The global burden of adolescent and young adult cancer in 2019: a systematic analysis for the Global Burden of Disease Study 2019

1Adolescents and young adults represent a heterogenous population consisting of individuals aged 15–39 years.,, This formative time in life is unique, with several physical, emotional, and psychosocial changes, and with individuals potentially beginning or advancing their careers, higher education, relationships, and having children. The definitions and cutoffs of the age range for adolescents and young adults vary, but this age group is generally described as a subpopulation that is in transition between childhood and older adulthood.

2Adolescents and young adults develop cancers commonly found and treated in the paediatric population as well as the more common carcinomas seen in adults., Additionally, some cancers are more prevalent in this age group than in younger or older individuals, such as Hodgkin lymphoma and gonadal germ cell tumours., As a consequence, from a health-care delivery perspective, adolescent and young adult patients with cancer might struggle to find care that is optimal for both their cancer type and their age-related treatment needs. Additionally, adolescent and young adult patients often face social and financial challenges, which might result in inequities in access to appropriate care, timely diagnosis, and treatment.,, Although adolescents and young adults have not seen the same improvements in cancer survival as younger and older cohorts for certain cancers, including acute myeloid leukaemia and soft tissue sarcomas, this population has not historically been a major focus of cancer control programmes and research development. Instead, based on historical precedent, adolescents and young adults are often grouped with adult patients in clinical care and clinical trials, and, as a consequence, comprehensive assessments of the cancer burden and epidemiological patterns in this age group are largely unknown or unreported in many settings.

3Evidence before this study Adolescents and young adults with cancer represent a transition population within the cancer continuum between children and older adults. As adolescents and young adults with cancer are treated by a variety of specialists, their unique epidemiology and clinical care needs are often overlooked. Although improvements in survival for children and adults with cancer are reported in high-income countries, less incremental progress has been observed among adolescents and young adults. Added complexities of cancer in this age group include the potential impact of a cancer diagnosis on starting or caring for their families and careers, access to care, diagnostic delays, and abandonment of therapy—issues that exist globally. Previous work assessing the global burden of adolescent and young adult cancer has focused on incidence and mortality, and has occasionally used a more restrictive age range than presented in this study. International adolescent and young adult cancer incidence patterns across time have been reported with data from Cancer Incidence in Five Continents reports, and national-level estimates have been reported from select, primarily high-income, countries. These publications have begun to raise awareness of adolescents and young adults as a distinctive population within the oncology community globally. However, to our knowledge, no previous publication has incorporated the impact of morbidity or done a comparative analysis of cancer within the broader context of the adolescent and young adult disease burden. We searched PubMed for English-language research articles describing the global burden of adolescent and young adult cancers between Jan 1, 2010, and Feb 1, 2021, using the terms “adolescent and young adult or adolescent or young adult or AYA” and “oncology or cancer or neoplasm or tumor or malignancy” and “global or worldwide or international” and “incidence or mortality or morbidity or burden or prevalence or survival”, and identified no additional comprehensive adolescent and young adult global cancer estimate reports. Added value of this study We share for the first time, the formal global analysis of the cancer burden in individuals aged 15–39 years in 2019, using disability-adjusted life-years (DALYs) estimated by the Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) 2019.GBD 2019 is a valuable global health resource used to inform government health policy decisions around the world when comprehensive data might be absent. The global burden of cancer in terms of mortality and DALYs is substantial in the adolescent and young adult population. The global distribution of the adolescent and young adult cancer burden is unique, reflecting the shift from cancers that primarily affect children (eg, acute lymphoblastic leukaemia) to those that primarily affect adults (eg, carcinomas), and including cancers that occur most often in adolescents and young adults (eg, testicular cancers). Although high Socio-demographic Index (SDI) countries had the highest age-standardised incidence rates, they also had the lowest age-standardised mortality rates when compared to non-high SDI (low, low-middle, middle, and high-middle SDI) countries. Implications of all the available evidence The relative burden of deaths and DALYs due to adolescent and young adult cancer is high globally, concentrated primarily in non-high SDI settings. These estimates are crucial for comparing the burden of cancer to other causes of deaths and DALYs in adolescents and young adults and might be used to inform health policy and resource allocation priorities. Focus on adolescents and young adults as a distinct cancer population in the development of cancer control programmes is crucial to improving outcomes.

4Evidence before this study

5Adolescents and young adults with cancer represent a transition population within the cancer continuum between children and older adults. As adolescents and young adults with cancer are treated by a variety of specialists, their unique epidemiology and clinical care needs are often overlooked. Although improvements in survival for children and adults with cancer are reported in high-income countries, less incremental progress has been observed among adolescents and young adults. Added complexities of cancer in this age group include the potential impact of a cancer diagnosis on starting or caring for their families and careers, access to care, diagnostic delays, and abandonment of therapy—issues that exist globally. Previous work assessing the global burden of adolescent and young adult cancer has focused on incidence and mortality, and has occasionally used a more restrictive age range than presented in this study. International adolescent and young adult cancer incidence patterns across time have been reported with data from Cancer Incidence in Five Continents reports, and national-level estimates have been reported from select, primarily high-income, countries. These publications have begun to raise awareness of adolescents and young adults as a distinctive population within the oncology community globally. However, to our knowledge, no previous publication has incorporated the impact of morbidity or done a comparative analysis of cancer within the broader context of the adolescent and young adult disease burden. We searched PubMed for English-language research articles describing the global burden of adolescent and young adult cancers between Jan 1, 2010, and Feb 1, 2021, using the terms “adolescent and young adult or adolescent or young adult or AYA” and “oncology or cancer or neoplasm or tumor or malignancy” and “global or worldwide or international” and “incidence or mortality or morbidity or burden or prevalence or survival”, and identified no additional comprehensive adolescent and young adult global cancer estimate reports.

