A nomogram to predict platinum-sensitivity and survival outcome in women with advanced epithelial ovarian cancer

Abstract

Objective: This study presents the development and validation of a nomogram aimed at predicting platinum-sensitivity and survival outcomes in women with advanced epithelial ovarian cancer (EOC). Materials and methods: Data from a retrospective cohort of women diagnosed with stage III/IV EOC between Jan 2011 and Dec 2021 treated at our institute were collected. Clinical and pathological characteristics were analyzed using logistic regression analysis to identify independent predictors of platinum-sensitivity. Impact on progression-free (PFS) and overall survival (OS) was determined by Kaplan–Meier and Cox regression analysis. A nomogram was constructed based on the significant predictors, and its performance was evaluated using calibration, discrimination, and validation analyses. Results: Of the 210 patients, 139 (66.19%) had platinum-sensitive and 71 (33.81%) were platinum-resistant disease. On multivariate analysis, platinum-resistance correlated with neoadjuvant chemotherapy (OR 2.15; 95% CI 1.10–4.21), clear cell/mucinous histology (OR 5.04; 95% CI 2.20–11.54), and sub-optimal debulking status (OR 3.37; 95% CI 1.44–7.91). Median PFS and OS were also significantly shorter for patients with neoadjuvant chemotherapy (23 vs. 10 months and 69 vs. 29 months, respectively), clear cell/mucinous histology (15 vs. 3 months and 63 vs. 11 months, respectively), and suboptimal debulking (26 vs. 5 months and 78 vs. 24 months, respectively). The nomogram demonstrated good predictive accuracy for platinum-sensitivity in the cohort as indicated by high concordance index of 0.745. Calibration plots showed excellent agreement and internal validation further confirmed the reliability of the nomogram's performance. Conclusion: A novel predictive nomogram based on type of initial treatment, histology, and debulking status was developed, which provides a friendly and reliable tool for predicting platinum-sensitivity and survival outcomes in women with advanced EOC. Its application may assist clinicians in individualizing treatment decisions.

Keywords

• Advanced ovarian cancer
• Platinum-sensitivity
• Prediction model