA). The BCCR ascertains all newly diagnosed cancer cases among residents of BC. The ascertainment of cancer incident cases from BCCR is estimated to be 95% or better.   Cancer survivors were eligible to participate if they 1) had a primary cancer included in the Classification of Cancers in Adolescent and Young Adults (AYA) [3] ; 2) had a primary cancer diagnosed from 1970 to 2004 when they were between 20 and 29 years old; 3) were resident in BC at time of diagnosis; 4) had survived at least 5 years from diagnosis, and were alive at the time of study; 5) understood English. Non-melanoma skin cancer cases were excluded from the study. Subjects whose physician denied approval to participate were also deemed ineligible.   The BCCR provided baseline demographic, diagnostic, and vital status information, including name, date of birth, sex, date of cancer diagnosis, and morphological and histological diagnosis codes based on International Classification of Disease for Oncology (ICDO), 3rd edition [121], as well as contact information on up to five referring physicians.   95  6.2.2 Survey questionnaire  The questionnaire included 20 items related to willingness to participate in late effects studies.  The first question was the primary outcome variable.  It asked the degree to which the subject was willing to participate in a late effects genetics study. This question was scored on a three-point scale of willingness with response options ?yes, always or most of the time,? ?sometimes,? or ?no, never or almost never.? Six items asked if different study methods affected how ?comfortable and willing? respondents would be to take part in a study using each methodology. Each item was scored on a five-point Likert Scale [157] of willingness, with response options very willing, somewhat willing, neutral (?little or no effect on participation?), somewhat unwilling, and very unwilling.  Six items assessed whether different study sponsorship (i.e., government, university, hospital, cancer agency, private research firm, drug company) affected comfort in participation, and seven items asked about other factors derived from the literature and expert opinion [144-150, 158] that could affect participants? comfort in participation. All these items were also scored on a five-point Likert Scale.   Nine additional questions assessed potential socio-demographic and medical factors identified in the literature [155, 156]. The questionnaire was pilot-tested among 12 volunteers in the same age range to ensure that the questions and methods were understandable and appropriately worded.  6.2.3 Study methods This was a cross-sectional survey. A flow diagram of participant accrual is shown in Figure 6.2. Prospective subjects (N=1100) were randomly selected from 4386 potentially eligible cases based on the recruitment criteria. For each subject, up to five recorded physicians shown in the registry database were contacted by fax to confirm the patient?s eligibility and latest known address and phone number. Three rounds of follow-up calls were made if the physician did not respond in one week. Subjects having no current address were removed from the follow-up. Physician confirmation prior to contact is routinely done in studies involving the BC Cancer Registry.  A survey package was sent to the prospective participants, including a cover letter explaining the objective of the study, a self-administered questionnaire, and a prepaid return envelope.  If no  96  answer was received two weeks after the initial contact letter, three reminder phone calls were made.   The study received ethical approval from the University of British Columbia /BC Cancer Agency Clinical Research Ethics Board.  6.2.4 Data analysis  The primary cancer diagnoses were classified into ten major diagnosis groups using the adolescents and young adults (AYA) classification system, based on morphologic and histological diagnosis codes [3]. Other demographic and clinical variables were coded as either binary variables or categorical variables, including age at diagnosis, length of survival, treatment received, and region of residence from the Registry and ethnicity, self-estimated health status, education, marital status, socio-economic status, and family history of cancer from the questionnaire. Categories were aggregated as needed depending on the frequency distributions.  The five-point scales were converted into binary responses for analysis: ?very willing? and ?not very willing? (includes: somewhat willing, neutral, somewhat unwilling, and very unwilling).   Willingness to participate in genetic studies was compared across socio-demographic characteristics, different study methods, study sponsorships and other factors affecting willingness, using chi-square test. To determine the demographic factors that were associated with the outcomes, a logistic regression model was applied. All of the analyses were performed using SAS Version 9.3 (SAS Institute Inc, Cary, NC, USA).  6.3 Results 6.3.1 Participant characteristics As shown in Figure 6.2, among 401 eligible subjects with physician -confirmed address, 45 mailing questionnaires were returned with wrong mailing address and were removed from the study; 5 subjects were further found to be ineligible and were excluded, including one deceased case not previously identified, two cases who did not think himself/herself as having had cancer,  97  and two cases that could not speak English. Among the remaining 351 subjects, 124 (35.3%) did not respond to the questionnaire or the three-round follow-up calls, 62 (17.7%) refused to participate, and 165 (47%) completed the questionnaire.  Table 6.1 summarizes the baseline characteristics of the entire cohort of 1100 potential participants and broken between participants and non-participants. The average age was 46.3 years (Standard Deviation (SD)=9.8), 53.8% were female, 16.8% had been diagnosed with lymphoma, and 34.1% with carcinomas. About a third (35.4%) had been diagnosed between the ages of 20 and 24 years, and 63.4% were survivors for more than 15 years. Among the 165 participants, 32.8% had university or higher education, 78.8% were currently married or living with a partner, and 75.8% had children. 87.3% of the participants were born in Canada and 89.7% were Caucasian. 54% reported having excellent or very good self-estimated health status, 67.3% had a family cancer history, and 47.3% had annual household income above $80,000 CAD.   Several demographic and medical factors were significantly different between participants and non-participants or subjects who did not have valid contact information. Participants were more likely to be survivors of lymphoma and have received chemotherapy, but less likely to be survivors of carcinoma and have received surgery.   6.3.2 Willingness to participate in genetic late effects research In total, 79.4% of the respondents (n=131) reported being always willing to participate in a study, 18.2% (n=30) reported being sometimes willing to participate, and 2.4% (n=4) reported never being willing to participate. Figure 6.1 summarizes the results regarding factors affecting willingness to participate in a study concerned with the effect genetic factors may have on future health among YACS. Study methodology affected willingness: subjects felt most comfortable with self-report questionnaire methods (72.1%) and most uncomfortable (including both very uncomfortable and somewhat uncomfortable) with donating a blood sample (15.8%). Notably, around 40% of subjects who were sometimes or never willing to participate reported being uncomfortable with donating a blood sample (P<0.001), indicating less comfort in participating  98  in research that involved an invasive procedure such as a blood donation than those who were always willing to participate.   When assessing sponsorship, participants felt most comfortable if the study was run by the BCCA (72.2%), followed by a hospital (55.8%) and a university (53.3%). In contrast, only 21.2% of the participants felt comfortable if the study was conducted by a drug company. Respondents who were sometimes willing or would never participate were even more likely to feel uncomfortable about research studies conducted by a drug company (64.8%, P<0.001) or a private research company (47.1%, P<0.001).   In general, the strongest motivation for participation was to provide future health benefits to society (76.4%) and individuals (70.3%). A significantly higher percentage (55.8%, P<0.001) of respondents who were sometimes or never willing to participate reported being uncomfortable with several visits to a research institute, indicating that this may be one of the major barriers for survivor participation in these research studies.  6.3.3 Factors affecting willingness to participate After adjusting for current age, sex, and whether or not the participant was born in Canada, only ethnicity (i.e., being Caucasian) and income (i.e., higher income) were independently associated with always being willing to participate in genetic studies (Table 6.3). Caucasian ethnicity was associated with a nearly three-fold increase in the rate of participants willing to take part in genetic studies compared with non- Caucasian respondents (Odds Ratio (OR) 2.8, 95% Confidence Interval (CI) 1.0-8.2). Participants who did not disclose their income were much less willing to participate, with an OR of 0.2 (95%CI 0.2-0.8) compared to subjects with higher income ($80,000 and above).  6.4 Discussion This study is one of the first to examine willingness to participate in future genetic studies among survivors of young adult cancer. Our data shows that a large majority of the respondents (n=165,  99  79.4%) were always willing to participate future genetic studies. Study methods, study sponsorship, and health concerns, including the risks and benefits of participation, affected the subjects? willingness to some degree, although individuals who indicated strong willingness were generally likely to take part regardless of these issues. Among individuals with less willingness to participate (those who would only ?sometimes? take part in a study) factors that would most affect them positively were studies using self-report questionnaires, sponsorship by the BCCA, or providing health benefit to society and individuals.  Studies requiring blood sample donation, sponsored by a drug company, or needing several visits were cited as factors that would make 40-50% of these participants uncomfortable to take part.   Study sponsorship affected willingness, with the highest comfort levels associated with studies conducted by BCCA.  