BackgroundCardiovascular disease (CVD) is the leading cause of non‐cancer death in cancer survivors, but the risk of CVD varies between cancers.ObjectivesTo synthesise available evidence on patterns and magnitude of CVD mortality risk.MethodsA systematic search of Medline (OVID), CINAHL and Scopus databases from 01‐January‐2000 to 16‐July‐2023 of studies of people with cancer, reporting CVD mortality in cancer population compared with a reference population (e.g. general population) as standardised mortality ratios (SMR). Meta‐analysis of SMRs across cancer and CVD types were pooled using a random‐effects model to allow for heterogeneity of the true effect size across studies.ResultsWe identified 136 studies from 16 countries. Sample sizes ranged from 157 to 7,529,481. The majority (n = 98; 72%) were conducted in the United States, followed by Europe (n = 22; 16%). The most common cancers studied were gastrointestinal (n = 34 studies), haematological (n = 31) and breast (n = 29). A total of 876 CVD SMRs were extracted across diverse CVD conditions. Of those, the majority (535; 61%) indicated an increased risk of CVD death (SMR >1), 109 (12%) a lower risk of CVD death (SMR <1) and 232 (27%) an equivalent risk (95% CI of SMR included 1) compared to the general population. The meta‐analysis of all reported SMRs showed an increased risk of CVD death (SMR = 1.55, 95% CI = 1.40–1.72) in cancer survivors compared with the general population. The SMR varied between CVD conditions and ranged from 1.36 (95% CI = 1.29–1.44) for heart diseases to 1.56 (95% CI = 1.39–1.76) for cerebrovascular diseases. SMR varied across cancer types, ranging from 1.14 (95% CI = 1.04–1.25) for testicular/germ cell tumours to 2.82 (95% CI = 2.20–3.63) for brain/central nervous system tumours.ConclusionsCancer survivors are at increased risk of premature CVD mortality compared to the general population, but the risk varies by cancer type and CVD. Future research should focus on understanding mechanisms behind the increased CVD risk to develop appropriate interventions.