6Added value of this study

7We share for the first time, the formal global analysis of the cancer burden in individuals aged 15–39 years in 2019, using disability-adjusted life-years (DALYs) estimated by the Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) 2019.GBD 2019 is a valuable global health resource used to inform government health policy decisions around the world when comprehensive data might be absent. The global burden of cancer in terms of mortality and DALYs is substantial in the adolescent and young adult population. The global distribution of the adolescent and young adult cancer burden is unique, reflecting the shift from cancers that primarily affect children (eg, acute lymphoblastic leukaemia) to those that primarily affect adults (eg, carcinomas), and including cancers that occur most often in adolescents and young adults (eg, testicular cancers). Although high Socio-demographic Index (SDI) countries had the highest age-standardised incidence rates, they also had the lowest age-standardised mortality rates when compared to non-high SDI (low, low-middle, middle, and high-middle SDI) countries.

8Implications of all the available evidence

9The relative burden of deaths and DALYs due to adolescent and young adult cancer is high globally, concentrated primarily in non-high SDI settings. These estimates are crucial for comparing the burden of cancer to other causes of deaths and DALYs in adolescents and young adults and might be used to inform health policy and resource allocation priorities. Focus on adolescents and young adults as a distinct cancer population in the development of cancer control programmes is crucial to improving outcomes.

10Previous studies have reported on global cancer incidence and mortality patterns of adolescents and young adults.,, One study used incidence and mortality estimates from GLOBOCAN 2012 for individuals aged 20–39 years, another reported incidence and mortality estimates from GLOBOCAN 2018 for individuals aged 15–39 years, and a third study reported international cancer incidence trends in individuals aged 15–39 years using data from the Cancer Incidence in Five Continents series, a publication comprising data from a subset of countries around the world with high-quality population-based cancer registries.,, However, global differences in measures that incorporate both morbidity and mortality due to adolescent and young adult cancers remain unexplored. Consideration of more comprehensive disease burden metrics is especially relevant in adolescents and young adults, whose disease burden might put a strain on their evolving careers and families.,

11The Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) is the only global disease burden estimation framework that provides estimates of disability-adjusted life-years (DALYs) for cancer as a metric to complement incidence and mortality data. DALYs are a key measure of disease burden that include both fatal and non-fatal impacts of disease, and are used in the development of national and global health policy. GBD estimates disease burden for more than 300 diseases and injuries, allowing for comparative analyses with other causes of morbidity and mortality in adolescents and young adults. To our knowledge, no formal GBD analysis has previously been done of the global burden of cancer in the adolescent and young adult population. In this study, we aimed to analyse and report adolescent and young adult cancer burden estimates, using the most encompassing definition of adolescents and young adults (ie, individuals aged 15–39 years),, with a focused analysis on DALY estimates. DALYs represent an important comprehensive assessment of cancer burden in this distinctive population, adding to existing estimates of disease burden with more classic measures, and are crucial to informing cancer control strategies that address health disparities and inequities in this population.

12GBD was established to provide global disease burden metrics that are comprehensive and comparable over time. Estimates produced include incidence, prevalence, mortality, years of life lost (YLLs), years lived with disability (YLDs), and DALYs, measures that can each be used to describe different aspects of the adolescent and young adult cancer burden. Estimates are generated for each disease and injury and are reported by age group, sex, location, and year. Each GBD iteration replaces the previous round of GBD estimates for the entire estimated time series, so that updates to data and methods in the new GBD round are applied consistently across time. The present analysis was based on GBD 2019 estimates., GBD 2019 was done in accordance with the Guideline for Accurate and Transparent Health Estimates Reporting (appendix pp 5, 6). Data sources used in GBD 2019 are available online and are further outlined in the appendix (p 10). This manuscript was produced as part of the GBD Collaborator Network and in accordance with the GBD Protocol. Analyses were completed with Python (versions 3.6.2 and 3.6.7), Stata (version 13), and R (versions 3.5.0 and 3.4.1).

13Although the definition of the age range for adolescents and young adults varies, particularly in the upper age limit,,, we used the age range of 15–39 years in this study, since this is the most encompassing age range definition recommended in oncology, is endorsed by the US National Cancer Institute and the AYA Working Group of the European Society for Medical Oncology and the European Society for Paediatric Oncology, and allows for comparability with other studies on adolescent and young adult cancer. Individuals aged 15–39 years have also experienced the least progress in survival outcomes in most countries. Data for this age range are available online with the GBD Results Tool and for subsets of this age range with the GBD Compare data visualisation tool or GBD Results Tool. As there are differences in the preferred age range used to define adolescents and young adults around the world, results of the narrower age range of 15–29 years are presented in the appendix (pp 115–122).