We need to be somewhat careful in drawing conclusions based on this finding, however, since the survey was conducted under the aegis of the BCCA and those who participated may have been positively disposed toward the BCCA or consider themselves under an obligation to participate. However, similar to previous studies [158], participants were more comfortable with studies conducted by a hospital and university, and less comfortable with those sponsored by the government, a private research company or a drug company.   Both study methods and perceived benefits/ risks were strongly associated with the willingness to participate. For example, respondents were more comfortable participating in studies requiring only self-report questionnaires. Additionally, studies providing future health benefits to the participants also were linked with greater comfort. Future researchers can build on these findings by ensuring that studies requiring biological samples or in-person participation stress how the research findings can lead to advances in health and health care.  We found that Caucasians were 2.8 times more likely to participate in future genetic studies than non-Caucasians. Respondents of Asian origin, including both East Asian and South Asian, were the major non-Caucasian ethnic group in our study. Previous studies had found that Asians were significantly more willing to participate in research [61], whereas Blacks and Latinos reported lack of trust in research  [61, 159, 160]. Asian participants had a low proportion of willingness to participate in studies involving both in-person and telephone interviews in this survey, but more  100  than three quarters of Asian participants reported feeling very/always comfortable with studies involving self-report questionnaires and blood / saliva donation, indicating that language may still be a concern here.  The results suggest that providing multi-language questionnaire and/or interpreter may help recruit minority participants. Meanwhile, Asian respondents showed a higher proportion of willingness than other ethnicities if the study was recommended by their own doctor. This result is consistent with the finding reported by Svenssson K et al, indicating that including family physician in research team may improve the recruitment of participants [61].  In previous studies, socio-demographic factors were associated with willingness to participate [144, 151-156]. In our study, only ethnicity and willingness to provide income information had an effect on willingness to participate. One potential explanation is that since we recruited only 165 subjects for the study, we may have had insufficient power to identify these associations.     Several limitations need to be mentioned. We found that only 31.9% of long-term survivors had valid current contact information using data from Cancer Registry, while more than 60% were considered ?lost to follow-up?, especially survivors diagnosed in the 1970s. This same problem has been found by other researchers using registries for survivor research [161]. This suggests the need for augmenting the cancer registry with updated contact information in order to conduct research on cancer survivors.  One option is linkage with other sources of information that contain valid and updated contact information, such as health insurance records or drivers? licences. In addition, the response rate in the study was 47%.  Although it is less than optimal, this rate is similar to other mailed questionnaire studies: e.g., Teschke et al.?s mailed survey [158], of randomly selected members of the general BC population reported a 49% response rate. Our analysis of participants and non-participants showed that individuals who returned the questionnaire were more likely to be survivors of lymphoma and have received chemotherapy, but less likely to be survivors of carcinoma and have received surgery. We also need to acknowledge that willingness to participate in future genetic factor studies was collected from a self-report questionnaire, and the response to a hypothetical question may not be the same as actual enrolment in a research study.   101  In this study, 98% of the respondents indicated that they would be willing to take part in future studies about genetic factors relating to risks of late effects, at least some of the time. This figure is almost identical to that reported by Teschke, et al. [158] in a general population survey about health research.  In another study of childhood cancer survivors, the rate of willingness to participate in long-term late effects study was 93% [162]. Together, these data provide strong support for interest in research among cancer survivors. Given the need for research about long-term sequelae in YACS, and the potential for genetic studies to help understand these risks, the findings suggest that this research is feasible and will achieve sufficient participation for robust results, if individuals can be contacted.  102  Tables Table 6.1. Descriptive summary of demographic and disease-related factors Variables Total  N(%) Responders N(%) Non-responders N(%) p-value Total 1100 165 (15.0) 935(85.0) Sex 0.0578 Female 592 ( 53.8 ) 100 ( 60.6 ) 492 ( 52.6 ) Male 508 ( 46.2 ) 65 ( 39.4 ) 443 ( 47.4 ) Original cancer diagnosis <0.001 Lymphomas 185 ( 16.8 ) 51 ( 30.9 ) 134 ( 14.3 ) Germ cell tumours 190 ( 17.3 ) 29 ( 17.6 ) 161 ( 17.2 ) Melanoma 183 ( 16.6 ) 25 ( 15.2 ) 158 ( 16.9 ) Carcinomas (except of skin) 375 ( 34.1 ) 42 ( 25.5 ) 333 ( 35.6 ) Others 167 ( 15.2 ) 18 ( 10.9 ) 149 ( 15.9 ) Age at diagnosis (years) 0.4423 20-24 389 ( 35.4 ) 54 ( 32.7 ) 335 ( 35.8 ) 25-29 711 ( 64.6 ) 111 ( 67.3 ) 600 ( 64.2 ) Diagnosis period 0.1927 1970-1979 222 ( 20.2 ) 25 ( 15.2 ) 197 ( 21.1 ) 1980-1989 334 ( 30.4 ) 54 ( 32.7 ) 280 ( 29.9 ) 1990-1999 361 ( 32.8 ) 52 ( 31.5 ) 309 ( 33.0 ) 2000-2004 183 ( 16.6 ) 34 ( 20.6 ) 149 ( 15.9 ) Age at time of study (years) 0.7597 25-39 337 ( 30.7 ) 47 ( 28.5 ) 290 ( 31.0 ) 40-49 366 ( 33.3 ) 55 ( 33.3 ) 311 ( 33.3 ) 50+ 396 ( 36 ) 63 ( 38.2 ) 333 ( 35.7 )  103  Table 6.1 Cont?d  Variables Total  N(%) Responders N(%) Non-responders N(%) p-value Length of survival (years) 0.5338 5-14 403 ( 36.6 ) 64 ( 38.8 ) 339 ( 36.3 ) 15+ 697 ( 63.4 ) 101 ( 61.2 ) 596 ( 63.7 ) Cancer treatment received 0.0027Surgery Yes 541 ( 49.2 ) 82 ( 49.7 ) 459 ( 49.1 ) No 138 ( 12.5 ) 33 ( 20.0 ) 105 ( 11.2 ) Not referred 421 ( 38.3 ) 50 ( 30.3 ) 371 ( 39.7 ) Radiation 0.188 Yes 231 ( 21 ) 41 ( 24.8 ) 190 ( 20.3 ) No 869 ( 79 ) 124 ( 75.2 ) 745 ( 79.7 ) Chemotherapy <0.001 Yes 249 ( 22.6 ) 57 ( 34.5 ) 192 ( 20.5 ) No 272 ( 24.7 ) 45 ( 27.3 ) 227 ( 24.3 ) Not referred 579 ( 52.6 ) 63 ( 38.2 ) 516 ( 55.2 ) SES at diagnosis 1(lowest) 28 ( 17.0 ) 2 30 ( 18.2 ) 3 25 ( 15.2 ) 4 32 ( 19.4 ) 5 (highest) 41 ( 24.8 ) Unknown 2 ( 1.2 )  104  Table 6.1 Cont?d Variables Total  N(%) Responders N(%) Non-responders N(%) p-value Region of residence Interior 27 ( 16.4 ) Fraser 37 ( 22.4 ) Vancouver Coastal 45 ( 27.3 ) Vancouver Island 35 ( 21.2 ) Northern 13 ( 7.9 ) Unknown 1 ( 0.6 ) Residential status at diagnosis Rural 21 ( 12.7 ) Small Community 23 ( 13.9 ) Large community 18 ( 10.9 ) Urban 94 ( 57.0 ) Unknown 2 ( 1.2 )  105  Table 6.2. Impact of study characteristics on comfort level by willingness     always willing to participate not always willing to participate* Total (%) p-value General willingness 131 (79.4) 34 (20.6)   study method, involves         completing a self-report questionnaire      very comfortable 112 (85.5) 7 (20.6) 119 (72.1) 0.0186  somewhat comfortable 18 (13.7) 19 (55.9) 37 (22.4)   Neutral 0 (0) 6 (17.6) 6 (3.6)   somewhat uncomfortable 1 (0.8) 2 (5.9) 3 (1.8)   very uncomfortable 0 (0) 0 (0) 0 (0)  completing an online survey       <0.001  very comfortable 99 (75.6) 6 (17.6) 105 (63.6)   somewhat comfortable 23 (17.6) 16 (47.1) 39 (23.6)   Neutral 4 (3.1) 6 (17.6) 10 (6.1)   somewhat uncomfortable 2 (1.5) 4 (11.8) 6 (3.6)    very uncomfortable 3 (2.3) 2 (5.9) 5 (3)  completing in-person interview       <0.001  very comfortable 84 (64.1) 4 (11.8) 88 (53.3)   somewhat comfortable 42 (32.1) 16 (47.1) 58 (35.2)   Neutral 4 (3.1) 6 (17.6) 10 (6.1)   somewhat uncomfortable 1 (0.8) 6 (17.6) 7 (4.2)   very uncomfortable 0 (0) 2 (5.9) 2 (1.2)  completing a telephone interview       <0.001  very comfortable 75 (57.3) 4 (11.8) 79 (47.9)   somewhat comfortable 37 (28.2) 11 (32.4) 48 (29.1)   Neutral 7 (5.3) 8 (23.5) 15 (9.1)   somewhat uncomfortable 10 (7.6) 8 (23.5) 18 (10.9)    very uncomfortable 2 (1.5) 3 (8.8) 5 (3.0)  donating DNA by providing saliva sample       <0.001  very comfortable 88 (67.2) 5 (14.7) 93 (56.4)   somewhat comfortable 34 (26.0) 11 (32.4) 45 (27.3)   neutral 2 (1.5) 7 (20.6) 9 (5.5)   somewhat uncomfortable 3 (2.3) 8 (23.5) 11 (6.7)   very uncomfortable 4 (3.1) 3 (8.8) 7 (4.2)                       106  Table 6.2 Cont?d      always willing to participate not always willing to participate* Total (%) p-value study method, involves     donating DNA by providing blood sample       <0.001  very comfortable 76 (58.0) 5 (14.7) 81 (49.1)   somewhat comfortable 38 (29.0) 7 (20.6) 45 (27.3)   neutral 4 (3.1) 9 ( 26.5 ) 13 (7.9)   somewhat uncomfortable 7 (5.3) 7 (20.6) 14 (8.5)    very uncomfortable 6 (4.6) 6 (17.6) 12 (7.3)  Study sponsorship, the study is run by         the government    <0.001  very comfortable 55 (42.0) 2 (5.9) 57 (34.5)   somewhat comfortable 33 (25.2) 8 (23.5) 41 (24.8)   neutral 32 (24.4) 16 (47.1) 48 (29.1)   somewhat uncomfortable 9 (6.9) 2 (5.9) 11 (6.7)   very uncomfortable 2 (1.5) 6 (17.6) 8 (4.8)  an university       <0.001  very comfortable 84 (64.1) 4 (11.8) 88 (53.3)   somewhat comfortable 24 (18.3) 12 (35.3) 36 (21.8)   neutral 22 (16.8) 15 (44.1) 37 (22.4)   somewhat uncomfortable 1 (0.8) 0 (0) 1 (0.6)    very uncomfortable 0 (0) 3 (8.8) 3 (1.8)  a hospital       <0.001  very comfortable 87 (66.4) 5 (14.7) 92 (55.8)   somewhat comfortable 25 (19.1) 18 (52.9) 43 (26.1)   neutral 17 (13.0) 8 (23.5) 25 (15.2)   somewhat uncomfortable 2 (1.5) 0 (0) 2 (1.2)   very uncomfortable 0 (0) 3 (8.8) 3 (1.8)  the BCCA       0.0555  very comfortable 109 (83.2) 11 (32.4) 120 (72.7)   somewhat comfortable 13 (9.9) 18 (52.9) 31 (18.8)   neutral 9 (6.9) 4 (11.8) 13 (7.9)   somewhat uncomfortable 0 (0) 0 (0) 0 (0)    very uncomfortable 0 (0) 1 (2.9) 1 (0.6)  a private research company       <0.001  very comfortable 45 (34.4) 2 (5.9) 47 (28.5)   somewhat comfortable 30 (22.9) 6 (17.6) 36 (21.8)   neutral 28 (21.4) 10 (29.4) 38 (23.0)   somewhat uncomfortable 19 (14.5) 9 (26.5) 28 (17.0)   very uncomfortable 9 (6.9) 7 (20.6) 16 (9.7)   107  Table 6.2 Cont?d      always willing to participate not always willing to participate* Total (%) p-value Study sponsorship, the study is run by     a drug company       <0.001  very comfortable 35 (26.7) 0 (0) 35 (21.2)   somewhat comfortable 27 (20.6) 1 (2.9) 28 (17.0)   neutral 20 (15.3) 11 (32.4) 31 (18.8)   somewhat uncomfortable 31 (23.7) 