Cardiovascular disease risk in cancer survivors: a population-based cohort study from the UK Biobank, and meta-analysis of cohort studies

Abstract

Introduction In addition to the well-recognised cardiotoxicity of cancer treatment, possible aetiological links between cancer diagnosis and cardiovascular disease (CVD) have gained growing research interests. We aimed to estimate the CVD burden among cancer survivors and illustrate population-level associations between these two conditions.Methods We first conducted a prospective cohort study in the UK Biobank and a meta-analysis of previous population-based cohorts. HRs were estimated in the cohort study to evaluate the effect of cancer diagnosis on the subsequent risk of CVD compared with that of non-cancer individuals. We then systematically searched Pubmed, Embase and Cochrane Library to retrieve previous cohorts. Random-effect meta-analysis was performed to pool relative risk estimates. A combination of multiple statistical metrics was employed to appraise the evidence.Results A total of 39 755 811 participants (5 898 597 cancer survivors vs 33 857 214 cancer-free controls) were identified in our study. In the cohort study, a 51% higher hazard of CVD risk was found among cancer survivors (95% CI 1.48 to 1.55, p<0.001). The hazard decreased to 29% after adjusting competing risk. The meta-analysis identified 104 published cohorts. We found a 1.34-fold increased CVD risk among patients with cancer (95% CI 1.22 to 1.47, p<0.001). The association remains significant among multiple cancer sites and multiple CVD subtypes. This association was consistent, irrespective of chemo or radiotherapy use. Evidence appraisal identified one convincing association between hematologic/lymphatic malignancies and ischaemic heart disease, along with 29 highly suggestive associations.Conclusions Our study provided comprehensive estimates of CVD incidence in cancer survivors and identified a significantly elevated CVD risk among patients with cancer, regardless of chemotherapy or radiotherapy. These findings underscore the need for routine assessment of CVD risk factors at cancer diagnosis to enhance the well-being and survival of patients with cancer.PROSPERO registration numberCRD42022307056