Interpretable AI to predict anal cancer local failure at 3 years with the FFCD Anabase prospective multicentric...

Machine learning (ML) has transformed oncological risk prediction by enabling personalized therapeutic strategies. Local tumor control remains a critical endpoint in anal cancer management. This study aimed to develop and validate an explainable ML model for predicting local recurrence at 3 years in patients with anal cancer.

We analyzed data from the prospective multicentric FFCD-Anabase cohort, comprising 1,015 patients with anal cancer treated with chemoradiotherapy across 60 French centers between January 2015 and April 2020. The endpoint was local recurrence at 3 years. An extreme gradient boosting model with an Accelerated Failure Time extension was developed to handle time-to-event data. Model inputs combined routinely available clinical, biological, and treatment variables, selected on the training set only. Model development incorporated cross-validation for hyperparameter optimization, followed by calibration to ensure reliable probability estimates. Performance was assessed on an independent test set using discrimination, calibration metrics, and clinical utility measures, with right-censored patients retained through inverse-probability-of-censoring weighting. Model interpretability was enhanced using Shapley Additive exPlanations values, providing global feature importance and individual patient-level prediction explanations.

The model demonstrated a C-index of 0.735 (95 % CI 0.65-0.82) and a time-dependent AUC at 3 years of 0.755 (0.66-0.85). The model achieved a sensitivity of 64 % and specificity of 74 %, with positive and negative predictive values of 39 % and 89 %, respectively. All 3-year classification metrics were computed on the full test set using inverse-probability-of-censoring weighting, so that censored patients were not discarded. Overall accuracy was 72 % with an F1-score of 0.49. Calibration performance was assessed using Brier score (0.135) and integrated Brier score (0.131). Calibration plots demonstrated good agreement between predicted and observed probabilities. Decision curve analysis revealed net clinical benefit across a range of threshold probabilities, with optimal risk stratification at a 37 % probability threshold for distinguishing low- and high-risk patients. Kaplan-Meier survival analysis confirmed statistically significant differences between risk groups (p = 2.78 × 10⁻5). Model interpretability analysis using SHAP value (SHapley Additive exPlanations, a method that quantifies each feature's contribution to a model's prediction) identified World Health Organization (WHO) performance status as the most influential predictor, followed by tumor size and age.

Our model yielded good performances on real-world data to predict the risk of local recurrence at 3 years for anal cancer.