11 (32.4) 42 (25.5)    very uncomfortable 18 (13.7) 11 (32.4) 29 (17.6)  Other factors that affect willingness         Recommended by your own doctor    <0.001  very comfortable 93 (71.0) 6 (17.6) 99 (60.0)   somewhat comfortable 28 (21.4) 14 (41.2) 42 (25.5)   neutral 8 (6.1) 11 (32.4) 19 (11.5)   somewhat uncomfortable 2 (1.5) 1 (2.9) 3 (1.8)   very uncomfortable 0 ( 0 ) 2 (5.9) 2 (1.2)  Provides future health benefits to yourself       0.009  very comfortable 107 (81.7) 9 (26.5) 116 (70.3)   somewhat comfortable 19 (14.5) 14 (41.2) 33 (20)   neutral 5 (3.8) 10 (29.4) 15 (9.1)   somewhat uncomfortable 0 (0) 0 (0) 0 (0)    very uncomfortable 0 (0) 1 (2.9) 1 (0.6)  Provides future health benefits to society       0.34  very comfortable 115 (87.8) 11 (32.4) 126 (76.4)   somewhat comfortable 12 (9.2) 17 (50.0) 29 (17.6)   neutral 4 (3.1) 5 (14.7) 9 (5.5)   somewhat uncomfortable 0 (0) 0 (0) 0 (0)   very uncomfortable 0 (0) 1 (2.9) 1 (0.6)  No physical pain involved in this study       0.0035  very comfortable 103 (78.6) 10 (29.4) 113 (68.5)   somewhat comfortable 13 (9.9) 15 (44.1) 28 (17.0)   neutral 15 (11.5) 7 (20.6) 22 (13.3)   somewhat uncomfortable 0 (0) 1 (2.9) 1 (0.6)    very uncomfortable 0 (0) 1 (2.9) 1 (0.6)                  108  Table 6.2 Cont?d      always willing to participate not always willing to participate* Total (%) p-value Other factors that affect willingness         Will receive compensation for your time       <0.001  very comfortable 67 (51.1) 9 (26.5) 76 (46.1)   somewhat comfortable 11 (8.4) 9 (26.5) 20 (12.1)   neutral 51 ( 38.9 ) 13 (38.2) 64 (38.8)   somewhat uncomfortable 1 (0.8) 1 (2.9) 2 (1.2)   very uncomfortable 1 (0.8) 2 (5.9) 3 (1.8)  Requires several visits to the research institute       <0.001  very comfortable 44 (33.6) 2 (5.9) 46 (27.9)   somewhat comfortable 43 (32.8) 6 (17.6) 49 (29.7)   neutral 27 (20.6) 7 (20.6) 34 (20.6)   somewhat uncomfortable 11 (8.4) 13 (38.2) 24 (14.5)   very uncomfortable 6 (4.6) 6 ( 17.6 ) 12 (7.3)  Will receive a copy of the results of the study       0.0016  very comfortable 101 (77.1) 10 (29.4) 111 (67.3)   somewhat comfortable 15 (11.5) 13 (38.2) 28 (17.0)   neutral 15 (11.5) 9 (26.5) 24 (14.5)   somewhat uncomfortable 0 (0) 0 (0) 0 (0)    very uncomfortable 0 (0) 2 (5.9) 2 (1.2)    109  Table 6.3. Impact of demographic factors on willingness to participate     Univariate OR Adjusted OR** Ethnicity white vs non-white 3.2(1.0-9.0) 2.8(1.0-8.2) Highest education level university+  vs not having univ 1.0(0.4-2.2) 1.0(0.4-2.4) Marital status Married/ Living with partne vs. not married 1.4(0.6-3.4) 1.6(0.6-3.8) Have had any children no vs. yes 0.8(0.4-2.0) 0.8(0.4-2.0) Self-estimated health status excellent/very good vs. others 1.2(0.6-2.6) 1.2(0.6-2.8) Family cancer history no vs. yes 1.2(0.6-2.6) 1.2(0.6-2.8) Income   low(<80,000)vs. high (80,000+) 1.3(0.5-3.2) 0.8(0.2-2.0) not disclose vs.high (80,000+) 0.4(0.2-0.9) 0.2(0.2-0.8)    110  Figures  Figure 6.1. Impact of study characteristics on comfort level with participation (all subjects) 0 20 40 60 80 100Run by a drug companyRequries several visits to the research instituteRun by a private research companyRun by the governmentWill receive compensation for your timeInvolves completing a telephone interviewInvolves donating DNA by providing blood sampleInvolves completing an in-person interviewRun by an universityRun by a hospitalInvolves donating DNA by providing saliva sampleRecommended by your own doctorInvolves completing an online surveyWill receive a copy of the results of the studyNo physical pain involved in this studyProvides future health benefits to yourselfInvolves completing a self-report questionnaireRun by the BC Cancer AgencyProvides future health benefits to societyVery Comfortable Somewhat Comfortable Neutral Somewhat Uncomfortable Very UncomfortablePercentage  111  Figure 6.2. Young adult survey study flow diagram  Randomly Selected from CAIS (n= 1100) Physician?s info is N/A (n= 182) Sent fax to Physician (n= 918)Enrollment?? Not Physician?s current pt/ or without subject?s current contact info (n=458) ? Subject can?t be contacted (n=59) Mail survey to subjects (n=401)   Wrong mailing address (n=45) Eligible subject (n= 351) ? No response (n=124, 35.3%) ? Refuse (n=62, 17.7%) Reply (n=165, 47.0%) Stop follow-up Contact?Physician?Contact?Subject? 112  Chapter 7: Conclusion This thesis is the first study that has been undertaken to evaluate the risk of late effects in a cohort of YACS from three aspects: mortality, second malignant neoplasm (SMN) and late morbidity, and the effects of characteristics influencing these risk. It is also the first study to understand YACS? willingness to participate late effects studies in the future, because accruing the sufficient number of participants is critical to ensure the results from such studies are reliable and representative.   7.1 Summary and discussion of the study findings  7.1.1 Late mortality and SMN among YACS Chapters 3 and 4 measure the overall and cause-specific risks of late mortality and SMN among YACS. YACS had increased risks of mortality and SMN compared with those of the general population. The overall mortality rate for the young adult survivors was almost six times higher than the corresponding rate for the BC population. Not surprisingly, survivors with a recurrence of their original cancer during the first five years after diagnosis experienced higher mortality rates. The risk of SMN among YACS was 3-fold higher compared with the age-adjusted cancer incidence in   the BC population. The results indicate that the diagnosis of a primary cancer had a significant impact on survivors? mortality and the probability of their having an SMN. The highest risk of mortality was observed among central nervous system (CNS) tumor survivors, who were three times more likely to die than lymphoma survivors, whereas the lowest risk was seen in survivors of germ cell tumors or carcinomas. Lymphoma survivors had the highest risk of experiencing an SMN.  Other factors also affected late mortality and SMN risks. Survivors having a longer follow-up time had a lower risk of mortality. We classified the cause of death (COD) into three major groups: death due to the original cancer, death due to SMN, and non-cancer death. We identified the original cancer as being the leading cause of death for survivors with less than 20 years of follow-up, and non-cancer related deaths as being the leading cause of death amongst survivors with more than 20 years of follow-up, followed by death due to SMNs, representing 43.3% and 33.3% of deaths, respectively. Male survivors showed a higher risk of death due to SMN and  113  non-cancer causes than females. Several other studies have suggested that the COD is potentially related to the length of follow-up [21, 83, 107]. A study of 15+ year survivors of childhood and adolescent cancers reported SMN to be the leading cause of death [21]. Taken together, these results suggest that the longer the patient has survived, the less likely she/he is to die from the original cancer.  Survival time also affected the risk of SMN. The risk of SMN was less in survivors at 5-19 years from the time of diagnosis than those 20+ years from the diagnosis time. However, the cumulative incidence of developing an SMN did not significantly vary over time.   Chemotherapy and RT were associated with increased risks of SMN [25, 93, 129, 130]. In our study, we found a 6-fold excess in risks of SMNs associated with chemotherapy and RT. 30% of SMNs developed in the RT exposure field, whereas 14 cases (22.6%) developed SMNs without either RT or chemotherapy, suggesting that genetics or other factors may also play an important role in the etiology of these SMNs.  Another important observation was that female survivors of young adult cancer, particularly those previously treated for Hodgkin lymphoma, were at an increased risk of breast cancer. Several previous studies conducted on childhood cancer survivors have reported that thoracic RT is associated with an increased risk of breast cancer, and it is estimated that at least 12% to 20% of women exposed to moderate to high dose thoracic RT will go on to develop breast cancer [25, 81, 93, 123]. The subject?s age at treatment did not appear to be significantly associated with the risk of subsequent breast cancer in most studies [95, 98, 99]. However, one study reported that the risk of breast cancer was not significant in girls treated between the ages of five and nine, but that the risk significantly increased in girls treated after age 10 [124]. Possible explanations for these findings  are that proliferating and developing breast tissue, rather than prepubertal breast tissue, may be more sensitive to the effects of radiation; familial cancer syndromes may have important effect in breast cancer risk; and the length of follow-up may not be sufficient for developing breast cancers in some studies [125]. Our findings are consistent with the recommendations for early and increased breast cancer screening, beginning at 25 years of age or  114  eight years after radiation, in women exposed to moderate to high dose thoracic RT as part of their treatment for cancer during young adulthood [126-128].  7.1.2 Late morbidity among YACS Chapter 5 focuses on measuring the risk of late morbidity leading to hospitalization among YACS. In this study, YACS were found to have a high risk for a wide range of late morbidities. 455 (50.4%) of 5-yr YACS and 3419 (37.9%) of individuals in the comparison group developed at least one type of late morbidity leading to hospitalization. Overall, YACS have increased risks of experiencing hospital-related morbidity compared with the general population. The highest risks were found for hospitalization due to blood diseases and neoplasms. We observed a 3-time higher risk of hospital-related morbidity associated with the use of chemotherapy, RT and surgery for all YACS, and a 2-time higher risk for YACS who had received treatment with RT and surgery.  We also observed that the female sex was associated with a higher risk of late morbidity leading to hospitalization, which was consistent with the results from the previous studies. Female sex was associated with a 1.5-time increased risk (95%CI, 1.2-1.9) of late morbidity leading to hospitalization compared with that of male YACS. Our results showed that the increased risk of late morbidity among female YACS was mainly caused by secondary cancers and diseases of the genitourinary system. However, in studies of late effects in childhood cancer survivors, being female was more likely to be associated with a higher risk of other adverse outcomes, including cardiac dysfunction, steroid-induced osteonecrosis, and primary hypothyroidism [16, 69, 71]. This discrepancy in results may indicate that the cancer treatment received during the young adult period may have a different impact on survivors compared with the treatment delivered at an earlier age.   