Abstract
Background: Treatment advances have significantly increased cure and survival rates in Hodgkin lymphoma (HL), with 5-year relative survival exceeding 90% in the United States (US). However, it is unknown whether this benefit has resulted in a similar mortality rate compared to the general population, particularly among long-term survivors. We conducted a population-level analysis comparing all-cause mortality among 5-year HL survivors to the general population.
Methods: We conducted a retrospective study using data from the population-based Surveillance, Epidemiology, and End Results registry, involving patients with first-primary HL diagnosed at age 20 years or older from 2000 to 2017. Follow-up was only available up to 2022. Patients with documented first-line chemotherapy with or without receiving radiotherapy who survived 5 years were included. Classical HL subtypes included were nodular sclerosis, mixed cellularity, lymphocyte-rich, and lymphocyte-depleted HL, as well as HL-not otherwise specified. The study endpoints included standardized mortality ratios (SMRs) to estimate relative risks and excess absolute risks (EARs) per 10,000 patient-years to assess the burden of the disease. Observed all-cause mortality was compared to the expected mortality based on age-, sex-, race/ethnicity-, and calendar period-matched general population. Subgroup analysis involved estimating the endpoints stratifying by sex, age groups, race/ethnicity, and calendar periods (2000-2007 [chemotherapy-based therapy] and 2008-2017 [reduced radiotherapy, PET-adapted therapy, and addition of immunotherapy]).
Results: During the 2000-2017 period, we identified 19,895 5-year HL survivors, with a median age of 36 and a male predominance (53%). With a median follow-up of 102 months (interquartile range 100-104 months), we observed 2,583 deaths. Overall, mortality was significantly increased (SMR=1.95, 95% confidence interval [CI]=1.87-2.03; EAR=80.8) compared to the general population, with the highest estimates in mixed cellularity HL (SMR=2.04, 95% CI=1.87-2.23; EAR=144.9), followed by nodular sclerosis (SMR=1.92, 95% CI=1.82-2.03; EAR=59.2), lymphocyte-depleted (SMR=1.79, 95% CI=1.25-2.49; EAR=126.1), and lymphocyte-rich (SMR=1.61, 95% CI=1.30-1.98; EAR=83.3) HL.
In subgroup analyses, SMR and EAR by sex stratification followed similar patterns to the overall analysis (P-heterogeneity=0.151). However, patients diagnosed during young adulthood (20-39 years) experienced the highest mortality risk (SMR=2.62, 95% CI=2.42-2.84; EAR=38.6; (P-heterogeneity<0.0001) compared to middle-aged (40-59 years; SMR=1.97, 95% CI=1.85-2.11; EAR=96.4) or older adults (≥60 years; SMR=1.69, 95% CI=1.59-1.79; EAR=333.4). According to race/ethnicity, non-Hispanic Asian and Pacific Islander patients experienced the highest mortality risk with an SMR of 3.21 (95% CI=2.59-3.94; EAR=87.1; (P-heterogeneity<0.0001), while SMRs for other race/ethnicity groups ranged from 1.83 to 2.13 (EAR range 75.8-100.8).
Over time, SMR significantly reduced from 2.05 (95% CI=1.96-2.15; EAR=85.5) in the 2000-2007 period to 1.77 (95% CI=1.65-1.89; EAR=71.6; P-trend<0.0001) in the 2008-2017 period. This trend was more pronounced for nodular sclerosis HL (P-trend<0.0001) and mixed cellularity HL (P-trend=0.0003). In contrast, the trend remained constant for lymphocyte-rich HL (P-trend=0.410) and lymphocyte-depleted HL (P-trend=0.407).
Conclusion: Although HL has high cure rates and treatment advancements have improved survival, long-term survivors continue to face increased mortality risk compared to the general US population. Although this finding has been mitigated in recent years, mortality rates remained elevated, with notable variations across demographic groups. Future studies should investigate the underlying pathways of therapy-related late effects (e.g., cellular senescence and aging) to effectively develop interventions and establish risk-based follow-up practices that improve outcomes among HL survivors.
