Introduction: With improved survival of patients with cardiovascular disease (CVD), the risks of non-CVD outcomes is of increasing interest. We investigated the risk of subsequent cancer in persons with clinical CVD compared to those without using a matched cohort study nested in the ARIC study.
Hypothesis: Persons with CVD may have higher risk of cancer compared to those without.
Methods: In ARIC participants (aged 45-64 years) without prevalent CVD or cancer at the study baseline (1987-1989), we first identified every participant who developed clinical CVD (myocardial infarction [MI], heart failure, or stroke as a combined outcome and separately) during follow-up (through 2012). For each of them, applying incidence density sampling, we selected up to two participants without clinical CVD matched on key confounders (age, sex, race, diabetes, hypertension, lipid-lowering therapy, and smoking). Then, we used the Cox proportional hazards regression to estimate the risk of total and site-specific cancer comparing CVD to no CVD further adjusting for conventional risk factors for CVD or cancer, socioeconomic status, insurance, and routine physical examination.
Results: We followed 2,565 with CVD and 4,622 without CVD from the date of sampling for a total of 62,772 person-years (mean of 9 years). The hazard ratio [HR] for total cancer risk was 1.13 (95% confidence interval, 1.00-1.32) for participants with CVD compared to those without. None of the HRs for site-specific cancers were statistically significant. When individual CVD subtypes were analyzed separately, participants with MI and heart failure had 2.1- and 1.3- folds significantly higher risk of total cancer compared to those without, respectively. Participants with MI had significantly increased HR in major types of cancer such as prostate, lung, and breast, with adjusted HR ranging from 2.5 to 3.5.
Conclusion: Persons with clinical CVD, especially MI, had a modestly increased risk of total cancer following that event compared to those without. Future studies are warranted to better understand its potential reasons.