Finally, this research examined late morbidity leading to hospitalization; therefore, conditions not leading to hospitalization, such as intellectual function and obesity, were not considered in the analysis. Second, when we select our comparison group from the general population for this study, a small number of survivors may be included. However, survivors will on average be  115  included in the control group in the same prevalence as they exist in the general population, and because of the small prevalence of survivors in the population, this should not compromise our ability to estimate the risk of late morbidity.   7.1.3 Willingness to participate in late effects studies among YACS The studies reported in Chapters 3, 4 and 5 were based on administrative and medical data.  While they yielded answers to some important questions, such as the risk of mortality and having SMNs compared with the general population, certain questions, such as the survivors? preference and health-related quality of life, cannot be answered in this way and will require collecting additional information directly from the survivors themselves. However, recruiting sufficient numbers of YACS participants in such studies is critical and challenging. Therefore, we conducted the survey study described in Chapter 6 to understand survivors? willingness to participate in late effects studies, including genetic research.   The study found that a large majority of its respondents were always willing to participate in future genetic studies. Study methods, study sponsorship, and health concerns affected their willingness to some degree, although individuals who indicated strong willingness were generally more likely to say that they would take part in the studies regardless of these factors. Among respondents who would only ?sometimes? take part in a study, using self-report questionnaires, sponsorship by the BCCA, and providing health benefits to the society and individuals were more important, when compared with other factors. Previous studies have found that respondents are more likely to participate in studies conducted by a hospital or university, and less likely to participate in studies sponsored by the government, private research companies or drug companies [158]. Our study showed similar results in that regarding study sponsorship, the respondents felt the highest comfort level for studies conducted by BCCA. However, we need to be careful in drawing conclusions based on this finding. Because the survey was conducted under the aegis of the BCCA, it is possible that those who participated may have been positively disposed toward the latter or considered themselves under an obligation to participate. Understanding participants? willingness and the factors affecting their willingness may assist  116  researchers in more effectively communicating with participants to increase accrual in future studies.   Participants? willingness varied significantly in terms of some socio-demographic characteristics. Two factors, ethnicity (being white versus non-white) and annual income (not disclosed versus a high $80,000 +), were independently associated with survivors? willingness to parcipate. A number of studies have found that Asian participants are more willing to participate in research than blacks and Latin Americans [61, 159, 160]. In our study, Asian respondents were less willing to participate in studies involving both in-person and telephone interviews than Caucasians, but a higher proportion were willing to participate in studies with self-report questionnaires or those that required blood or saliva donations. Language barriers might be a concern among these participants, and providing them with multi-language questionnaires and/or interpreters might help recruit those of minority groups. However, the willingness to participate in late effects research is a complex behavioural issue and the rates of participation and factors affecting enrolment may vary in different populations, creating the need to target different groups, such as ethnicity, with different strategies.   In this study, we used a mailed questionnaire study design. We found that only 31.9% of long-term survivors had valid current contact information using data from Cancer Registry, while more than 60% were considered ?lost to follow-up?, given lack of current contact information. This suggests the need for augmenting the cancer registry with updated contact information in order to conduct research on cancer survivors.  One option is linkage with other sources of information that contain valid and updated contact information, such as health insurance records or drivers? licences. The comparison between respondents and non-respondents showed that individuals who returned the questionnaire were more likely to be survivors of lymphoma and have received chemotherapy, but less likely to be survivors of carcinoma and have received surgery. It needs to be pointed out that our measure of the willingness of participants to take part in future genetic studies was determined from a self-report questionnaire regarding future behavioural intentions, and that the response to a hypothetical question may not be the same as the actual enrolment numbers in a research study.   117  7.2 Strengths, limitations and methodological concerns 7.2.1 Strengths This study utilized an epidemiologic approach to understand the risks of the late effects in a geographically-defined YACS population. The study also employed a data linkage method to obtain detailed socio-demographic and clinical-related factors, and the outcomes from multiple data sources, including administrative databases and medical records. The current research is among the first to have accessed the overall burden of late effects among YACS and to have examined the risk factors that influence these late effects, using a multivariable statistical approach. In this study, survey methodology was also applied to assess the feasibility of conducting late effects studies in the future.   Traditional survey studies collecting late effects outcomes and using questionnaires often face potential inaccuracy in disease specification and identification, very low response rates, and a high loss to follow-up rate. The current research integrated data from multiple data sources using the data linkage method. The medical records with accurate and specific for treatment information and administrative data from provincial health insurance plan with a comprehensive population coverage provide better information.   The medical records are more accurate and specific for treatment information, and administrative data from provincial health insurance plan with a comprehensive population coverage and medical services utilization. The availability of data from these data sources covering the entire study period allowed us to investigate the incidence and severity of the late effects in YACS, and to determine the risks of late effects in the survivors; as compared with the general population. Meanwhile, we identified subgroups of survivors among YACS at the highest risks for severe late effects, such as mortality, SMN and late morbidity leading to hospitalization. A better understanding of the overall late effects secondary to the type of cancer and its treatment can provide important insights into how one might design an applicable, reliable, and valid follow-up program to identify high risk groups among YACS, and to follow up on survivors to detect and reduce late effects.   118  The linkage rates (72% for this study [80]) were high compared with the participation rates for studies using conventional identification, recruitment, tracking, and data collection methods (e.g. recruitment and data collection rate = 61.4% in the CCSS  [77]). Furthermore, the rate of loss to follow-up (4% for survivors vs. 6% for the comparison group [80]) was low. The difference in response rates due to the different data collection methods may partially explain the discrepancy in results among studies.  The retrospective cohort study design allows us to assess the risks of multiple outcomes at the same time. Having tumour treatment information on the majority of subjects and the long follow-up time were also strengths of the study. It provides the advantage of examining many different risk factors simultaneously, and the potential to avoid selection bias.   The hypothetical nature of the willingness to participate in future genetic studies may raise some concerns on methodology issues related to the validity and reliability of the results. Therefore, we improved the validity of the results from several aspects. First, at the study design level, we randomly selected all prospective subjects from a population-based British Columbia Cancer Registry database. Each subject?s eligibility and contact information including both address and phone number was confirmed by the patients? last recorded physicians. Second,  at the questionnaire design level, questions were carefully chosen from the research question of interest, along with the literature and experts? opinions. Third, we conducted a pilot test among 12 volunteers in the same age range to ensure that the questions and methods were understandable and appropriately worded. Fourth, each prospective participant received a mailed survey package. If we did not receive any answer two weeks after the initial contact letter, we made three rounds of reminder phone calls to each participant. Fifth, although we could not receive direct responses for subjects that we were unable to contact, there was some information available from indirect sources, such as baseline data from the BC Cancer Registry, that we could use to compare the baseline characteristics of the non-responders with the information of the responders. It is possible that individuals replying to the survey may have different characteristics from those who ignored the survey requests or refused to participate. Comparing the differences in the baseline risk factors between responders and non-responders helped us to understand the factors influencing the participation and willingness rates. We found that individuals who returned the  119  questionnaire were more likely to be female, to be older than 50 years at the time of study, to have survived for more than 15 years after diagnosis, and to be living in more urban areas of BC.   7.2.2 Limitations There are limitations to any research study. However, great efforts were taken to minimize the impact of these limitations at every step in our studies.   In the mortality and SMN studies, one limitation was the missing data on treatment modality. Information regarding chemotherapy and surgery was missing for approximately 28% of the cases who had not been referred to any of the tertiary cancer centers, as we did not have access to information for cases treated outside this system. Therefore, the multiple imputation method for missing data was utilized under the assumption of missing at random [111].   Some limitations of this research are related to the record linkage methodology. Information on vital status and SMN were obtained retrospectively from administrative databases in British Columbia through data linkage. Therefore, we did not have access to events that may have occurred outside of Canada. In the late morbidity study, because only subjects actively on MSP coverage were included, information about subjects who had moved out of the province and might be healthier was not available, which may have led to overestimates for the morbidity for survivors.   This research used data from administrative databases, and, as always, the information in these databases was not originally collected for research purposes. As such, important risk factor information and some outcome data was not available. For example, health-behaviour-related factors, such as smoking and alcohol consumption, had been highlighted as the risk factors of many late effects in the literature, such as lung cancer and cardiovascular diseases [163]. However, information on these factors could not be obtained from either medical records or administrative databases, and therefore, could not be included in our analyses. Moreover, the existing data sources do not cover the full breadth of subjects? health status. Some outcomes related to survivors? subjective perceptions of the impact the cancer and its treatment had had on  120  the survivors? physical, psychological, and social aspects, such as quality of life, were not captured. These data sources are also not designed for survivorship research, and as such, a lot of components that we would like to have, such as lifestyle information, and regular patient-reported outcome assessment, are not available.   Sample size is always a concern for any population study. Because of the relatively small numbers of deceased and SMN cases (138 deceased cases and 62 SMN cases) observed, we did not have sufficient study power to assess treatment-related dose-response for mortality and SMN risks. In the survey study, only ethnicity and the willingness to provide income information had significant effects on the willingness to participate; whereas, in previous studies, other socio-demographic factors, such as education level, were also associated with the willingness to participate [144, 151-156]. One potential explanation for the discrepancy may be the small sample size (165 subjects) of our study.   Another limitation of the survey study is that participants who were not always willing to participate might be under represented in the  study. Among the subjects who ignored the survey request or refused to reply to the survey, a large proportion of them might have already made up their minds to not participate in genetic studies. We were unable to estimate the difference in willingness between participants who replied to the questionnaire and those who refused to reply to the questionnaire.   7.3 Impact, contribution and implications Even though the impacts of each study were described in each research chapter, taken together, this research has added several unique contributions to the literature.   Our research is the first study to investigate late effects among YACS (1248 for late mortality and SMN studies, 902 for late morbidity study). We observed higher risks of late effects of mortality, SMN and late morbidity among YACS. Most important, our findings emphasize the importance of future monitoring, screening, early detection, and eventually prevention among  121  this population. The results from this research have been shared with a broad audience in presentations and posters in academic conferences and in published papers in journals.   This study has identified a unique study population. This particular YACS population was identified by the CAYACS research team who recognized the distinctive characteristics of this cohort from late effects studies among childhood and adolescent cancer survivors. In this thesis, in Chapter 3-5, we examine the prevalence of late effects, emphasize the increased risks of these late effects on YACS as compared with those of the general population, and identify risk factors associated with survivors at high risk of late effects. The survey study in Chapter 6 highlights the feasibility of conducting future late effects studies among this population. We believe that these results will shape our understanding that a shift in care and support for young adult cancer survivors is necessary to meet their needs in living as healthy and active a life as possible.   Some novel aspects of this research are its multiple study designs, including a retrospective cohort and a cross-sectional survey, the data linkage method, and the usage of administrative data, medical records and survey data. The diversity in methods provides information on the effects of multiple outcomes [80], and strengthens the conclusions formed through a consistency across the results. This study supports the importance of conducting late effects studies of YACS to better meet the needs of this population. Combining administrative data and medical records in a population-based retrospective cohort study, the results from Chapters 3 to 5 demonstrate the increased risks of late effects in multiple health outcomes among YACS, and confirm the needs for research concerning long-term sequelae in YACS and the potential for late effects studies, including genetic studies, to help understanding these risks. When we explored the feasibility of conducting future late effects studies in Chapter 6, the results from the survey study highlighted that such research is feasible and would likely achieve sufficient participation for robust results, we also demonstrated the preferences in study design, study sponsorship and study method in this population. Therefore, the consistency across the results reinforces the fact that farther late effects research, with the goal to improve health outcomes among YACS, are essential and are likely feasible.   122  To our knowledge, the study presented in Chapter 6 was one of the first to examine willingness to participate in future genetic studies among YACS. Our findings confirmed that the majority of the respondents indicated that they would be willing to take part in future studies about genetic factors related to the risks of late effects, at least some of the time. It provided strong support for an interest in research among cancer survivors. We also found significant differences in willingness among responders, refusers, and non-responders in terms of demographic and disease-related factors. The socio-demographic related factors, especially ethnicity and income, might also play roles in participants? willingness. Our results might help researchers to improve response rates and help healthcare providers to communicate with YACS effectively when providing late effects related surveillance, screening and treatment.   The origins of the current research were grounded in the ongoing needs of YACS to achieve high-quality cancer survivorship care. Recently, the prominent cancer survivorship researcher Sir Michael Richards initiated ten research questions to improve health outcomes for cancer survivors [164]. These addressed the importance of understanding: 1) the number of cancer survivors; 2) the health and well-being of cancer survivors; 3) the problems and concerns of cancer survivors; 4) the consequences of treatment; 5) health service utilization; 6) perceptions and preferences for care; 7) interventions to enhance health outcomes; 8) service delivery levels; 9) priorities for future survivorship research; and 10) support for survivorship research. The findings of the current research have provided information for the YACS population related to several of these questions. This confirms that YACS are at increased risk of late effects and willing to participate in future late effects studies. Future studies could examine their preferences and perceptions in late effects, health services utilization, and direct health policy makers on how to meet cancer survivors? needs at the policy level.    In addition to their contributions to academic research, the findings from this work have health policy implications. Our results provide key directions on who is at risk, the size of the risk, the types of risks that exist, and the determinants of the risks. Key messages were: 1) YACS are at increased risks of late effects, including late mortality, SMN and late morbidity involving multiple organ systems; 2) the risks were varied by the type of primary diagnosis; 3) survivors with a later year of diagnosis and receipt of radiation therapy had the highest risks of late  123  mortality and SMN; 4) the risk of late morbidity was higher among survivors receiving three treatment modalities; 5) most survivors are willing to participate in future late effects studies, especially genetic studies, but fewer survivors are willing to participate in studies requiring blood sample donations, sponsored by a drug company, or needing several visits.   Knowledge translation is also important to promote our research findings and widespread the support of key late effects issues for cancer survivors. The results from this research have been presented in different conference posters, presentations, and manuscripts [165, 166]. A website for childhood, adolescent and young adult cancer survivors from the CAYACS program (http://www.cayacs.ca/) was established to provide up-to-date research results not only for cancer survivors and their families, but also for researchers, healthcare providers, and health policymakers, and my work is a part of the results presented.   7.4 Future research  This research yielded many novel results that increase our understanding of late effects among YACS and which explore the feasibility of conducting future survey studies. But there are still many research questions of interest that need to be explored further. It is our hope that the challenges faced in this work may pave the way for future research.   