Key result
RISKHEART machine learning model predicts 10-year cardiovascular mortality in gynecologic cancer survivors with ~0.79 AUC.
Why the study?
General-population cardiovascular disease risk scores underperform in cancer survivors by failing to capture treatment-related cardiotoxicity, particularly in gynecologic cancer survivors who have distinct phenotypes defined by older age and heterogeneous therapies.
Does the RISKHEART machine learning model improve prediction of 5-year and 10-year cardiovascular mortality in women with primary gynecologic malignancies?
Cohort (n=104,384)
Yes
Does the RISKHEART machine learning model improve prediction of 5-year and 10-year cardiovascular mortality in women with primary gynecologic malignancies?
Effect estimate: AUC 0.786
The RISKHEART machine learning model provides clinically meaningful discrimination for predicting 5- and 10-year cardiovascular mortality in gynecologic cancer survivors, outperforming general-population risk scores.
No takes yet. Share an insight, caveat, or question.
May support tailored survivorship care in gynecologic cancer survivors; leaves open prospective validation of RISKHEART.
Silva et al. (2026) conducted a cohort in Gynecologic cancer (n=104,384). RISKHEART machine learning model was evaluated on 10-year cardiovascular mortality (AUC 0.786). The RISKHEART machine learning model demonstrated good discrimination for predicting 10-year (AUC 0.786) and 5-year (AUC 0.796) cardiovascular mortality in gynecologic cancer survivors.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: