Article Text

Download PDFPDF
466 Precision surgery meets artificial intelligence: upper abdominal peritonectomy‘s key role in advanced ovarian cancer cytoreduction
  1. Alexandros Laios1,
  2. Marios Evangelos Mamalis2,
  3. Evangelos Kalampokis2,
  4. Amudha Thangavelu1,
  5. Timothy Broadhead1,
  6. David Nugent1 and
  7. Diederick Dejong1
  1. 1Department of Gynaecologic Oncology, ESGO Centre of Excellence for ovarian cancer surgery, St James’s University Hospital, Leeds, UK
  2. 2Information Systems Lab, Department of Business Administration, University of Macedonia, Thessaloniki, Greece


Introduction/Background The Surgical Complexity Score (SCS) has been widely used to characterize the surgical effort in the cytoreduction of advanced-stage epithelial ovarian cancer (EOC). It effectively integrates numerical data with the intricacy of the individual sub-procedures. Nevertheless, not all such procedures are encompassed by this score. It is necessary to define the relevance of each of these surgical sub-procedures in the context of complete cytoreduction (CC0). Explainable Artificial Intelligence (XAI) holds promise to elucidate how real-time features impact the prediction of CC0.

Methodology We analysed data from 560 surgically cytoreduced advanced-stage EOC patients at a UK tertiary referral centre. To ensure a comprehensive assessment of surgical sub-procedures, we customarily tailored the ESGO ovarian cancer report template. Our modelling approach involved the application of the XGBoost algorithm, coupled with the implementation of the SHAP framework to enhance cohort interpretability. For survival analysis, we employed Cox regression and generated Kaplan-Meier curves to visualize the results.

Results The XGBoost model demonstrated a satisfactory level of accuracy in predicting CC0 (AUC=0.70). When we visually assessed the importance of surgical sub-procedures for CC0 prediction, upper abdominal peritonectomy (UAP) emerged as the most critical factor, followed closely by regional lymphadenectomies. Incorporating UAP into a composite model, alongside engineered features, significantly bolstered its predictive performance (AUC = 0.80). UAP was predictive of poorer progression-free survival (Hazard Ratio [HR] = 1.76, 95% Confidence Interval [CI] 1.14–2.70, p: 0.01), although it did not significantly impact overall survival (HR = 1.06, CI 0.56–1.99, p: 0.86). The Surgical Complexity Score (SCS) did not exhibit a significant effect on survival outcomes.

Conclusion Our study identified UAP as the foremost procedural indicator for achieving CC0 in advanced-stage EOC women undergoing surgical cytoreduction (figure 1). It underscores the importance of meticulous assessment of the upper abdominal quadrants to ensure the attainability of CC0.

Disclosures There are non conflicts of interest.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.