This research was able to utilize a relatively long follow-up time. Results in Chapters 3 and 4 were based on up to 33 years follow-up among YACS diagnosed from 1970 and results in Chapter 5 were based on up to 21 years among YACS diagnosed from 1981. It is believed that some exposures may have cumulative effects that are expressed over long periods of time, and some late-onset effects may occur many years after the original treatment. A few pediatric studies have gathered data from survivors diagnosed before 1970. However, cancer treatment has improved dramatically since the 1970s and, as a result, the late effects have changed as well. For example, the overall mortality rate has decreased from 11-18% for people treated before 1970 [73, 74] to 9-13% for those treated after 1970 [75, 76]. Therefore, the information obtained from survivors before the 1970s may not be applicable to survivors following this period. In addition, it is also possible that changes in therapy over recent years, such as conformal irradiation and  124  laparoscopy in surgical techniques, may be associated with a different spectrum of late effects in young adult cancer survivors than those in our current study cohort. Hence, it is imperative that we continue to prospectively study survivors of young adult cancer with ongoing long-term follow-up to identify potential late toxicity related to their disease and therapy.   As previously mentioned in the limitations, despite the fact that the YACS cohort utilized in our research was large compared with that of other studies (N=1248 in Chapter 3 & 4; N = 902 in Chapter 5), the numbers of cases observed with certain outcomes, such as death (N=138) and SMN (N=62 ), were small. Therefore, future studies with larger sample sizes are needed to better understand late effects. Because of the relatively low incidence of cancer in the young adult population compared to its overall incidence in the older adult population, the number of cancer patients is limited. A combined research effort across regions and institutions is therefore needed to gain a large, diverse, and well-characterized cohort of long-term YACS. This approach has been used in other studies, such as the CCSS, to serve as a resource for identifying important late effects in childhood and adolescent cancer survivors. Future studies utilizing multi-institutional collaborations for YACS are needed to assess the treatment-related dose-response for late effects, and the late effects according to type of cancer.   Given the limitations of administrative data, other research studies designed to characterize the risks of related behavioral factors and assess survivors? subjective perceptions are necessary. Therefore, future studies with direct survivor contact collecting information, including both self-reports and biological samples, are recommended. Other studies, such as an examination of  survivors? preferences for follow-up care and preventive treatment, may contribute to healthcare practice.  7.5 Conclusion Although a great deal of research has been done with the survivors of pediatric and adult cancers, late effects studies among YACS are a relatively new research area. This research identified the high prevalence of multiple and significant late effects among YACS, and it also supported the feasibility of conducting late effects studies in the future. Evidence from this work points to the  125  importance of careful monitoring for late effects, and for the need to develop effective clinical programs and guidelines to meet the needs of this population.  126  Bibliography 1. Bleyer WA. Cancer in older adolescents and young adults: epidemiology, diagnosis, treatment, survival, and importance of clinical trials. Med Pediatr Oncol 2002;38:1-10. 2. Cancer Care Ontario: Cancer in Young Adult in Canada. Toronto, Canada, 2010. 3. Bleyer A, Viny A, and Barr R. Cancer in 15- to 29-year-olds by primary site. Oncologist 2006;11:590-601. 4. Bleyer A, Barr R. Cancer in Adolescents and Young Adults. 1st ed. Springer Berlin Heidelberg New York, 2007. 5. Bleyer A. Adolescent and young adult (AYA) oncology: the first A. Pediatr Hematol Oncol 2007;24:325-36. 6. Haimi M, Peretz Nahum M, and Ben Arush MW. Delay in diagnosis of children with cancer: a retrospective study of 315 children. Pediatr Hematol Oncol 2004;21:37-48. 7. Canadian Childhood Cancer Surveillance and Control Program (2004) Diagnosis and Initial Treatment of Cancer in Canadian Adolescents 15 to 19 Years, 1995-2000, Health Canada: Ottawa, Canada. 8. Pollock BH, Krischer JP, and Vietti TJ. Interval between symptom onset and diagnosis of pediatric solid tumors. J Pediatr 1991;119:725-32. 9. Saha V, Love S, Eden T, et al. Determinants of symptom interval in childhood cancer. Arch Dis Child 1993;68:771-4. 10. Carvalho AL, Pintos J, Schlecht NF, et al. Predictive factors for diagnosis of advanced-stage squamous cell carcinoma of the head and neck. Arch Otolaryngol Head Neck Surg 2002;128:313-8. 11. Boissel N, Auclerc MF, Lheritier V, et al. Should adolescents with acute lymphoblastic leukemia be treated as old children or young adults? Comparison of the French FRALLE-93 and LALA-94 trials. J Clin Oncol 2003;21:774-80. 12. de Bont JM, Holt B, Dekker AW, et al. Significant difference in outcome for adolescents with acute lymphoblastic leukemia treated on pediatric vs adult protocols in the Netherlands. Leukemia 2004;18:2032-5. 13. Nachman J, Sather HN, Buckley JD, et al. Young adults 16-21 years of age at diagnosis entered on Childrens Cancer Group acute lymphoblastic leukemia and acute myeloblastic leukemia protocols. Results of treatment. Cancer 1993;71:3377-85. 14. Klein-Geltink J, Shaw AK, Morrison HI, et al. Use of paediatric versus adult oncology treatment centres by adolescents 15-19 years old: the Canadian Childhood Cancer Surveillance and Control Program. Eur J Cancer 2005;41:404-10. 15. Barr RD, Greenberg ML. Cancer surveillance and control in adolescents--similarities and contrasts between Canada and the United States. Pediatr Blood Cancer 2006;46:273-7. 16. Oeffinger KC, Mertens AC, Sklar CA, et al. Chronic health conditions in adult survivors of childhood cancer. N Engl J Med 2006;355:1572-82. 17. Mertens AC, Liu Q, Neglia JP, et al. Cause-specific late mortality among 5-year survivors of childhood cancer: the Childhood Cancer Survivor Study. J Natl Cancer Inst 2008;100:1368-79. 18. Armstrong GT, Liu Q, Yasui Y, et al. Late mortality among 5-year survivors of childhood cancer: a summary from the Childhood Cancer Survivor Study. J Clin Oncol 2009;27:2328-38.  127  19. Garwicz S, Anderson H, Olsen JH, et al. Late and very late mortality in 5-year survivors of childhood cancer: changing pattern over four decades--experience from the Nordic countries. Int J Cancer 2012;131:1659-66. 20. Cardous-Ubbink MC, Heinen RC, Langeveld NE, et al. Long-term cause-specific mortality among five-year survivors of childhood cancer. Pediatr Blood Cancer 2004;42:563-73. 21. Lawless SC, Verma P, Green DM, et al. Mortality experiences among 15+ year survivors of childhood and adolescent cancers. Pediatr Blood Cancer 2007;48:333-8. 22. MacArthur AC, Spinelli JJ, Rogers PC, et al. Mortality among 5-year survivors of cancer diagnosed during childhood or adolescence in British Columbia, Canada. Pediatr Blood Cancer 2007;48:460-7. 23. Mertens AC. Cause of mortality in 5-year survivors of childhood cancer. Pediatr Blood Cancer 2007;48:723-6. 24. Armstrong GT, Pan Z, Ness KK, et al. Temporal trends in cause-specific late mortality among 5-year survivors of childhood cancer. J Clin Oncol 28:1224-31. 25. Cardous-Ubbink MC, Heinen RC, Bakker PJ, et al. Risk of second malignancies in long-term survivors of childhood cancer. Eur J Cancer 2007;43:351-62. 26. de Vathaire F, Hawkins M, Campbell S, et al. Second malignant neoplasms after a first cancer in childhood: temporal pattern of risk according to type of treatment. Br J Cancer 1999;79:1884-93. 27. Friedman DL, Whitton J, Leisenring W, et al. Subsequent neoplasms in 5-year survivors of childhood cancer: the Childhood Cancer Survivor Study. J Natl Cancer Inst 2010;102:1083-95. 28. Hammal DM, Bell CL, Craft AW, et al. Second primary tumors in children and young adults in the North of England (1968-99). Pediatr Blood Cancer 2005;45:155-61. 29. Oeffinger KC, Tonorezos ES. The cancer is over, now what?: Understanding risk, changing outcomes. Cancer 2011;117:2250-7. 30. Hodgson DC. Late effects in the era of modern therapy for Hodgkin lymphoma. Hematology Am Soc Hematol Educ Program 2011;2011:323-9. 31. Aleman BM, van den Belt-Dusebout AW, De Bruin ML, et al. Late cardiotoxicity after treatment for Hodgkin lymphoma. Blood 2007;109:1878-86. 32. Myrehaug S, Pintilie M, Tsang R, et al. Cardiac morbidity following modern treatment for Hodgkin lymphoma: supra-additive cardiotoxicity of doxorubicin and radiation therapy. Leuk Lymphoma 2008;49:1486-93. 33. Hull MC, Morris CG, Pepine CJ, et al. Valvular dysfunction and carotid, subclavian, and coronary artery disease in survivors of hodgkin lymphoma treated with radiation therapy. JAMA 2003;290:2831-7. 34. Glanzmann C, Kaufmann P, Jenni R, et al. Cardiac risk after mediastinal irradiation for Hodgkin's disease. Radiother Oncol 1998;46:51-62. 35. Bhatia S, Landier W, Toogood AA, et al., Adolescent and Young Adult Cancer Survivors: Late Effects of Treatment, in Cancer in Adolescents and Young Adults, A. Bleyer and R.D. Barr, Editors. 2007, Springer Berlin Heidelberg New York. p. 411-430. 36. Barnes ND. Effects of external irradiation on the thyroid gland in childhood. Horm Res 1988;30:84-9. 37. Hancock SL, Cox RS, and McDougall IR. Thyroid diseases after treatment of Hodgkin's disease. N Engl J Med 1991;325:599-605.  128  38. Sklar C, Whitton J, Mertens A, et al. Abnormalities of the thyroid in survivors of Hodgkin's disease: data from the Childhood Cancer Survivor Study. J Clin Endocrinol Metab 2000;85:3227-32. 39. van Dorp W, van Beek RD, Laven JS, et al. Long-term endocrine side effects of childhood Hodgkin's lymphoma treatment: a review. Hum Reprod Update 2012;18:12-28. 40. Diller L, Chow EJ, Gurney JG, et al. Chronic disease in the Childhood Cancer Survivor Study cohort: a review of published findings. J Clin Oncol 2009;27:2339-55. 41. van Beek RD, van den Heuvel-Eibrink MM, Hakvoort-Cammel FG, et al. Bone mineral density, growth, and thyroid function in long-term survivors of pediatric Hodgkin's lymphoma treated with chemotherapy only. J Clin Endocrinol Metab 2009;94:1904-9. 42. Howell SJ, Shalet SM. Effect of cancer therapy on pituitary-testicular axis. Int J Androl 2002;25:269-76. 43. Waxman JH, Terry YA, Wrigley PF, et al. Gonadal function in Hodgkin's disease: long-term follow-up of chemotherapy. Br Med J (Clin Res Ed) 1982;285:1612-3. 44. Clark ST, Radford JA, Crowther D, et al. Gonadal function following chemotherapy for Hodgkin's disease: a comparative study of MVPP and a seven-drug hybrid regimen. J Clin Oncol 1995;13:134-9. 45. Heikens J, Behrendt H, Adriaanse R, et al. Irreversible gonadal damage in male survivors of pediatric Hodgkin's disease. Cancer 1996;78:2020-4. 