Cardiovascular Phenotypes Across Major Cancer Types: Distribution and Association with All-Cause Mortality in a...

Background/Objectives: Cardiovascular comorbidities are frequent in patients with solid malignancies, but their distribution across tumor types and their association with recorded all-cause mortality remain insufficiently defined. This study assessed clinically defined cardiovascular phenotypes across major cancer groups and examined their association with recorded all-cause mortality. Methods: We performed a retrospective cohort study of 2020 consecutive adult patients hospitalized with malignancies at University Hospital "Tsaritsa Yoanna-ISUL" between September 2023 and December 2025. Patients were classified as having respiratory, gastrointestinal, genitourinary, breast, or other solid malignancies. Four non-mutually exclusive cardiovascular phenotypes were predefined: atherosclerotic, cardiometabolic, structural heart disease, and thromboembolic. Mortality status was ascertained through the National Health Insurance Fund database and hospital electronic records. As complete time-to-event data were unavailable, mortality was analyzed as a binary outcome using multivariable logistic regression. Results: Recorded all-cause mortality occurred in 471 patients (23.3%) and differed significantly across solid tumor types (p < 0.001). Mortality was highest in respiratory cancers (42.3%), followed by gastrointestinal cancers (30.2%), other malignancies (23.5%), genitourinary cancers (14.9%), and breast cancer (9.1%). Cardiovascular phenotypes also differed across tumor groups. In the fully adjusted model, structural heart disease was associated with recorded all-cause mortality status after multivariable adjustment (OR 1.61, 95% CI 1.16-2.23, p = 0.004). The atherosclerotic phenotype showed a borderline association (OR 1.42, 95% CI 1.00-2.02, p = 0.051), whereas cardiometabolic and thromboembolic phenotypes were not associated with recorded mortality after multivariable adjustment. Model AUC was 0.771. When cardiovascular burden was modeled continuously, each additional cardiovascular factor was associated with higher odds of recorded all-cause mortality status (OR 1.12, 95% CI 1.03-1.21, p = 0.005). Conclusions: Cardiovascular profiles differed across cancer types. Structural heart disease was the most consistent cardiovascular phenotype associated with recorded all-cause mortality, while overall cardiovascular burden provided additional but modest mortality-related information. In our cohort, routinely recorded cardiovascular diagnoses helped characterize mortality-related clinical profiles. However, these findings should be viewed as observational and hypothesis-generating, and further longitudinal studies are needed before any clinical use can be recommended.