14All malignant cancer types, as defined in the tenth revision of the International Classification of Diseases, chapter II (Neoplasms), were categorised into 32 cancer groups in this analysis, called causes in GBD and this Article. Non-melanoma skin cancers were excluded, since they are not a major cause of mortality in this age range. The cause “other malignant neoplasms” in GBD includes estimates for cancers not included in any other GBD cancer cause, such as bone cancers and soft tissue sarcomas (see appendix p 11 for more details about cancer mapping). The adolescent and young adult age group was compared to children (aged 0–14 years) and older adults (aged ≥40 years) in specific analyses. The focus of this analysis was on global and regional estimates, although GBD 2019 also produces estimates at the national and, for select countries, subnational level. National and subnational estimates are available in the GBD Compare and GBD Results tools online. Select results are presented by quintiles of the Socio-demographic Index (SDI; countries ranked and divided into five equal SDI groups), which is a composite measure of income per capita, total fertility rate (age <25 years), and average educational fulfilment (for those aged ≥15 years), and is a useful summary measure of a country's overall social and economic development that allows for analyses of disease burden patterns across different resource contexts (appendix p 56). All cancer rates were reported per 100 000 person-years. The GBD world population standard was used for the calculation of age-standardised rates (appendix p 56).

15The GBD cancer estimation process begins with a focus on mortality. Data sources include vital registration systems, verbal autopsies, and population-based cancer registration systems. Some cancer registries report incidence only; therefore, mortality-to-incidence ratios (MIRs) were used to convert cancer registry incidence data to estimates of mortality, increasing data availability in locations that might not have mortality data, but have active cancer registries. Using a spatiotemporal Gaussian process regression, MIRs were modelled for all combinations of age, sex, year, and location with incidence data from cancer registries and mortality data from cancer registries or high-quality vital statistics registries (elaborated in the appendix pp 25, 26). Estimates of mortality obtained with MIRs were combined with vital registration and verbal autopsy mortality data and used as inputs in cancer type and sex-specific Cause of Death Ensemble models (CODEm). The CODEm methodology uses all available mortality data to select the optimal model or models on the basis of out-of-sample predictive validity (appendix p 28). Cause-specific mortality estimates were then scaled to independently modelled all-cause mortality with CoDCorrect to ensure consistency.

16Incidence estimates were obtained by dividing the mortality estimates by the corresponding MIR for each cancer type. Survival estimates based on MIRs were used to model 10-year prevalence for each cancer cause (appendix pp 50, 51). Prevalence for each cancer cause was divided into distinct phases of cancer treatment to estimate YLDs. For cohorts that survived beyond 10 years from diagnosis, two phases were estimated for the 10-year time period after diagnosis: diagnosis or treatment; and remission. After the 10-year period, the disability risk was returned to the baseline of the general population without a cancer diagnosis. For cohorts that did not survive beyond 10 years from diagnosis, two additional phases were estimated: the metastatic or disseminated phase; and the terminal phase. YLD estimates were generated by multiplying each phase prevalence by a phase-specific disability weight, representative of the health loss magnitude associated with a specified health outcome. Disability weights are measured on a scale of 0 (full health) to 1 (equivalent to death; appendix p 55). YLLs were calculated as the standard life expectancy at the age of death multiplied by age-specific cancer deaths. DALY estimates were the sum of the YLD and YLL estimates. Proportional DALYs for each cancer cause and 5-year age group were calculated as the mean of 1000 proportion draws of the absolute number of DALYs for each cancer cause and age group divided by the total number of cancer DALYs within the same age group. Proportional DALYs for each SDI were calculated as the mean of 1000 proportion draws of the absolute number of DALYs for each cancer cause within each SDI quintile and divided by the total number of DALYs in each SDI quintile (appendix p 56). An additional analysis was done to identify the proportion of adolescent and young adult cancer cases covered by the WHO Global Initiative for Childhood Cancer (appendix p 56). Further detailed descriptions of the methods are provided in the appendix (pp 7–57) and in GBD 2019 summary publications.,

17Final point estimates are reported with 95% uncertainty intervals (UIs). 95% UIs are 95% ranges calculated as the range from the 2·5th to the 97·5th percentile on the basis of the distribution of 1000 draws at each GBD cancer estimation step, with uncertainty propagated through each step (appendix p 57).

18The funders of this study had no role in the design of the GBD cancer estimation process, collection or analysis of data, interpretation of results, or in the writing of this manuscript.

19There were an estimated 1·19 million (95% UI 1·11–1·28) incident cancer cases and 396 000 (370 000–425 000) deaths among individuals aged 15–39 years worldwide in 2019 (table). The highest age-standardised incidence rates were seen in high SDI (59·6 [54·5–65·7] per 100 000 person-years) and high-middle SDI (53·2 [48·8–57·9] per 100 000 person-years) countries, while the highest age-standardised mortality rates from cancer in adolescents and young adults were seen in middle SDI (13·6 [12·6–14·8] per 100 000 person-years) and low-middle SDI (14·2 [12·9–15·6] per 100 000 person-years) regions. Adolescent and young adult cancers contributed 23·5 million (21·9–25·2) DALYs to the global burden of disease in 2019 (table), of which 2·7% (1·9–3·6) came from YLDs and 97·3% (96·4–98·1) from YLLs (appendix p 79). The majority (91·4% [91·0–91·8]) of the worldwide absolute adolescent and young adult cancer DALY burden is concentrated in non-high SDI (low, low-middle, middle, and high-middle SDI) quintiles. Overall, high SDI settings have the highest age-standardised incidence rate (59·6 [54·5–65·7] per 100 000 person-years), but the lowest age-standardised DALY rate (564·3 [542·8–590·1] per 100 000 person-years). Breast cancer (10·6% [10·0–11·2]), followed by brain and CNS cancer (7·4% [6·0–8·0]), colon and rectum cancer (7·0% [6·6–7·3]), and stomach cancer (6·7% [6·5–7·0]) were the four greatest contributors to the DALY burden globally for both sexes combined, of separately categorised cancers (appendix p 81). If leukaemias were considered as a single group, given that they are treated by haematologist-oncologists and have a similar diagnostic approach, rather than as individual leukaemia subtypes, leukaemias would be the largest categorised cancer group contributing to the global cancer DALY burden (12·0% [10·9–12·8]), greater than that of breast cancer. The “other malignant neoplasms” category, the aggregated cancer cause category for cancers not separately estimated in the GBD framework, comprised the highest proportion of the adolescent and young adult cancer DALY burden globally (13·7% [12·8–14·5]; appendix p 81). A focused analysis of individuals aged 15–29 years is provided in the appendix (pp 115–122).

