O-RADS plus CA125, HE4, and CEA yields high diagnostic accuracy for early ovarian cancer: an interpretable model

In 674 patients with adnexal masses, an interpretable model integrating O-RADS ultrasound with CA125, HE4, CEA, and maximum tumor diameter detected early ovarian cancer with validation AUC 0.973 (95% CI 0.952–0.994), sensitivity 0.897, specificity 0.980.

• A total of 674 patients with adnexal masses (270 early ovarian cancer, 404 benign) were randomized to training (n=438) and validation (n=236) sets.
• O-RADS alone achieved AUC 0.854; the combined model significantly outperformed the marker-only combination (AUC 0.918) and all individual indices (DeLong all P<0.05).
• SHAP analysis identified O-RADS classification, CA125, and HE4 as the most influential predictors; calibration was robust and decision curve analysis showed net benefit.

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.