Purpose: Hodgkin's lymphoma (HL) survivors are known to be at substantially increased risk of solid cancers (SC). However, no investigation has used multivariate modeling to estimate the relative risk (RR), excess absolute risk (EAR), and cumulative incidence for specific attained ages and ages at HL diagnosis.
Patients and methods: We identified 18,862 5-year HL survivors from 13 population-based cancer registries in North America and Europe. Poisson regression was used to evaluate the effects of age at diagnosis, attained age, latency, sex, treatment, and year of diagnosis on the RR and EAR of SC.
Results: Among 1,490 identified SC, 850 were estimated to be in excess. For most cancer sites, both RR and EAR decreased with age at HL diagnosis and showed strong dependencies on attained age. For a patient diagnosed at age 30 years and survived to > or = 40 years, modeled risks were significantly elevated for cancers of the breast (RR = 6.1), other supradiaphragmatic sites (RR = 6.0), and infradiaphragmatic sites (RR = 3.7); the largest RR (20-fold) was observed for malignant mesothelioma. Thirty-year cumulative risks of SC for men and women diagnosed at 30 years were 18% and 26%, respectively, compared with 7% and 9%, respectively, in the general population. For young HL patients, risks of breast and colorectal cancers were elevated 10 to 25 years before the age when routine screening would be recommended in the general population.
Conclusion: Multivariable modeling demonstrates for the first time temporal changes in SC risk not evident in unadjusted analyses, and can facilitate the development of individualized risk assessment and the creation of screening strategies for early detection.