20The greatest burden of cancer in adolescents and young adults in 2019, as represented by age-standardised DALY rates, was concentrated in parts of Asia, southern sub-Saharan Africa, and South America (figure 1A; appendix p 84). The distribution of DALYs due to cancer in adolescents and young adults is distinct from that of children (figure 1B) and older adults (figure 1C). The geographical pattern of age-standardised DALY rate quintiles for adolescent and young adult cancer was similar to the geographical pattern of childhood cancers in high SDI countries and resembled the distribution of adult cancer in low and middle SDI countries (figure 1).

21Of all age groups, individuals aged 35–39 years had the largest contribution to the adolescent and young adult global cancer DALYs (8·4 million [95% UI 7·8–9·0], with corresponding DALY rates of 1547·6 [1441·3–1658·0] per 100 000 person-years; figure 2A). The proportion of DALYs attributed to leukaemias declined with increasing age across the adolescent and young adult population (26·7% [24·8–28·8] of total age group DALYs, corresponding to 0·64 million [0·56–0·72] DALYs in those aged 15–19 years vs 6·2% [5·6–6·7] of total age group DALYs, corresponding to 0·52 million [0·46–0·58] DALYs in those aged 35–39 years; appendix p 107). The proportion of DALYs attributed to carcinomas increased with increasing age across the adolescent and young adult population (18·1% [17·3–19·3] of total age group DALYs in those aged 15–19 years, corresponding to 0·43 million [0·40–0·47] DALYs vs 73·6% [72·7–75·2] of total age group DALYs, corresponding to 6·2 million [5·7–6·6] DALYs in those aged 35–39 years; figure 2B; appendix p 107). There was a notable proportion of “other malignant neoplasms” across the adolescent and young adult population, which was highest in those aged 15–19 years (30·6% [28·6–32·2] of total age group DALYs, corresponding to 0·73 million [0·65–0·81] DALYs), and lowest in those aged 35–39 years (7·1% [6·5–7·5] of total age group DALYs, corresponding to 0·59 million [0·54–0·65] DALYs; figure 2B). In direct comparisons of the proportional DALY burden for the 15–29-year age group with that of the 30–39-year age group, there is a transition in the predominant cause from leukaemias and lymphomas to carcinomas, especially breast and cervical cancer (appendix p 122).

22When assessed by SDI quintile, age-standardised DALY rates and the proportional DALY burden varied by cancer type (figure 3). Individuals in the high SDI quintile had a lower age-standardised DALY rate (564·3 [95% UI 542·8–590·1] DALYs per 100 000 person-years; figure 3A) than other SDI quintiles. Estimates of the proportion of the DALY burden due to cervical cancer increased with decreasing SDI quintile, having the lowest proportional burden in the high SDI setting (4·1% [3·7–4·4]; figure 3B) and the highest burden in the low SDI setting (12·1% [10·4–14·4]; figure 3B). The adolescent and young adult cancer burden attributed to brain and CNS cancer was highest in the high SDI (10·7% [8·8–11·6]) and high-middle SDI (9·0% [7·2–9·7]) quintiles, compared to the low-middle SDI (6·1% [5·0–6·9]) and low SDI (5·2% [4·0–6·2]) quintiles. The proportion of adolescent and young adult cancers that were in the “other malignant neoplasms” category was highest in the low SDI quintile (20·1% [18·7–22·2]) and lowest in the high-middle SDI quintile (9·9% [9·5–10·6]).

23The top five causes by absolute DALY burden in females globally in 2019 were breast cancer (2·46 million [95% UI 2·23–2·70] DALYs), cervical cancer (1·56 million [1·32–1·78] DALYs), “other malignant neoplasms” (1·35 million [1·21–1·51] DALYs), stomach cancer (732 000 [653 000–814 000] DALYs), and brain and CNS cancer (722 000 [536 000–827 000] DALYs; figure 4; appendix pp 66–69). The five cancers with the highest absolute DALY burden in males were “other malignant neoplasms” (1·88 million [1·64–2·12] DALYs); brain and CNS cancer (1·03 million [0·76–1·19] DALYs); colon and rectum cancer (973 000 [887 000–1 070 000] DALYs); tracheal, bronchus, and lung cancer (856 000 [766 000–952 000] DALYs); and stomach cancer (842 000 [767 000–928 000] DALYs; figure 4; appendix pp 62–65). In 2019, females had a higher overall incidence of cancer than males globally (686 000 [622 000–751 000] vs 509 000 [469 000–549 000] incident cancer cases), but had similar absolute mortality (202 000 [184 000–222 000] vs 194 000 [179 000–209 000] deaths; appendix pp 62, 66, 82–83). Breast and cervical cancer combined made up a substantial proportion of the DALY burden globally in females (33·6% [32·3–35·1]). Among the non-sex-specific cancer causes, males had higher absolute DALYs globally in 24 of 27 cancer groups, representing a 13·7% (3·5–25·1) overall higher absolute number of DALYs than females.

