Lifelong disease burden of chemotherapy in Hodgkin lymphoma (HL): A simulation study from the St. Jude Lifetime (SJLIFE) Cohort and HL International Study for Individual Care (HoLISTIC).

Abstract

12068 Background: Current emphasis for childhood and young adults with HL is to maintain high cure rates while concurrently identifying regimens to reduce excess long-term mortality/morbidity. Thus, understanding the late effects (LE) of contemporary clinical trials (CCT) for HL is critical. Methods: We used simulation to estimate the projected life expectancy (LExp), quality adjusted life-expectancy (QALE) & cause of death (COD) in a large cohort of HL CCT patients (pts) in the recently established HoLISTIC consortium by linking long-term risk models from the SJLIFE cohort. Individual patient data (IPD) on bleomycin, alkylating agents and anthracycline were extracted & harmonized for 982 HL pts in 5 prospective CCT (mean diagnosis age 19y, range 3-30y; 51% male; all treated with chemotherapy only; progression-free survival [PFS] >5y) in the HoLISTIC database. LExp, QALE & COD were projected using a previously developed microsimulation model (Bhakta, Blood [Supplement], 2019) that incorporated mortality & incidence of LEs by diagnosis age, sex, race, treatment exposures & attained age estimated from 5,522 adult 10-y survivors of childhood cancers in the SJLIFE cohort (56% male; mean age at last follow-up 35y, range 19-68). Microsimulation was applied to 10,000 randomly selected survivors of HL CCT cohort, from 10y after HL diagnosis until death to project the LExp, QALE & COD. Results: Assuming 10-y PFS, LExp and QALEs projected for the HL CCT cohort using adjusted US general population rates linked with the SJLIFE microsimulation model, COD and trial-specific exposures are shown in the Table. Conclusions: A novel lifetime simulation approach was used to project LExp, QALE & COD by linking together IPD from CCTs with the long-term risk model of the SJLIFE survivorship cohort. Despite differences in PFS, reflecting in part the variation in risk/stage status, the projected long-term outcomes were similar. Our approach highlights a new opportunity to inform future clinical trial design and aid provider & patient decision-making. [Table: see text]