ISHL12 Abstract 127

The Effect of Heart Failure on Management and Outcomes of Older Patients with Hodgkin Lymphoma

Background: Anthracycline-containing regimens are recommended as first-line treatment for Hodgkin lymphoma (HL); however, the management and outcomes of patients with cardiomyopathy or heart failure (HF) at the time of lymphoma diagnosis is not known.

Methods: US Surveillance, Epidemiology, and End Results (SEER)-Medicare data from 1999-2016 were used to identify newly diagnosed HL in patients 66 years and older with one year of Medicare A and B prior to lymphoma diagnosis. Prevalent HF and comorbidities in the year prior to lymphoma diagnosis were identified using International Classification of Diseases codes. Cancer treatment, including doxorubicin and cardioprotective medications (i.e., liposomal doxorubicin and dexrazoxane) were assessed using Healthcare Common Procedure Coding Systems codes. Cause of death was defined using the SEER Cause of Death Recodes. The association between prevalent HF and cancer treatment was estimated using logistic regression with adjustment for comorbidities, social determinants of health and hospital level variables. The association between prevalent HF and cause specific mortality was evaluated using competing risk Cox proportional hazards models with sequential adjustment for comorbidities and cancer treatment.

Results: Among 3,348 individuals with newly diagnosed HL, prevalent HF was present in 13.1%. Patients with prevalent HF were less likely to be treated with an anthracycline in the first year after diagnosis (OR 0.42, 95% CI 0.29, 0.60). Among patients with HF who received an anthracycline, dexrazoxane or liposomal doxorubicin was used in only 4.5%. For those with prevalent HF, 1-year lymphoma mortality was 37.4% (95% CI 35.5, 39.5%) [Fig]. In multivariable models adjusting for clinical covariates, prevalent HF was associated with higher lymphoma mortality (HR 1.21, 95% CI 1.03, 1.41); however, the effect of prevalent HF on lymphoma mortality was no longer significant when adding cancer treatment variables to the model (HR 1.05, 95% CI 0.71, 1.56) [Fig]. Prevalent HF was also associated with cardiovascular mortality in fully adjusted models (HR 1.77, 95% CI 1.34, 2.32).

Conclusion: HF is common in older patients with HL and is associated with less anthracycline use and higher lymphoma mortality. Dexrazoxane and liposomal doxorubicin are used infrequently. Future randomized trials are needed to determine strategies to decrease lymphoma and cardiac mortality in this high-risk population.