24Rankings of the burden of absolute DALYs and deaths due to adolescent and young adult cancer compared to other diseases in individuals aged 15–39 years, both globally and by SDI quintile, are shown in figure 5. Adolescent and young adult cancer had the tenth highest DALY burden globally (23·5 million [95% UI 21·9–25·2] DALYs; figure 5A) among 22 causes of DALYs at this level in the GBD hierarchy. The inter-category rankings show that cancer ranks higher than other prominent causes of DALYs in high, high-middle, and middle SDI quintiles, compared to low-middle and low SDI quintiles. In adolescents and young adults, deaths from cancer ranked fourth globally (396 000 [370 000–425 000]; figure 5B), among 21 causes of death at this level in the GBD hierarchy, with a higher intra-SDI-quintile ranking in high, high-middle, and middle SDI regions, compared to low-middle and low SDI regions. In comparison, deaths due to cancer ranked 11th globally in those younger than 15 years and second in those older than 39 years (appendix pp 105, 106). In 2019, deaths due to cancer in the adolescent and young adult population were lower than those estimated for transport injuries and cardiovascular and circulatory diseases, but higher than those estimated for HIV/AIDS and sexually transmitted infections, respiratory infections and tuberculosis, and unintentional injuries (figure 5B). More detailed findings are summarised in the appendix (pp 62–114). An additional analysis showed that 8·6% (95% UI 8·2–9·1) of all adolescent and young adult cancer cases are included in the WHO Global Initiative for Childhood Cancer (appendix p 56).

25In our analysis of adolescent and young adult cancer, based on data from GBD 2019, we show, to the best of our knowledge, for the first time that the global burden of adolescent and young adult cancer is substantial in terms of DALYs, a measure that is frequently used by governments to inform policy and resource allocation needs. From a descriptive perspective, the age-standardised distribution of adolescent and young adult DALYs was unique compared to both childhood and adult cancers, reflecting an expected but ill-described transition from childhood to adult cancer epidemiological patterns.,, Additionally, when the overall disease burden is studied cross-sectionally within the age range encompassing adolescents and young adults, the global burden of cancer contributed more DALYs to the global disease burden than some high-profile communicable diseases such as HIV/AIDS and sexually transmitted infections. This comparison of cancer with other leading causes of global mortality and DALYs in adolescents and young adults has not been previously documented. These results highlight that cancer is an important contributor to premature death and the disease burden in adolescents and young adults globally, even when compared with some communicable diseases that that are the focus of more active global funding, research, and advocacy efforts., The findings also underscore the need to develop a global strategy to address the cancer burden in this population, which should include the integration of adolescent and young adult cancer into overall cancer control planning and universal health coverage plans.

26Because of the substantial burden of adolescent and young adult cancers globally, with the majority of DALYs occurring on the lower end of the SDI spectrum, broader attention to the unique determinants driving cancer outcomes in this age range is needed. In 2017, the World Health Assembly accepted the global cancer challenge resolution, which stated the importance of including children and adolescents in the development of cancer control programmes. The World Health Assembly noted in particular that these populations often experience delays and difficulties in accessing care. Unfortunately, the resolution did not address the unique needs of young adults separately, thus reinforcing a gap in current global cancer control paradigms. There is an opportunity for advocates to directly address this gap, petitioning member states and developing an amendment specific to adolescents and young adults by emphasising the barriers faced by these patients.

27The psychosocial challenges adolescents and young adults face is an important issue since these challenges are truly unique across the age spectrum and require resources and skills that are often not available to cancer treatment teams. The age range of adolescents and young adults encompasses their formative years in life and spans the time from completing education, to possibly starting a career and raising children, and potentially contributing to society more broadly. A cancer diagnosis during these years can have a considerable impact on individuals' future life trajectory through major stressors, including feelings of isolation, anxiety and depression, concerns about infertility, discontinuing schooling or work, and financial hardship.,,,, Efforts to mitigate the issues distinct to this age group have resulted in the formation of organisations to help support adolescent and young adult patients with cancer. However, although these oncology advocacy efforts focused on adolescents and young adults have been successful in creating awareness campaigns and implementing adolescent and young adult programmes at cancer centres, these efforts have largely been limited to high-income countries. These initiatives need to be expanded globally, particularly in low SDI settings—which carry a disproportionate burden of adolescent and young adult cancer DALYs—with appropriate local knowledge and champions.

28The array of cancer types is also unique in adolescents and young adults compared to children and adults. Even what seems to be the same cancer is often biologically different in adolescents and young adults than in patients of other age groups and thereby might benefit from a different approach to therapy. For these and other reasons, survival improvements in adolescent and young adult patients with cancer have lagged behind those of children and adults for several cancer types. Delivery of cancer care to adolescents and young adults should be prioritised and optimised, especially in non-high SDI settings, where the majority of DALYs are reported. At present, adolescent and young adult patients often do not have an obvious health-care home and are frequently grouped into adult oncology service programmes because of age restrictions in paediatric wards or facilities.,, Where a patient receives care has important clinical and policy ramifications, as there is evidence of improvement in survival outcomes for some cancer types (eg, acute lymphoblastic leukaemia) when adolescents and young adults are treated according to paediatric protocols, which are often complex and might be unavailable in adult cancer centres., Furthermore, treatment by specialised adolescent and young adult oncology teams has been associated with improved survival of adolescents and young adults with cancer in some high-income countries, possibly as a result of access to cancer expertise, clinical trials, and multidisciplinary care. Although access to these centres and programmes is not currently possible in many settings, most adolescent and young adult patients with cancer might benefit from a multidisciplinary treatment approach involving close collaboration between paediatric and medical oncologists.

29To improve outcomes in this unique population, a new approach to global cancer control in adolescents and young adults is required. Faced with similar challenges for children and adolescents, the recently launched WHO Global Initiative for Childhood Cancer provides one implementation framework for addressing gaps in access and care. This initiative includes adolescents up to 19 years of age, bridging the lowest ages included in adolescent and young adult oncology, and at least one cancer that predominantly occurs in adolescents and young adults—Hodgkin lymphoma—is an index cancer in this initiative. Although this is excellent news for the younger bounds of the adolescent and young adult spectrum, the Global Initiative for Childhood Cancer initiative covers only 8·6% (95% UI 8·2–9·1) of all adolescent and young adult cancer cases, and the unique needs of and potential synergies with adolescent and young adult cancer care are not specifically addressed. A dedicated initiative similar to the Global Initiative for Childhood Cancer is unlikely in the near future. Therefore, integration of adolescent and young adult cancer policies within WHO cancer initiatives such as the Global Initiative for Childhood Cancer and the WHO Cervical Cancer Elimination Initiative, a cancer that comprises approximately 10·0% (8·5–10·9) of adolescent and young adult cancer cases globally, could be prioritised in the short term. A strategy to integrate specific objectives of relevance to the adolescent and young adult population in these initiatives would immediately cover almost one-fifth of adolescent and young adult cancer cases and provide a template for future global cancer initiatives. Potential areas for collaboration could include integration of human papillomavirus (HPV) vaccination efforts into the Global Initiative for Childhood Cancer, an as-yet untapped opportunity, and inclusion of policies specific to adolescent and young adult patients in the WHO technical packages, such as provisions for referrals and access to expert adolescent and young adult cancer care and appropriate treatment regimens, psychosocial support, and universal health coverage to reduce financial hardship. Intentional collaboration with other WHO cancer initiatives could facilitate progress in both areas and highlight other potential areas of synergy for improving cancer outcomes in adolescents and young adults.

30The adolescent and young adult cancer burden estimates presented in this study also underscore the limitations of GBD and possible opportunities to improve future assessments of the global adolescent and young adult cancer burden. The classification of adolescent and young adult cancers in this study is based on the GBD cancer cause list, which has historically focused on cancers occurring in adulthood. As such, GBD 2019 did not differentiate some of the most common adolescent and young adult cancer types, such as soft tissue sarcomas and bone tumours. These cancers contribute to the substantial proportion of “other malignant neoplasms” in this age range, cancers that do not have their own individual GBD cancer causes. Many of the rarer cancers that fall into this “other malignant neoplasms” category rely on complex multidisciplinary therapy (eg, provided by medical, radiation, and surgical oncologists), and resource allocation could be improved if their global burden was accurately known. Future studies should use the recently updated recommendations for classification of adolescent and young adult cancers to better characterise the cancer burden in this age group and minimise the number of cancer types falling into the “other malignant neoplasms” category. Additionally, the quality of the data obtained, especially from low-resource settings, might cause challenges due to underestimates or miscategorisation of less common cancer types. For instance, there was an observed decrease in the proportion of adolescent and young adult cancer DALYs due to brain and CNS cancers across the SDI spectrum, with the lowest proportion in low SDI settings. As many lower SDI countries do not have population-based cancer registries or robust referral mechanisms, the data upon which these estimates are drawn might be subject to underdiagnosis, misdiagnosis, or under-reporting. Therefore, results in lower SDI settings should be interpreted with caution. However, these modelled results provide a useful contribution towards determining the global burden of adolescent and young adult cancer, especially in regions where such data do not exist or are scarce. An additional limitation of the present analysis is that SDI was applied at the national level, but within-country socio-economic status can vary greatly. Improving global adolescent and young adult cancer burden estimates must be rooted in capacity-building efforts that consider the local context, to ensure identification of incident cancer cases and deaths in the adolescent and young adult population, as well as expansion of and support for population-based cancer registration systems. Another potential limitation of the present analysis is the current approach to YLD estimation, which accounts for only 10 years after cancer diagnosis. Previous studies have shown that late effects, such as cardiomyopathy, can affect the adolescent and young adult population beyond the 10-year cutoff point., This limits the ability to determine the long-term chronic disease burden and competing risks for survivors in this population, which have the potential to be substantial. Additionally, the experience of disability for survivors of childhood cancer might be different to that of the general population. Thus, GBD 2019 might be underestimating the YLDs and DALYs associated with cancer in adolescents and young adults, and future efforts might be needed to identify ways to account for this limitation. Finally, this study focused on estimates from 2019, and thus did not incorporate the direct and indirect effects of the COVID-19 pandemic on the global adolescent and young adult cancer burden. This will be an important area of consideration in future studies as the data become available.

