Luigi Sanasi, Antonio Iaconelli, Danila Azzolina, Pasquale Dolce, Christian Cardile, Lorenzo Cacioli, Alessandro Forni, Matteo Busti, Luca Sensini, Nicola Riccardo Pugliese, Domenico D'Amario, Giovanni Pecorini, Mattia Galli, Giovanna Liuzzo, Francesco Burzotta, Riccardo Rinaldi, Moustafa I Morsy, Pasquale Maffia, Pierpaolo Pellicori, Jocelyn M Friday, Mark Petrie, Bernhard Haring, Michael Böhm, Markus S Anker, Javed Butler, Filippo Crea, John G F Cleland
HF is associated with a higher incidence of cancer, although with marked heterogeneity among studies and cancer types. Mortality and cause of death were rarely reported. The observed HF-cancer association is influenced by statistical methodology, shared risk factors, differences in surveillance intensity and cancer type, that confound analyses investigating possible causal biological links. More granular and harmonised studies are required.
BACKGROUND: Cancer and heart failure (HF) frequently coexist due to population ageing and improved survival for both conditions. HF is a well-recognised complication of oncologic therapies and a potential association between HF and subsequent cancer incidence has been recently noticed. We therefore conducted a meta-analysis to quantify cancer incidence and mortality in patients with and without HF.
METHODS: Databases were searched from inception to 15 February 2026. Data from observational studies and randomised trials on individuals with and without HF and reporting cancer incidence and/or mortality during follow-up were included. The main outcome of interest was the incidence of any cancer; secondary outcomes included site-specific cancer incidence (lung, colorectal, breast, and prostate) and all-cause, cardiovascular, and cancer-related mortality.
RESULTS: Twelve studies comprising 8,979,195 individuals were included. HF was associated with a higher incidence of cancer (hazard ratio [HR] 1.33, 95% confidence interval [CI] 1.14-1.54), with substantial heterogeneity (I2 = 99.7%). Cancer site-specific analyses showed increased incidences of lung (HR 1.70, 95% CI 1.26-2.29), colorectal (HR 1.28, 95% CI 1.11-1.48), and breast (HR 1.19, 95% CI 1.04-1.37), but not prostate, cancer. Effect estimates varied according to analytical strategy: matched HF and non-HF cohorts reported higher risk estimates than those using covariate adjustment without matching. Mortality data were sparse and heterogeneous.
CONCLUSIONS: HF is associated with a higher incidence of cancer, although with marked heterogeneity among studies and cancer types. Mortality and cause of death were rarely reported. The observed HF-cancer association is influenced by statistical methodology, shared risk factors, differences in surveillance intensity and cancer type, that confound analyses investigating possible causal biological links. More granular and harmonised studies are required.