Beyond survival: the risk of subsequent primary melanoma in childhood cancer survivors

In this issue of the Journal, Shojaie and colleagues1 report long-term risks of subsequent primary melanoma in a large cohort of 68 002 five-year childhood cancer survivors across 12 European countries. Survivors were 2.3 times more likely to develop subsequent primary melanoma compared with the general population. Although the standardized incidence ratios (SIRs) decreased as survivors age, the absolute excess risks continued to rise, peaking after age 50 years—a period for which this study provides important new estimates. Risk was highly heterogeneous across subgroups. Survivors of heritable retinoblastoma experienced a 16.5-fold increased risk and a 5.5% cumulative incidence by age 60 years, while those who received radiotherapy had a 4-fold increased risk after age 50 years. These findings highlight that cancer type and attained age are important determinants of subsequent primary melanoma risk. Importantly, the study underscores that survivorship care requires a lifelong perspective that accounts for changes in absolute risk with age. The authors also advocate for enhanced dermatologic surveillance in high-risk groups to facilitate early detection, as well as general melanoma risk-reduction strategies for all survivors, including skin self-examination and sun protection. The PanCareSurFup study has several key strengths, notably its unprecedented scale, long-term follow-up, and ability to evaluate rare outcomes across diverse subgroups. It represents the largest assembled cohort of childhood cancer survivors worldwide, including 68 002 individuals from 12 European countries and data contributions from 11 population-based cancer registries. This scale provides the statistical power needed to analyze rare outcomes and specific subgroups. The cohort spans a long period, including survivors diagnosed between 1940 and 2008, and contributes more than 1.2 million person-years of follow-up, enabling robust assessment of late effects that may take decades to emerge. Loss to follow-up was low (<6% in any participating country), supporting data completeness and internal validity. Methodologically, the authors further strengthen their inferences using multivariable Poisson regression with random effects to account for between-country heterogeneity, as well as directed acyclic graphs to guide confounder selection. Importantly, the study provides more granular estimates of melanoma risk across specific childhood cancer types, including central nervous system tumors, Wilms tumor, neuroblastoma, leukemia, and other solid tumors, substantially expanding on prior work. Unlike earlier studies with limited follow-up into older adulthood, this study extends risk estimates beyond age 50 years, including nearly 3 times as many melanoma events in this age group as the previous largest study.2 This extended follow-up confirms that, although standardized incidence ratios decrease with age, the absolute excess risks continue to rise in older survivors. This study, although extensive, has several limitations related to data granularity, missing clinical details, and potential misclassification of genetic status. Limited treatment detail constrained deeper interpretation of the findings. A clinically significant portion (n = 21 887) of the cohort was excluded from treatment-specific analyses because treatment data were unavailable. In addition, the study lacks detailed treatment information, limiting the assessment of risks by cumulative doses of specific chemotherapeutic agents, radiation dose, or specific radiation fields. Despite this lack of treatment data, the findings are consistent with those from the North American Childhood Cancer Survivor Study, which included more detailed treatment information.2 Although childhood cancer survivors and the general population appear to share a similar overall distribution of melanoma involving the trunk (36.0% vs 33.7%, respectively),3 the absence of detailed anatomical and radiotherapy field data may obscure important differences in subanatomic localization, particularly within the back, chest, and abdomen. Furthermore, because of data protection constraints inherent to a multicountry study, detailed pathological characteristics, such as Breslow thickness and staging, were unavailable. As a result, the aggressiveness of subsequent primary melanoma in survivors could not be assessed, although these are key determinants of prognosis. Because genetic testing was not routinely performed for all survivors in this historical cohort, the heritable status of some retinoblastoma survivors may have been misclassified, potentially inflating the estimates for the nonheritable group. Similar misclassification may occur across other childhood cancer types. Pathogenic germline variants associated with cancer predisposition are identified in approximately 6%-12% of childhood cancer survivors and represent a major driver of subsequent malignant neoplasms in this population. Harboring cancer predisposition syndrome is the strongest independent risk factor for subsequent malignant neoplasms, with a hazard ratio of 7.8.4 Twenty years postdiagnosis, the cumulative incidence of a subsequent malignant neoplasm reaches 23% in survivors with a cancer predisposition syndrome compared with only 3% in noncarriers.4 Despite these limitations, the study provides unprecedented statistical power and robust long-term risk data on subsequent malignant neoplasms in childhood cancer survivors. Before this study, data on subsequent malignant neoplasms among survivors aged older than 50 years were extremely limited. This study provides estimates for a population reaching the median age of melanoma diagnosis in the general population, thereby