Key result
Oncology-based machine learning score achieves 0.71 AUC for predicting 10-year CVD in prostate cancer.
Why the study?
Conventional cardiovascular risk scores are poorly calibrated for prostate cancer and require laboratory tests not routinely available in oncology clinics.
Does a machine learning-based risk classification score accurately predict cardiovascular risk in men with prostate cancer?
Cohort (n=7,566)
Does a machine learning-based risk classification score accurately predict cardiovascular risk in men with prostate cancer?
Effect estimate: TDAUC 0.71 (95% CI 0.64-0.78)
A novel machine learning-based cardiovascular risk score using oncology-available variables provides pragmatic risk stratification for patients with prostate cancer.
Signals need for CV risk assessment during ADT; leaves open whether ML models enhance stratification beyond standard scores.
PURPOSE Patients with prostate cancer (PC) face elevated cardiovascular (CV) risk. Conventional CV risk scores are often poorly calibrated for PC and require laboratory tests that are not routinely available in PC clinics, limiting oncology use. METHODS The development cohort (DC) included men with PC (age ≥18 years) diagnosed in 2010-2014. Validation cohort 1 (V1) included men age 45 years and older within 1 year of PC diagnosis or near initiation of androgen deprivation therapy. Validation cohort 2 (V2) included men with PC diagnosed in 2006-2019. Outcomes were cardiovascular disease (CVD), atherosclerotic CVD (ASCVD), and heart failure (HF). Predictors were selected with XGBoost and LASSO, modeled with Fine-Gray and penalized Cox regression, and scaled to additive point scores (low/intermediate/high risk). Performance was assessed using time-dependent AUC (TDAUC), which quantifies the model's ability to correctly rank patients by risk at each time point, and compared with conventional scores. RESULTS The DC included 1,815 patients, and V1 and V2 included 4,022 and 1,729 patients, respectively (median follow-up 10.3 years). At 10 years, cumulative incidence was 15.0% for composite CVD, 9.4% for ASCVD, and 10.1% for HF. The composite CVD score (age, race, smoking history, high-risk Gleason score, and household members) achieved a 10-year TDAUC of 0.71 (95% CI, 0.64 to 0.78) in the DC and 0.59 (95% CI, 0.52 to 0.65) in V2, with a 2-year TDAUC of 0.66 (95% CI, 0.56 to 0.75) in V1. In the DC, ASCVD and HF scores achieved 10-year TDAUCs of 0.66 (95% CI, 0.58 to 0.75) and 0.70 (95% CI, 0.62 to 0.79), respectively, and retained discrimination in validation. CONCLUSION To our knowledge, these are the first PC-specific CV risk scores derived from oncology-available variables, supporting pragmatic CV risk stratification in PC clinics and motivating further validation and implementation studies.
No takes yet. Share an insight, caveat, or question.
Stabellini et al. (2026) conducted a cohort in Prostate cancer (n=7,566). Machine learning-based cardiovascular risk score vs. Conventional cardiovascular risk scores was evaluated on Composite cardiovascular disease (CVD) (TDAUC 0.71, 95% CI 0.64-0.78). A machine learning-based cardiovascular risk score using oncology-available variables achieved a 10-year time-dependent AUC of 0.71 (95% CI, 0.64-0.78) for composite CVD in men with prostate cancer.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: