Differentiating Borderline from Malignant Ovarian-Adnexal Tumours: A Multimodal Predictive Approach Joining Clinical, Analytic, and MRI Parameters

TL;DR

This study improves the ability to distinguish borderline ovarian tumors from malignant ones using a combination of clinical, blood, and MRI data, reducing misclassification and overtreatment.

Contribution

A new multimodal predictive model enhances diagnostic accuracy for borderline ovarian tumors in the indeterminate O-RADS MRI 4 category.

Findings

The new model increased overall diagnostic accuracy from 0.856 to 0.955 when used alongside O-RADS MRI.

The positive predictive value for borderline tumors in O-RADS MRI 4 improved from 0.49 to 0.90 with the full model.

The model maintains high accuracy for benign and malignant lesions while improving risk stratification in indeterminate cases.

Abstract

Borderline ovarian-adnexal tumours (BOTs) have a much better prognosis than invasive ovarian cancer but are frequently misclassified as malignant by MRI examination applying the “Ovarian-Adnexal Reporting Data System for Magnetic Resonance Imaging (O-RADS MRI)”, especially in score 4. This sometimes leads to overtreatment and potential loss of fertility. In this retrospective single-centre study, we explored whether combining clinical information, blood tumour markers, and MRI features could improve this distinction in indeterminate cases. Our multimodal, simple, rule-based predictive model—used as a second step after O-RADS MRI—significantly improves the diagnostic performance for BOTs. This approach could optimise patient management and directly address a major limitation of current O-RADS MRI classification. Further validation in larger, multicentre studies is required before routine clinical use.

Objectives: To improve the differentiation of borderline ovarian-adnexal tumours (BOTs) from malignant ovarian-adnexal masses, most of which fall into the indeterminate O-RADS MRI 4 category, by developing a multimodal predictive model that integrates clinical, analytic, and MRI parameters. Methods: This retrospective, single-centre study included 248 women who underwent standardised MRI for ovarian-adnexal mass characterisation between 2019 and 2024. Of these, 201 had true ovarian-adnexal masses (114 benign, 22 borderline, and 65 malignant), confirmed by histopathology or stability after ≥12-month follow-up. Forty-one clinical, laboratory, and imaging variables were initially assessed, and after a bivariate evaluation, 18 final predictors with clinical relevance were selected for model construction with thresholds learned from the data. A classification and regression tree (CART) model (“Full Model”) was applied as a second-stage tool after O-RADS MRI scoring, using 10-fold cross-validation to prevent overfitting. A pruned “Simplified Model” was also derived to enhance interpretability. Results: O-RADS MRI performed well at the extremes (scores 2–3 and 5) but showed limited discrimination between BOTs and malignancies within category 4 (PPV for borderline = 0.50). The decision-tree models significantly improved diagnostic performance, increasing overall accuracy from 0.856 with O-RADS MRI alone to 0.905 (Simplified Model) and 0.955 (Full Model). The PPV for BOTs within the intermediate O-RADS MRI 4 category increased from 0.49 with O-RADS MRI alone to 0.77 and 0.90 with the simplified and full models, respectively, while maintaining high accuracy for benign and malignant lesions. Conclusions: In this retrospective single-centre cohort, the addition of an interpretable rule-based predictive model as a second-line tool within O-RADS MRI category 4 was associated with improved discrimination between borderline and invasive malignant ovarian-adnexal tumours. These findings suggest that multimodal integration of clinical, laboratory, and MRI features may help refine risk stratification in indeterminate cases; however, external validation in prospective multicentre cohorts is required before clinical implementation.

Figures