filling a critical knowledge gap in survivorship research. Moreover, the study offers a practical path forward for medical care. It demonstrates that when detailed treatment records are missing—a common issue for older survivors—childhood cancer type and attained age can serve as effective risk-stratifying factors to help primary care providers determine who requires more intensive screening. This is particularly important given that 85% of childhood cancer survivors are seen in community settings.5 These findings suggest several clear paths forward for the clinical care of childhood cancer survivors and for future research, including a shift from a treatment-focused approach to a more comprehensive risk-stratification model. Current guidelines from the Children’s Oncology Group Long-Term Follow-Up Guidelines for Survivors of Childhood, Adolescent, and Young Adult Cancer primarily focus on treatment history, such as radiotherapy or hematopoietic stem cell transplantation, to determine subsequent primary melanoma risk. These guidelines could be strengthened by incorporating childhood cancer type and recognizing it as an independent indicator of elevated risk. However, potential confounding between diagnosis and treatment exposures must be carefully considered, as the 2 are closely linked. In addition, ongoing surveillance should remain a priority, particularly for those aged older than 50 years, when the absolute excess risk of subsequent primary melanoma is highest, even though the standardized incidence ratio declines. This study also highlights the need to broaden surveillance and prevention efforts. Regular professional skin checks should be incorporated into long-term survivorship care for high-risk subgroups to facilitate early detection, but all survivors should be encouraged to practice regular skin self-examinations, use sun protection, and avoid excessive sun exposure. However, adherence to surveillance guidelines remains suboptimal, and skin cancer screening rates are no exception, at approximately 22%.6 Barriers to adherence include knowledge gaps among survivors regarding treatment exposures, associated risks, and the need for screening.7 Provider-related challenges further compound the issue, particularly given that most adult survivors of childhood cancer received care in community settings.5,8 Recently, several interventional trials have been designed to address suboptimal adherence to surveillance guidelines, often leveraging behavioral theories and digital health strategies. For example, the Advancing Survivors Knowledge (ASK) study uses print materials, text messaging, and a dedicated website to promote skin cancer screening,9 and the ENGaging and Activating Cancer Survivors in GEnetic Services (ENGAGE) study evaluates remote telegenetic services to increase uptake of genetic counseling and testing.10 In the Technology-Enabled Patient and Physician Activation to Enhance Skin Cancer Screening After HCT (TEACH) study, which targeted survivors of hematopoietic stem cell transplantation, technology-enabled patient activation improved skin self-examination, while physician-directed activation further increased clinician-performed skin examinations (Armaian et al, revised manuscript under review). This highlights the complementary roles that patients and providers can play in implementing survivorship guidelines. More interventional studies focused on improving adherence to surveillance guidelines are urgently needed to support early detection and prevention. To further refine risk stratification and model development, future efforts should address the limitations identified in this study by linking specific radiation doses and fields to the anatomical sites where subsequent primary melanomas develop and by incorporating detailed pathological characteristics, such as Breslow thickness and staging, to determine whether these melanomas are more aggressive than those in the general population and how they relate to clinical outcomes. The findings from the PanCareSurFup cohort study on the risk of subsequent primary melanoma point to a necessary shift in survivorship care—from a treatment-centered model toward a more comprehensive, risk-adapted approach that incorporates cancer type, attained age, and, where available, genetic predisposition. As survivors age, the growing burden of subsequent primary melanoma underscores the importance of sustained surveillance strategies that extend well beyond early adulthood. At the same time, interventional studies to improve adherence to surveillance recommendations are needed to support early detection and prevention. Gaps in treatment detail and anatomical specificity highlight critical areas for future research, particularly in understanding how radiation exposure and other etiological factors may influence subsequent primary melanoma development and mortality. Moving forward, integrating epidemiologic insights with more granular clinical and biological data, along with developing interventions to improve surveillance adherence, will be essential to refine risk stratification and optimize long-term care for this expanding population of survivors. Lindsay F. Schwartz (Conceptualization, Writing—original draft, Writing—review & editing), Kevin C. Oeffinger (Conceptualization, Writing—review & editing), and Tara O. Henderson (Conceptualization, Writing—original draft, Writing—review & editing) No funding or support was used in this editorial. L.F.S. and K.C.O. have no conflicts of interest to disclose. T.O.H., who is a JNCI/JNCI Cancer Spectrum Associate Editor/Deputy Editor/Editor-in-Chief and co-author on this paper, was not involved in the editorial review or decision to publish the manuscript. No data were generated or analyzed for this editorial.