Mortality after Diagnosis of Cardiovascular Disease in Survivors of Adolescent and Young Adult Lymphoma

Background: Due to cardiotoxic treatment, cardiac dysfunction is a major morbidity among survivors of adolescent and young adult (AYA) cancer, defined as a primary cancer diagnosis between the ages of 15-39. Although high mortality after cardiovascular disease (CVD) diagnosis has been shown in other cancer survivor populations, this has not been investigated in survivors of AYA cancer. Furthermore, these relationships have been described in mostly White populations, with little focus on individuals from diverse backgrounds. Objective: To evaluate overall survival in survivors of AYA Hodgkin lymphoma (HL) and non-Hodgkin lymphoma (NHL) after a diagnosis of CVD and to investigate differences in the CVD-survival relationship by race/ethnicity. Methods: Medical records were reviewed for 825 patients treated at MD Anderson Cancer Center between 2000-2016 who had a primary lymphoma cancer diagnosis between the ages of 15-39. Data abstracted included anthropometric characteristics, demographics, treatment course, vital status, and CVD diagnoses occurring at least 3 months after diagnosis of cancer. CVD diagnoses collected included coronary artery disease, myocardial infarction, angina pectoris, cardiomyopathy, congestive heart failure, pericardial effusion, pericarditis or myocarditis, valvular heart disease, arrhythmia, and transient ischemic attack. Overall survival (OS) was the duration from date of diagnosis to last visit or day of death. OS rates and age at death were calculated based on CVD status. Kaplan-Meier curves were used to estimate survival durations over time by CVD status, and Cox proportional hazards models were fitted to evaluate main effect and stratified associations with OS. Models were adjusted for age at diagnosis, cancer type, sex, race/ethnicity, anthracycline dose > 300 mg/m2, chest radiation exposure, stem cell transplant status, and baseline BMI. Results: Of the 825 patients included in the analysis, 73.7% (608) had HL and 26.3 % (217) had NHL. The most common subtypes of NHL were diffuse large B-cell lymphoma (69.6%) and follicular lymphoma (18.2%). The median age at diagnosis was 28.0 years. 58.5% of patients were White, 26.9% Hispanic, and 14.5% Black. Nearly 25% (23.6%; N=195) had a CVD diagnosis after their cancer diagnosis during the median follow-up time of 9.8 years with the median age at CVD diagnosis being 33.0 years. Patients with CVD had worse survival outcomes than those without (HR: 2.61, 95% CI: [1.70-4.02]). In univariate analysis of risk of death among those with CVD, NHL diagnosis, Black race, underweight at baseline, transplant recipients, and those without chest radiation exposure significantly increased risk factors for death after CVD. However, in multivariable analyses, only transplant status was an independent risk factor. This adverse impact of CVD on survival was consistently observed among patient subgroups when stratified by race/ethnicity, cancer subtype, baseline BMI, and treatment exposures. Of note, the largest effect conferred by CVD on survival when compared to no CVD (HR > 4.0) was among Black patients, NHL survivors, those with a healthy baseline BMI, individuals who did not receive a transplant, patients with an anthracycline dose ≤ 300 mg/m², and those without chest radiation exposure. Conclusion: Survival outcomes were inferior for survivors of AYA lymphoma diagnosed with CVD subsequent to their cancer diagnosis. Our analysis suggests that this burden may not be shared equally based on race/ethnicity and other patient subpopulations.