46. Sieniawski M, Reineke T, Josting A, et al. Assessment of male fertility in patients with Hodgkin's lymphoma treated in the German Hodgkin Study Group (GHSG) clinical trials. Ann Oncol 2008;19:1795-801. 47. van Beek RD, Smit M, van den Heuvel-Eibrink MM, et al. Inhibin B is superior to FSH as a serum marker for spermatogenesis in men treated for Hodgkin's lymphoma with chemotherapy during childhood. Hum Reprod 2007;22:3215-22. 48. Viviani S, Bonfante V, Santoro A, et al. Long-term results of an intensive regimen: VEBEP plus involved-field radiotherapy in advanced Hodgkin's disease. Cancer J Sci Am 1999;5:275-82. 49. van der Kaaij MA, Heutte N, Le Stang N, et al. Gonadal function in males after chemotherapy for early-stage Hodgkin's lymphoma treated in four subsequent trials by the European Organisation for Research and Treatment of Cancer: EORTC Lymphoma Group and the Groupe d'Etude des Lymphomes de l'Adulte. J Clin Oncol 2007;25:2825-32. 50. Byrne J. Infertility and premature menopause in childhood cancer survivors. Med Pediatr Oncol 1999;33:24-8. 51. Byrne J, Fears TR, Gail MH, et al. Early menopause in long-term survivors of cancer during adolescence. Am J Obstet Gynecol 1992;166:788-93. 52. Sanders JE, Buckner CD, Amos D, et al. Ovarian function following marrow transplantation for aplastic anemia or leukemia. J Clin Oncol 1988;6:813-8. 53. Gleeson HK, Darzy K, and Shalet SM. Late endocrine, metabolic and skeletal sequelae following treatment of childhood cancer. Best Practice & Research Clinical Endocrinology & Metabolism. 2002;16:335-48. 54. Muller J. Disturbance of pubertal development after cancer treatment. Best Pract Res Clin Endocrinol Metab 2002;16:91-103.  129  55. Chow EJ, Friedman DL, Yasui Y, et al. Timing of menarche among survivors of childhood acute lymphoblastic leukemia: a report from the Childhood Cancer Survivor Study. Pediatr Blood Cancer 2008;50:854-8. 56. Schmiegelow M, Hertz H, Schmiegelow K, et al. Insulin-like growth factor-I and insulin-like growth factor binding protein-3 during maintenance chemotherapy of acute lymphoblastic leukemia in children. J Pediatr Hematol Oncol 1999;21:268-73. 57. Geller BM, Mace J, Vacek P, et al. Are cancer survivors willing to participate in research? J Community Health 2011;36:772-8. 58. Pakilit AT, Kahn BA, Petersen L, et al. Making effective use of tumor registries for cancer survivorship research. Cancer 2001;92:1305-14. 59. Smith T, Stein KD, Mehta CC, et al. The rationale, design, and implementation of the American Cancer Society's studies of cancer survivors. Cancer 2007;109:1-12. 60. Harlan LC, Lynch CF, Keegan TH, et al. Recruitment and follow-up of adolescent and young adult cancer survivors: the AYA HOPE Study. J Cancer Surviv 2011;5:305-14. 61. Svensson K, Ramirez OF, Peres F, et al. Socioeconomic determinants associated with willingness to participate in medical research among a diverse population. Contemp Clin Trials 2012;33:1197-205. 62. Haynes MA SB. The Unequal Burden of Cancer: An Assessment of NIH Research and Programs for Ethnic Minorities and the Medically Underserved. 2010/09/17ed., Washington (DC): National Academies Press (US): 1999. 63. Woodward E, Jessop M, Glaser A, et al. Late effects in survivors of teenage and young adult cancer: does age matter? Ann Oncol 22:2561-8. 64. Bleyer A. The adolescent and young adult gap in cancer care and outcome. Curr Probl Pediatr Adolesc Health Care 2005;35:182-217. 65. Lorenzi MF, Xie L, Rogers PC, et al. Hospital-related morbidity among childhood cancer survivors in British Columbia, Canada: report of the childhood, adolescent, young adult cancer survivors (CAYACS) program. Int J Cancer 128:1624-31. 66. Bradley NM, Lorenzi MF, Abanto Z, et al. Hospitalisations 1998-2000 in a British Columbia population-based cohort of young cancer survivors: report of the Childhood/Adolescent/Young Adult Cancer Survivors (CAYACS) Research Program. Eur J Cancer 2010;46:2441-8. 67. Kurt BA, Nolan VG, Ness KK, et al. Hospitalization rates among survivors of childhood cancer in the Childhood Cancer Survivor Study cohort. Pediatr Blood Cancer 2012;59:126-32. 68. Armenian SH, Bhatia S. Chronic health conditions in childhood cancer survivors: is it all treatment-related--or do genetics play a role? J Gen Intern Med 2009;24 Suppl 2:S395-400. 69. Geenen MM, Cardous-Ubbink MC, Kremer LC, et al. Medical assessment of adverse health outcomes in long-term survivors of childhood cancer. Jama 2007;297:2705-15. 70. Mulrooney DA, Yeazel MW, Kawashima T, et al. Cardiac outcomes in a cohort of adult survivors of childhood and adolescent cancer: retrospective analysis of the Childhood Cancer Survivor Study cohort. Bmj 2009;339:b4606. 71. Armstrong GT, Sklar CA, Hudson MM, et al. Long-term health status among survivors of childhood cancer: does sex matter? J Clin Oncol 2007;25:4477-89. 72. Pollock BH, Birch JM. Registration and classification of adolescent and young adult cancer cases. Pediatr Blood Cancer 2008;50:1090-3.  130  73. Hawkins MM, Kingston JE, and Kinnier Wilson LM. Late deaths after treatment for childhood cancer. Arch Dis Child 1990;65:1356-63. 74. Li FP, Myers MH, Heise HW, et al. The course of five-year survivors of cancer in childhood. J Pediatr 1978;93:185-7. 75. Robertson CM, Hawkins MM, and Kingston JE. Late deaths and survival after childhood cancer: implications for cure. Bmj 1994;309:162-6. 76. Mertens AC, Yasui Y, Neglia JP, et al. Late mortality experience in five-year survivors of childhood and adolescent cancer: the Childhood Cancer Survivor Study. J Clin Oncol 2001;19:3163-72. 77. Robison LL, Armstrong GT, Boice JD, et al. The Childhood Cancer Survivor Study: a National Cancer Institute-supported resource for outcome and intervention research. J Clin Oncol 2009;27:2308-18. 78. Olsen JH, Moller T, Anderson H, et al. Lifelong cancer incidence in 47,697 patients treated for childhood cancer in the Nordic countries. J Natl Cancer Inst 2009;101:806-13. 79. Jazbec J, Ecimovic P, and Jereb B. Second neoplasms after treatment of childhood cancer in Slovenia. Pediatr Blood Cancer 2004;42:574-81. 80. McBride ML, Rogers PC, Sheps SB, et al. Childhood, adolescent, and young adult cancer survivors research program of British Columbia: objectives, study design, and cohort characteristics. Pediatr Blood Cancer 2010;55:324-30. 81. MacArthur AC, Spinelli JJ, Rogers PC, et al. Risk of a second malignant neoplasm among 5-year survivors of cancer in childhood and adolescence in British Columbia, Canada. Pediatr Blood Cancer 2007;48:453-9. 82. Tukenova M, Guibout C, Oberlin O, et al. Role of cancer treatment in long-term overall and cardiovascular mortality after childhood cancer. J Clin Oncol 2010;28:1308-15. 83. Moller TR, Garwicz S, Barlow L, et al. Decreasing late mortality among five-year survivors of cancer in childhood and adolescence: a population-based study in the Nordic countries. J Clin Oncol 2001;19:3173-81. 84. Kaldor JM, Day NE, Clarke EA, et al. Leukemia following Hodgkin's disease. N Engl J Med 1990;322:7-13. 85. Tucker MA, Coleman CN, Cox RS, et al. Risk of second cancers after treatment for Hodgkin's disease. N Engl J Med 1988;318:76-81. 86. Robison LL, Bhatia S. Late-effects among survivors of leukaemia and lymphoma during childhood and adolescence. Br J Haematol 2003;122:345-59. 87. Garwicz S, Anderson H, Olsen JH, et al. Second malignant neoplasms after cancer in childhood and adolescence: a population-based case-control study in the 5 Nordic countries. The Nordic Society for Pediatric Hematology and Oncology. The Association of the Nordic Cancer Registries. Int J Cancer 2000;88:672-8. 88. Swerdlow AJ, Barber JA, Hudson GV, et al. Risk of second malignancy after Hodgkin's disease in a collaborative British cohort: the relation to age at treatment. J Clin Oncol 2000;18:498-509. 89. Travis LB, Gospodarowicz M, Curtis RE, et al. Lung cancer following chemotherapy and radiotherapy for Hodgkin's disease. J Natl Cancer Inst 2002;94:182-92. 90. Birdwell SH, Hancock SL, Varghese A, et al. Gastrointestinal cancer after treatment of Hodgkin's disease. Int J Radiat Oncol Biol Phys 1997;37:67-73.  131  91. Swerdlow AJ, Higgins CD, Smith P, et al. Second cancer risk after chemotherapy for Hodgkin's lymphoma: a collaborative British cohort study. J Clin Oncol 2011;29:4096-104. 92. Metayer C, Lynch CF, Clarke EA, et al. Second cancers among long-term survivors of Hodgkin's disease diagnosed in childhood and adolescence. J Clin Oncol 2000;18:2435-43. 93. Neglia JP, Friedman DL, Yasui Y, et al. Second malignant neoplasms in five-year survivors of childhood cancer: childhood cancer survivor study. J Natl Cancer Inst 2001;93:618-29. 94. Horwich A, Swerdlow AJ. Second primary breast cancer after Hodgkin's disease. Br J Cancer 2004;90:294-8. 95. Bhatia S, Yasui Y, Robison LL, et al. High risk of subsequent neoplasms continues with extended follow-up of childhood Hodgkin's disease: report from the Late Effects Study Group. J Clin Oncol 2003;21:4386-94. 96. van Leeuwen FE, Klokman WJ, Veer MB, et al. Long-term risk of second malignancy in survivors of Hodgkin's disease treated during adolescence or young adulthood. J Clin Oncol 2000;18:487-97. 97. Ng AK, Bernardo MV, Weller E, et al. Second malignancy after Hodgkin disease treated with radiation therapy with or without chemotherapy: long-term risks and risk factors. Blood 2002;100:1989-96. 98. Travis LB, Hill DA, Dores GM, et al. Breast cancer following radiotherapy and chemotherapy among young women with Hodgkin disease. Jama 2003;290:465-75. 99. Inskip PD, Robison LL, Stovall M, et al. Radiation dose and breast cancer risk in the childhood cancer survivor study. J Clin Oncol 2009;27:3901-7. 100. Sigurdson AJ, Ronckers CM, Mertens AC, et al. Primary thyroid cancer after a first tumour in childhood (the Childhood Cancer Survivor Study): a nested case-control study. Lancet 2005;365:2014-23. 101. O'Brien MM, Donaldson SS, Balise RR, et al. Second malignant neoplasms in survivors of pediatric Hodgkin's lymphoma treated with low-dose radiation and chemotherapy. J Clin Oncol 2010;28:1232-9. 102. Bergmann MM, Calle EE, Mervis CA, et al. Validity of self-reported cancers in a prospective cohort study in comparison with data from state cancer registries. Am J Epidemiol 1998;147:556-62. 103. Bergmann MM, Byers T, Freedman DS, et al. Validity of self-reported diagnoses leading to hospitalization: a comparison of self-reports with hospital records in a prospective study of American adults. Am J Epidemiol 1998;147:969-77. 104. Messite J, Stellman SD. Accuracy of death certificate completion: the need for formalized physician training. Jama 1996;275:794-6. 105. Smith Sehdev AE, Hutchins GM. Problems with proper completion and accuracy of the cause-of-death statement. Arch Intern Med 2001;161:277-84. 