Analysis of prognostic factors in uterine cervical malignancy treated with concurrent chemoradiation therapy: A retrospective study in two centers in Korea.

125 Background: Cervical cancer is prevalent in most developing countries, and immune checkpoint inhibitors are very expensive to be standard treatment in limited-resource conditions. In these circumstances, concurrent chemoradiation therapy (CCRT) remains a preferred treatment. This study aims to evaluate prognosis and identify prognostic markers in cervical cancer patients treated with CCRT in Korea. Methods: This retrospective study utilized data from two centers (KNUH and KNUCH) January 2012 to September 2023. Disease stage was determined based on the FIGO 2018 system, including patients between stages IIB to IIIC-2. PET/CT assessed lymph node status. All patients received CCRT as the first-line treatment with cisplatin (40 to 60 mg/m 2 ) chemotherapy and EBRT with or without ICR, guided by gynecologic radiation oncologist. Pelvic MRI or CT within 5 months after CCRT was used to evaluate radiosensitivity, including the size of the residual primary cervical tumor. The primary endpoint was 5-year disease-free survival (DFS), determined via systemic imaging studies like concurrent chest and abdomen CT or PET/CT. Median DFS was evaluated using Kaplan-Meier survival analysis. Prognostic factors were stratified and analyzed using Cox’s proportional hazard ratio regression. Multivariate analysis was conducted using only significant factors identified in the univariate analysis. Receiver operating characteristic curve and Youden’s index were used to determine the cutoff value. Results: Ninety-seven patients were enrolled. The median DFS within 5 years was not reached; mean DFS was 41.5 months (95% CI=36.2–46.8). Multivariate Cox’s proportional hazard regression identified two significant clinical factors for 5-year DFS; residual primary tumor size after CCRT and initial serum CA-125 level (p=0.001, p=0.007, respectively). The cutoff values were 1.85 cm for residual primary tumor size and 24.650 U/mL for CA-125 to predict the 5-year DFS rate (CR or NED vs. PR, SD, PD or recurrence). Sensitivity, specificity, positive predictive value, and negative predictive value were calculated for residual tumor size (36.4%, 98.4%, 92.3%, 73.4%, respectively) and CA-125 (67.9%, 60.7%, 46.3%, and 79.1%, respectively). Conclusions: The size of the residual primary tumor or initial serum CA-125 level can be useful in predicting 5-year DFS rate after CCRT. [Table: see text]