31This report of the adolescent and young adult cancer burden from GBD 2019 identified a considerable burden of DALYs due to cancer in the global adolescent and young adult population. The absolute mortality burden in adolescents and young adults is highest in non-high SDI settings, underscoring the need for a global effort to improve outcomes in this population, with collaboration at the regional and country levels, as well as between governments, institutions, academic societies, and patient advocacy and non-profit organisations. Efforts to comprehensively estimate the global burden of cancer in adolescents and young adults are a crucial first step., Adolescent and young adult oncology has historically been less prioritised than cancer disciplines in younger and older patients. Increased awareness of the burden of cancer in this population could lead to targeted interventions for improved outcomes.

32To download the data used in these analyses, please visit the Global Health Data Exchange GBD 2019 website at http://ghdx.healthdata.org/gbd-2019.

33R Ancuceanu reports consulting fees from Abbvie; and payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events from Abbvie, Sandoz and B. Braun, all outside the submitted work. H Ariffin reports payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events from Amgen, all outside the submitted work. M Atashzar reports support for the present manuscript from medical writing and analysis of data. M Atashzar reports paid consulting fees; and receipt of equipment, materials, drugs, medical writing, gifts or other services from medical writing, all outside the submitted work. P Atorkey reports support for the present manuscript from the School of Medicine and Public Health, University of Newcastle, Australia, Hunter Medical Research Institute, University of Newcastle, Australia, and Hunter New England, Population Health. A Aujayeb reports grants or contracts from Rocket Medical Plc; payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events from Rocket Medical Plc for talks given on pneumothorax and work done on digital suction device; and leadership or fiduciary role in board, society, committee or advocacy group, paid or unpaid with Mesothelioma UK as Trustee, all outside the submitted work. M Ausloos reports grants or contracts from the Romanian National Authority for Scientific Research and Innovation, CNDS-UEFISCDI, project number PN-III-P4-ID-PCCF-2016-0084 “Understanding and modelling time-space patterns of psychology-related inequalities and polarization”; all outside the submitted work. T Bärnighausen reports grants or contracts from the European Union (Horizon 2020 and EIT Health), German Research Foundation (DFG), US National Institutes of Health, German Ministry of Education and Research, Alexander von Humboldt Foundation, Else-Kräner-Fresenius-Foundation, Wellcome Trust, Bill & Melinda Gates Foundation, KfW, UNAIDS, and WHO; consulting fees from KfW on the OSCAR initiative in Vietnam; participation on a Data Safety Monitoring Board or Advisory Board with NIH-funded study “Healthy Options” as Chair of the Data Safety and Monitoring Board (DSMB), German National Committee on the “Future of Public Health Research and Education”; Chair of the scientific advisory board to the EDCTP Evaluation; Member of the UNAIDS Evaluation Expert Advisory Committee; National Institutes of Health Study Section Member on Population and Public Health Approaches to HIV/AIDS (PPAH), US National Academies of Sciences, Engineering, and Medicine's Committee for the “Evaluation of Human Resources for Health in the Republic of Rwanda under the President's Emergency Plan for AIDS Relief (PEPFAR)”, University of Pennsylvania (UPenn) Population Aging Research Center (PARC) as an External Advisory Board Member; leadership or fiduciary role in board, society, committee or advocacy group, paid or unpaid as a Co-chair of the Global Health Hub Germany (which was initiated by the German Ministry of Health); all outside the submitted work. N Bekele reports participation on a Data Safety Monitoring Board or Advisory Board as Ethical review board member for two years; and leadership or fiduciary role in board, society, committee or advocacy group, paid or unpaid with Wollo University as an unpaid graduate program coordinator for three years, all outside the submitted work. S Bhaskar reports grants or contracts from NSW Ministry of Health, NSW Brain Clot Bank; and leadership or fiduciary role in board, society, committee or advocacy group, paid or unpaid with Rotary Club of Sydney, Australia as Board Director, International Rotary Fellowship of Rotarian Healthcare Professionals (IRFRHP), UK as Board Director, and BMC Neurology as Editorial Board Member, all outside the submitted work. J Conde reports grants or contracts from European Research Council Starting Grant, ERC-StG-2019-848325; patents planned, issued or pending, as Functionalized nanoparticles and compositions for cancer treatment and methods, U.S. Application No. 62/334538 and TRPV2 Antagonists, WO Application No. PCT/PT2018/050035; and support from TargTex S.A. as co-founder and shareholder, all outside the submitted work. X Dai reports support for the present manuscript from Bloomberg Philanthropies and the Bill and Melinda Gates Foundation through their employment at IHME. I Filip reports payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events from Avicenna Medical and Clinical Research Institute, all outside the submitted work. L Force reports support for the present manuscript from the Bill and Melinda Gates Foundation and American Lebanese Syrian Associated Charities for providing funding, related to their employment at IHME. L. Force reports grants or contracts from St Baldrick's Foundation; leadership or fiduciary role in board, society, committee or advocacy group, unpaid with the Lancet Oncology International Advisory Board; and payments towards federal student loans from the NIH Loan Repayment Award, all outside the submitted work. F Ghassemi reports support for the present manuscript for medical writing and literature review. N Ghith reports grants or contracts from NovoNordisc Foundation through salary covered by grant NNF16OC0021856, all outside the submitted work. A Guha reports grants or contracts from American Heart Association-Strategically Focused Research Network Grant in Disparities in Cardio-Oncology (#847740,#863620), all outside the submitted work. V Gupta reports grants or contracts from National Health and Medical research Council (NHMRC) Australia, all outside the submitted work. J Haro reports grants or contracts from Eli Lilly and Co., all outside the submitted work. H Henrikson reports support for the present manuscript from the Bill and Melinda Gates Foundation, American Lebanese Syrian Associated Charities, and Saint Baldrick's Foundation, all for providing funding, related to their employment at IHME. C Herteliu reports grants or contracts from Romanian National Authority for Scientific Research and Innovation, CNDS-UEFISCDI, project number PN-III-P4-ID-PCCF-2016-0084, grant title “Understanding and modelling time-space patterns of psychology-related inequalities and polarization” as research team member, and from Romanian National Authority for Scientific Research and Innovation, CNDS-UEFISCDI, project number PN-III-P2-2.1-SOL-2020-2-0351, grant title “Approaches within public health management in the context of COVID-19 pandemic,” as project manager, all outside the submitted work. K Innos reports support for the present manuscript from Estonian Research Council, Grant No PRG722. S M S Islam reports grants or contracts from the NHMRC Emerging Leadership Fellowship and the National Heart Foundation of Australia Fellowship, all outside the submitted work. N E Ismail reports leadership or fiduciary role in board, society, committee or advocacy group, paid or unpaid with Malaysian Academy of Pharmacy as an unpaid Council Member, all outside the submitted work. I Karaye reports support for the present manuscript from the Bill and Melinda Gates Foundation, American Lebanese Syrian Associated Charities, and Saint Baldrick's Foundation, all for providing funding, related to their employment at IHME. J Kauppila reports grants or contracts from Sigrid Juselius Foundation, Finnish Cancer Foundation, and Päivikki and Sakari Sohlberg Foundation, all outside the submitted work. T Ketterl reports consulting fees from Fennec Pharmaceuticals, Inc for advisory services, all outside the submitted work. J Kocarnik reports support for the present manuscript from the Bill and Melinda Gates Foundation and American Lebanese Syrian Associated Charities for providing funding, related to their employment at IHME. M-C Li reports support for the present manuscript from Ministry of Science and Technology, Taiwan (MOST 110-2314-B-003-001). J A Loureiro reports support for the present manuscript from Scientific Employment Stimulus (FCT), CEECINST/00049/2018, for salary support and Base Funding, UIDB/00511/2020 of the LEPABE, funded by national funds through the FCT/MCTES (PIDDAC) for research support. M Mahmoudi reports support from the Academic Parity Movement, as co-founder and director; support from Partners in Global Wound Care (PGWC) as Founding Partner; and reports royalties/honoraria for published books, plenary lectures, and licensed patent, all outside the submitted work. O Odukoya reports support for the present manuscript from the Fogarty International Center of the National Institutes of Health under the Award Number K43TW010704. A Pana reports grants or contracts from ARPIM, Amgen, Janssen, Astra Zeneca, Novartis Oncology, BMS, Angelini, and Servier; and participation on a Data Safety Monitoring Board or Advisory Board with Novartis Oncology and Pfizer, all outside the submitted work. M Postma reports stock or stock options in Pharmacoeconomics Advice Groningen and Health-Ecore, all outside the submitted work. A Radfar reports payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events from Avicenna Medical and Clinical Research Institute, all outside the submitted work. A Riad reports grants or contracts from Masaryk University; and support from Cochrane Collaboration as deputy director of Cochrane Czech Republic center, all outside the submitted work. M Saylan reports support from their employer Bayer, all outside the submitted work. M Šekerija reports payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events from Roche and Johnson & Johnson, all outside the submitted work. D A Santos Silva reports support for the present manuscript from Coordenação de Aperfeiçoamento de Pessoal de Nível Superior—Brazil (CAPES) and National Council for Scientific and Technological Development (CNPq), Brazil. D A Santos Silva reports grants or contracts from the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior—Brazil (CAPES)—Finance Code 001 and is supported in part by National Council for Scientific and Technological Development (CNPq), Brazil (302028/2018-8), all outside the submitted work. J A Singh reports consulting fees from Crealta/Horizon, Medisys, Fidia, PK Med, Two labs Inc, Adept Field Solutions, Clinical Care options, Clearview healthcare partners, Putnam associates, Focus forward, Navigant consulting, Spherix, MedIQ, Jupiter Life Science, UBM LLC, Trio Health, Medscape, WebMD, and Practice Point communications, and the National Institutes of Health and the American College of Rheumatology; payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events from Simply Speaking; support for attending meetings and/or travel from OMERACT, an international organization that develops measures for clinical trials and receives arm's length funding from 12 pharmaceutical companies, when traveling to OMERACT meetings; participation on a Data Safety Monitoring Board or Advisory Board as a member of the FDA Arthritis Advisory Committee; leadership or fiduciary role in other board, society, committee or advocacy group, paid or unpaid, with OMERACT as a member of the steering committee, with the Veterans Affairs Rheumatology Field Advisory Committee as a chair member, and with the UAB Cochrane Musculoskeletal Group Satellite Center on Network Meta-analysis as Director and editor; stock or stock options in TPT Global Tech, Vaxart pharmaceuticals, Atyu biopharma, Charlotte's Web Holdings Inc. and previously owned stock options in Amarin, Viking, and Moderna pharmaceuticals; all outside the submitted work.