106. Dama E, Pastore G, Mosso ML, et al. Late deaths among five-year survivors of childhood cancer. A population-based study in Piedmont Region, Italy. Haematologica 2006;91:1084-91. 107. Reulen RC, Winter DL, Frobisher C, et al. Long-term cause-specific mortality among survivors of childhood cancer. Jama 304:172-9.  132  108. Barr RD, Holowaty EJ, and Birch JM. Classification schemes for tumors diagnosed in adolescents and young adults. Cancer 2006;106:1425-30. 109. World Health Organization. International classification of diseases: Manual of the international classification of diseases, injuries and causes of death, ninth revision. Geneva, Switzerland. 1979, WHO. 110. Fritz A, Percy C, Jack A, et al., eds. International Classification of Diseases for Oncology, 3rd edn. 3rd editioned. Geneva, Switzerland, World Health Organization: 2000. 111. Little RJA, Rubin DB. Statistical Analysis with Missing Data. 2nd ed. John Wiley & Sons, 2002. 112. Moller TR, Garwicz S, Perfekt R, et al. Late mortality among five-year survivors of cancer in childhood and adolescence. Acta Oncol 2004;43:711-8. 113. Greenland S. Basic methods for sensitivity analysis of biases. Int J Epidemiol 1996;25:1107-16. 114. SEER Cancer Statistics Review, 1975-2008, National Cancer Institute. Bethesda, MD, http://seer.cancer.gov/csr/1975_2008/, based on November 2010 SEER data submission, posted to the SEER web site, 2011.,  115. Cancer Care Ontario: Cancer in Young Adults in Canada. Toronto, Canada: 2006. 116. Soliman H, Agresta SV. Current issues in adolescent and young adult cancer survivorship. Cancer Control 2008;15:55-62. 117. Bhatia S, Robison LL, Oberlin O, et al. Breast cancer and other second neoplasms after childhood Hodgkin's disease. N Engl J Med 1996;334:745-51. 118. Travis LB, Hill D, Dores GM, et al. Cumulative absolute breast cancer risk for young women treated for Hodgkin lymphoma. J Natl Cancer Inst 2005;97:1428-37. 119. Dores GM, Metayer C, Curtis RE, et al. Second malignant neoplasms among long-term survivors of Hodgkin's disease: a population-based evaluation over 25 years. J Clin Oncol 2002;20:3484-94. 120. Bhatia S, Sather HN, Pabustan OB, et al. Low incidence of second neoplasms among children diagnosed with acute lymphoblastic leukemia after 1983. Blood 2002;99:4257-64. 121. Fritz A, Percy C, Jack A, et al., eds. International Classification of Diseases for Oncology. 3rded. Geneva, Switzerland, World Health Organization: 2000. 122. Fine J, Gray R. A Proportional Hazards Model for the Subdistribution of a Competing Risk. J Amer Stat Assoc 1999;94:496-509. 123. Wilson CL, Cohn RJ, Johnston KA, et al. Late mortality and second cancers in an Australian cohort of childhood cancer survivors. Med J Aust 193:258-61. 124. Kenney LB, Yasui Y, Inskip PD, et al. Breast cancer after childhood cancer: a report from the Childhood Cancer Survivor Study. Ann Intern Med 2004;141:590-7. 125. Ng AK, Kenney LB, Gilbert ES, et al. Secondary malignancies across the age spectrum. Semin Radiat Oncol 20:67-78. 126. Landier W, Bhatia S, Eshelman DA, et al. Development of risk-based guidelines for pediatric cancer survivors: the Children's Oncology Group Long-Term Follow-Up Guidelines from the Children's Oncology Group Late Effects Committee and Nursing Discipline. J Clin Oncol 2004;22:4979-90. 127. Oeffinger KC, Ford JS, Moskowitz CS, et al. Breast cancer surveillance practices among women previously treated with chest radiation for a childhood cancer. Jama 2009;301:404-14.  133  128. Long-Term Follow-Up Guidelines for Survivors of Childhood, Adolescent, and Young Adult Cancer. Children's Oncology Group, 2008. 129. Bassal M, Mertens AC, Taylor L, et al. Risk of selected subsequent carcinomas in survivors of childhood cancer: a report from the Childhood Cancer Survivor Study. J Clin Oncol 2006;24:476-83. 130. Jenkinson HC, Hawkins MM, Stiller CA, et al. Long-term population-based risks of second malignant neoplasms after childhood cancer in Britain. Br J Cancer 2004;91:1905-10. 131. Childhood Cancer Survivorship: Improving Care and Quality of Life, ed. M. Hewitt, S.L. Weiner, and J.V. Simone. The National Academies Press, 2003. 132. Bleyer A. Older adolescents with cancer in North America deficits in outcome and research. Pediatr Clin North Am 2002;49:1027-42. 133. International statistical classification of diseases and related health problems. 1992, World Health Organization: Geneva. 134. Wilkins R, PCCF+  Version 4J User's Guide. Automated geographic coding based on the Statistics Canada Postal Code Conversion files, including postal codes to September 2006., in Catalogue 82F0086-XDB. 2007, Statistics Canada: Ottawa. 135. Rambeau. S, K T, Census metropolitan area and census agglomeration influenced zones (MIZ) with census data. 2000, Geography Division, Statistics Canada: Ottawa. 136. National Cancer Institute(US), Common Terminology Criteria for Adverse Events v3.0 (CTCAE). 2003, National Cancer Institute. 137. Garre ML, Gandus S, Cesana B, et al. Health status of long-term survivors after cancer in childhood. Results of an uniinstitutional study in Italy. Am J Pediatr Hematol Oncol 1994;16:143-52. 138. Stevens MC, Mahler H, and Parkes S. The health status of adult survivors of cancer in childhood. Eur J Cancer 1998;34:694-8. 139. Oeffinger KC, Eshelman DA, Tomlinson GE, et al. Grading of late effects in young adult survivors of childhood cancer followed in an ambulatory adult setting. Cancer 2000;88:1687-95. 140. Hudson MM, Mertens AC, Yasui Y, et al. Health status of adult long-term survivors of childhood cancer: a report from the Childhood Cancer Survivor Study. JAMA 2003;290:1583-92. 141. Oeffinger KC, Mertens AC, Sklar CA, et al. Obesity in adult survivors of childhood acute lymphoblastic leukemia: a report from the Childhood Cancer Survivor Study. J Clin Oncol 2003;21:1359-65. 142. Canadian Cancer Society's Steering Committee: Canadian Cancer Statistics 2009. Toronto: Canadian Cancer Society, 2009. 143. Bleyer A. Young adult oncology: the patients and their survival challenges. CA Cancer J Clin 2007;57:242-55. 144. Murphy EJ, Wickramaratne P, and Weissman MM. Racial and ethnic differences in willingness to participate in psychiatric genetic research. Psychiatr Genet 2009;19:186-94. 145. Katz RV, Green BL, Kressin NR, et al. Willingness of minorities to participate in biomedical studies: confirmatory findings from a follow-up study using the Tuskegee Legacy Project Questionnaire. J Natl Med Assoc 2007;99:1052-60. 146. Teschke K, Marino S, Chu R, et al. Public opinions about participating in health research. Can J Public Health 2010;101:159-64.  134  147. Maayan-Metzger A, Kedem-Friedrich P, and Kuint J. Motivations of mothers to enroll their newborn infants in general clinical research on well-infant care and development. Pediatrics 2008;121:e590-6. 148. Tait AR, Voepel-Lewis T, Siewert M, et al. Factors that influence parents' decisions to consent to their child's participation in clinical anesthesia research. Anesth Analg 1998;86:50-3. 149. Katz RV, Kegeles SS, Kressin NR, et al. The Tuskegee Legacy Project: willingness of minorities to participate in biomedical research. J Health Care Poor Underserved 2006;17:698-715. 150. Shavers VL, Lynch CF, and Burmeister LF. Knowledge of the Tuskegee study and its impact on the willingness to participate in medical research studies. J Natl Med Assoc 2000;92:563-72. 151. Crowley R, Casarett D. Patients' willingness to participate in symptom-related and disease-modifying research: results of a research screening initiative in a palliative care clinic. Cancer 2003;97:2327-33. 152. Brody JL, Annett RD, Scherer DG, et al. Comparisons of adolescent and parent willingness to participate in minimal and above-minimal risk pediatric asthma research protocols. J Adolesc Health 2005;37:229-35. 153. Katz RV, Kegeles SS, Kressin NR, et al. Awareness of the Tuskegee Syphilis Study and the US presidential apology and their influence on minority participation in biomedical research. Am J Public Health 2008;98:1137-42. 154. Klug SJ, Hetzer M, and Blettner M. Screening for breast and cervical cancer in a large German city: participation, motivation and knowledge of risk factors. Eur J Public Health 2005;15:70-7. 155. St. Jude Children's Research Hospital. Long-term Follow-up Study. [survey questionnaire]. Available at: http://ltfu.stjude.org/. Accessed 2008.  156. Blatt J, Olshan A, Gula MJ, et al. Second malignancies in very-long-term survivors of childhood cancer. Am J Med 1992;93:57-60. 157. Likert R. A technique for the measurement of attitudes. Archives of Psychology 1932;1-55. 158. Teschke K, Chu R, Marino S, et al. Understanding attitudes towards health research, or "Who can phone you at dinnertime?". in EPICOH, Banff. 2007. 159. Bottomley SJ, Kassner E. Late effects of childhood cancer therapy. J Pediatr Nurs 2003;18:126-33. 160. Doescher MP, Saver BG, Franks P, et al. Racial and ethnic disparities in perceptions of physician style and trust. Arch Fam Med 2000;9:1156-63. 161. Gotay CC, Farley JH, Kawamoto CT, et al. Adaptation and quality of life among long-term cervical cancer survivors in the military health care system. Mil Med 2008;173:1035-41. 162. Hudson MM, Ness KK, Nolan VG, et al. Prospective medical assessment of adults surviving childhood cancer: study design, cohort characteristics, and feasibility of the St. Jude Lifetime Cohort study. Pediatr Blood Cancer 2011;56:825-36. 163. Krull KR, Huang S, Gurney JG, et al. Adolescent behavior and adult health status in childhood cancer survivors. J Cancer Surviv 2010;4:210-7.  135  164. Richards M, Corner J, and Maher J. The National Cancer Survivorship Initiative: new and emerging evidence on the ongoing needs of cancer survivors. Br J Cancer 2011;105 Suppl 1:S1-4. 165. Zhang Y, Lorenzi MF, Goddard K, et al. Late morbidity leading to hospitalization among 5-year survivors of young adult cancer: A report of the childhood, adolescent, and young adult cancer survivors (CAYACS) research program. Int J Cancer 2013. 166. Zhang Y, Goddard K, Spinelli JJ, et al. Risk of Late Mortality and Second Malignant Neoplasms among 5-Year Survivors of Young Adult Cancer: A Report of the Childhood, Adolescent, and Young Adult Cancer Survivors Research Program. J Cancer Epidemiol 2012;2012:103032.    136  Appendices Appendix A  AYA Group Classification AYA Group Classification 1. Leukemias 2. Lymphomas 3. Central nervous system tumors 4. bone tumours 5. Soft tissue sarcomas 6. Germ cell tumours 7. Melanoma and skin carcinoma 8. Carcinomas (except of skin) 9. Miscellaneous specified neoplasms NEC 10. Unspecified malignant neoplasms NEC  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0107198/manifest

Comment

Related Items