Diagnostic performance of the IOTA SR xCA-125 model in predicting ovarian malignancy in women undergoing surgery for adnexal masses.
Garima Yadav205
Background: The preoperative differentiation of adnexal masses is crucial for appropriate patient management. While ultrasound is the primary imaging modality, standardized risk models are needed to improve accuracy and reduce operator dependency. The International Ovarian Tumor Analysis (IOTA) group has developed models like the Simple Rules Risk (SRR) assessment. This study aims to evaluate the diagnostic performance of the IOTA SRR model and a model combining the Simple Rules with serum CA-125, stratified by menopausal status (SRR-CA125). The primary objective was to study the diagnostic performance of the IOTA SRR-CA125 model in differentiating benign versus malignant/borderline ovarian neoplasms preoperatively in women undergoing surgery for adnexal masses. Methods: This was a prospective observational study conducted over a two-year period. The study included 139 women with an adnexal mass who were planned for surgery, excluding pregnant patients. A detailed history and clinical examination were performed, followed by standardized transvaginal and/or transabdominal ultrasonography. Serum CA-125 levels were analyzed for all patients. The malignancy risk for each adnexal mass was calculated using the IOTA SRR model and the stratified SRR-CA125 model. The final histopathological diagnosis was considered the gold standard. Results: The IOTA SRR model demonstrated excellent overall performance with an Area Under the ROC Curve (AUROC) of 0.953. At an optimal cutoff of ≥10.1%, it achieved a sensitivity of 94.4%, specificity of 83.5%, a Positive Predictive Value (PPV) of 66.7%, and a Negative Predictive Value (NPV) of 97.7%. The IOTA SRR-CA125 model showed distinct advantages when stratified: In pre-menopausal women, it yielded an AUROC of 0.936, with a sensitivity of 95.2% and a significantly improved specificity of 89.2%. In post-menopausal women, it yielded an AUROC of 0.907, with a lower sensitivity of 76.2% but an excellent specificity of 95.0%. Conclusions: Both the IOTA SRR and SRR-CA125 models are highly effective tools for the preoperative risk stratification of adnexal masses. The IOTA SRR model is an excellent, simple, and highly sensitive first-line test. The IOTA SRR-CA125 model provides a more nuanced assessment that significantly enhances specificity, making it valuable for reducing false-positive results. However, this benefit is coupled with a trade-off of lower sensitivity in post-menopausal women, highlighting the importance of selecting the appropriate model based on the clinical context and menopausal status.