Development of heart failure in long-term survivors of childhood cancer: results from the cvss study

Abstract Background Long-term survivors of childhood cancer (CCS) are at increased risk for cardiovascular sequelae, including heart failure (HF) as the largest non-malignant contributor to excess death. Yet, little is known about the risk factors, mechanisms of cardiac dysfunction and prevalence of different stages of HF in these patients. Purpose To investigate the development and prevalence of HF phenotypes in CCS compared to the general population. Methods The CVSS study is a prospective single-center cohort study investigating cardiovascular sequelae of CCS. Patients were eligible for the study when they were diagnosed with a neoplasia according to the International Classification of Childhood Cancer (ICCC 3) at an age prior to 15 years between 1980 and 1990, survived more than five years after initial cancer diagnosis and received antineoplastic treatment. Between 2013 and 2016, study individuals underwent a comprehensive, standardized clinical investigation in the CVSS cohort study including echocardiographic examination. HF was categorized as stages A to D according to current HF guidelines of the American Heart Association (AHA). A population-based sample free of cancer (age- and sex matched) serves as control group. Results From 1,002 individuals, all CCS with history of chemo- or radiotherapy and no subsequent neoplasia (n=877) were included (mean age 34.2 (±5.5) years, 44.7% (N=392) female). Age at diagnosis was 6.28 years (±4.24) and the mean interval from the date of diagnosis of cancer to the date of completion of baseline examination 28.5±3.2 years. Based on echocardiographic examination, clinical data and biomarker assessment, 26.6% of CCS were diagnosed with HF stage A, 21.1% with HF stage B and 2.1% with symptomatic HF, i.e. HF stage C/D. Importantly, prevalence of different HF stages varied strongly by specific tumor history. Compared to the population, the prevalence ratio (PR) was 1.16 [95% confidence interval 1.02/1.31] for stage A HF and 1.91 [1.63/2.23] for the composite of stage B to D HF in an age- and sex-adjusted Poisson regression model. Multivariable linear regression with the systolic marker left ventricular ejection fraction as dependent variable and adjustment for tumor entities, age, sex, and cardiovascular risk factors (CVRF) revealed a lower EF in patients with history of bone tumors (β −6.0 [−8.1/−3.0]), soft tissue sarcoma (β −2.2 [−4.1/−0.35]), leukemia (β −0.84 [−1.8/0.08]) and renal tumors (β −1.8 [−4.0/0.27]) compared to the population. In contrast, the same model for the diastolic marker E/E', showed an association only with CVRF, but not with tumor entities. Conclusion The prevalence of stage B to D HF was significantly higher among long-term CCS in the 3rd to 5th age decade compared to the population and varied strongly by tumor entity. Systolic dysfunction was primarily associated with history of tumor entities, whereas diastolic dysfunction was associated with the higher burden of CVRF in CCS. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): This study is funded by the Deutsche Forschungsgemeinschaft (DFG) (SP 1381/2-1&2, FA 1038/2-1&2, WI 3881/2-1&2)