Splenectomy during cytoreductive surgery in advanced epithelial ovarian cancer can be predicted

Introduction

Splenectomy may be necessary for complete cytoreductive surgery (CRS) in advanced stage epithelial ovarian cancer (AS-EOC), potentially raising perioperative morbidity and necessitating specific patient management.

Objective

This study aimed to develop a predictive score of splenectomy in CRS of AS-EOC.

Materials and methods

Data from histologically confirmed AS-EOC (FIGO IIB-IV) before CRS and diagnosed between 01/01/2000 and 01/06/2017, were extracted from the FRANCOGYN multicentric database (14 French hospitals). After identifying predictive factors of splenectomy, we performed a logistic regression to develop a prediction model and construct a risk score, allowing identification of a high-risk group. Model discrimination was assessed using a Receiver Operating Characteristic (ROC) curve. Decision Curve Analysis (DCA) was then conducted to evaluate the model’s net clinical benefit across a range of threshold probabilities.

Results

Among 1288 patients included, 7 % (n = 91) underwent splenectomy. Four independent variables statistically associated with splenectomy were identified: age < 60 years (aDOR = 1.76, 95 % CI [1.13–2.75], p = 0.015), omental cake (aDOR = 2.12, 95 % CI [1.11–4.08], p = 0.024), diaphragmatic carcinosis (aDOR = 2.36, 95 % CI [1.34–4.18], p = 0.001), and digestive involvement at initial CT and/or laparoscopy (aDOR = 3.24, 95 % CI [1.93–5.43], p < 0.001). The ROC-AUC of this prediction model was 0.76. Patients meeting all 4 criteria with a maximum of 10 points defined the high-risk group and had a splenectomy probability of 32 % (95 % CI [22.00–44.31]), with a specificity of 95.8 % (95 % CI [94.5–96.9]) and a positive likelihood ratio of 6.31 (95 % CI [4.08–9.78]). The DCA showed a positive net clinical benefit of the model between 15 % and 40 % threshold probabilities.

Conclusion

Using a simple 4 – variable predictive score, patients at high risk of splenectomy during CRS in AS-EOC could be identified to improve patients’ preoperative information and perioperative management.