Altaf A, Munir MM, Endo Y, Khan MMM, Rashid Z, Khalil M, Guglielmi A, Aldrighetti L, Bauer TW, Marques HP, Martel G, Lam V, Weiss MJ, Fields RC, Poultsides G, Maithel SK, Endo I, Pawlik TM
OBJECTIVE : We sought to develop an artificial intelligence (AI) based model to predict early recurrence (ER) following curative-intent resection of neuroendocrine liver metastases (NELM).
METHODS : Patients with NELM, who underwent resection were identified from a multi-institutional database. ER was defined as recurrence within 12 months of surgery. Different AI based models were developed to predict ER using 10 clinicopathological factors.
RESULTS : Overall, 473 NELM patients were included. Among 284 (60.0%) patients with recurrence, 118 (41.5%) patients developed an ER. An ensemble AI model demonstrated the highest area under ROC curves (AUC) of 0.763 and 0.716 in the training and testing cohorts, respectively. Maximum diameter of the primary neuroendocrine tumor, NELM radiologic tumor burden score (TBS), and bilateral liver involvement were the factors most strongly associated with risk of NELM ER. Patients predicted to develop ER had worse 5-year recurrence free survival and overall survival (21.4% vs. 37.1%; p=0.002 and 61.6% vs. 90.3%; p=0.03, respectively) versus patients not predicted to recur. An easy-to-use tool was made available online: (https://altaf-pawlik-nelm-earlyrecurrence-calculator.streamlit.app/).
CONCLUSION : An AI-based model demonstrated excellent discrimination to predict ER of NELM following resection. The model may help identify patients who may benefit the most from curative-intent resection, risk-stratify patients according to prognosis, and guide tailored surveillance and treatment decisions, including consideration of non-surgical treatment options.