Chemotherapy Response Score and Survival Outcomes After Neoadjuvant Chemotherapy in High-Grade Serous Tubo-Ovarian...

Background and Objectives: In advanced high-grade serous tubo-ovarian carcinoma (HGSOC), the chemotherapy response score (CRS) assesses histopathologic response to neoadjuvant chemotherapy (NACT). CRS1 and CRS2 are commonly grouped for similar prognostic outcomes, but whether partial response carries prognostic value and which factors predict pathologic response remain unclear. This multicenter study evaluated predictors of pathologic response and the association of CRS with recurrence-free survival (RFS) and overall survival (OS) after NACT followed by interval debulking surgery (IDS). Materials and Methods: We retrospectively analyzed 157 patients with FIGO stage IIIC-IV HGSOC treated with NACT followed by IDS between 2015 and 2021. Pathologic response was defined as CRS2-3 versus CRS1; logistic regression identified predictors of response. Survival was assessed using Kaplan-Meier analysis, and Cox regression evaluated factors associated with RFS and OS. Secondary analyses evaluated the conventional CRS1-2 versus CRS3 classification, direct CRS3-versus-CRS2 contrasts, and ordinal trends for RFS and OS. Results: CRS1, CRS2, and CRS3 were observed in 33 (21.0%), 97 (61.8%), and 27 (17.2%) patients, respectively. Median RFS was 10.4, 16.4, and 24.1 months for CRS1, CRS2, and CRS3, respectively (p < 0.001). Median OS was 21.2, 39.1, and 47.0 months, respectively (p < 0.001). In multivariate Cox analysis, CRS2 and CRS3 were independently associated with better RFS (HR 0.36 and 0.24) and OS (HR 0.48 and 0.41) compared with CRS1. In the conventional CRS1-2 versus CRS3 model, CRS3 remained associated with lower recurrence risk after multivariate adjustment (HR 0.60; p = 0.042), but not OS (HR 0.74; p = 0.354). Direct CRS3-versus-CRS2 contrasts were nonsignificant for RFS and OS (p = 0.131 and p = 0.628), whereas ordinal trends toward lower hazards were significant for both endpoints (p = 0.001 and p = 0.041). A higher CA-125 decrease rate (OR 1.04) and more than three NACT cycles (OR 4.89) independently predicted pathologic response, whereas FIGO stage IV predicted a lower response (OR 0.15). Conclusions: Partial pathological response (CRS2) was independently associated with improved RFS and OS compared with no or minimal response (CRS1), suggesting that prognostic information is not confined to complete or near-complete pathological response. The conventional CRS1-2 versus CRS3 classification remained independently prognostic for RFS after multivariate adjustment but did not reach statistical significance for OS, while direct CRS3-versus-CRS2 contrasts were nonsignificant for both outcomes. Thus, separate evaluation of CRS2 may provide additional RFS stratification within CRS1-2. However, the incremental contribution of CRS3 beyond CRS2 could not be established in the present cohort and requires validation in larger cohorts.