The preoperative recurrence score: Predicting early recurrence in peri-hilar cholangiocarcinoma

Introduction: Despite advances in surgical techniques, the rate of early recurrence in perihilar cholangiocarcinoma (PCC) remains high. We sought to develop the Preoperative Recurrence Score (PRS), a model to estimate the risk of early recurrence after resection based on preoperative radiological characteristics. Materials and methods: Data of patients who underwent surgery for PCC were retrospectively collected, and preoperative imaging was reviewed to assess tumor characteristics. A model to assess the risk of early recurrence based on preoperative radiologic characteristics was internally developed and externally validated on two cohorts of patients from two European major hepatobiliary surgery referral centers. Results: A total of 215 patients among three different patient cohorts were included in the study. Tumor size ≥18 mm (HR 2.70, 95 % CI 1.48-4.92, p = 0.001), macroscopic portal vein involvement (HR 2.28, 95%CI 1.19-4.34, p = 0.013), hepatic arteries involvement (HR 2.44, 95%CI 1.26-4.71, p = 0.008), and presence of suspicious lymph nodes (HR 1.98, 95%CI 1.02-3.83, p = 0.043) were significantly associated with recurrence-free survival (RFS). The model showed excellent discrimination both on the internal (AUC 0.83) and external validation cohorts (external 1: AUC 0.84; external 2: AUC 0.70). High PRS was associated with worse RFS among all three cohorts, with a 1-year recurrence probability of 80.1 %, 100.0 %, and 54.2 % in the internal and external validation cohorts 1 and 2, respectively. Conclusions: The PRS is a simple tool that can accurately assess the risk of early recurrence in patients with PCC. Up-front surgery should be carefully evaluated in patients with high PRS, as it could result in a futile resection.