Clinicopathologic Predictors of Recurrence and Postrecurrence Survival Following Curative Resection of Perihilar Cholangiocarcinoma: Insights from a 24-Year Institutional Cohort

Document Type

Conference Proceeding

Publication Date

2-20-2026

Publication Title

Ann Surg Oncol

Keywords

adult, aged, cohort analysis, conference abstract, controlled study, female, follow up, hepatectomy, hepatobiliary system cancer, human, Klatskin tumor, lymph node metastasis, major clinical study, male, overall survival, primary sclerosing cholangitis, radiotherapy, recurrence free survival, recurrence risk, recurrent disease, retrospective study, surgery, surgical margin, tumor volume

Abstract

Perihilar cholangiocarcinoma (pCCA) remains one of the most aggressive hepatobiliary malignancies, with high recurrence rates even after curative-intent resection. Contemporary data defining recurrence patterns, clinicopathologic predictors, and post-recurrence outcomes in a homogeneous surgical cohort are limited. METHODS: We conducted a 24-year retrospective cohort study of patients undergoing curative-intent liver resection for de novo pCCA. Patients requiring vascular reconstruction, transplantation, or with primary sclerosing cholangitis were excluded. Clinicopathologic features, recurrence patterns, recurrence-free survival (RFS), overall survival (OS), and post-recurrence survival (PRS) were analyzed using Kaplan-Meier and multivariable Cox regression. RESULTS: Among 163 patients, the median follow-up was 7 years. The median RFS was 4.7 years (95% CI, 3.6-5.9), with 1-, 3-, and 5-year rates of 78.5%, 68.1%, and 58.9%. The median OS was 5.4 years (95% CI, 3.6-7.2), with 1-, 3-, and 5-year rates of 85.3%, 68.1%, and 58.9%. Recurrence occurred in 66 patients (40.5%) at a median of 1.5 years, most commonly distant (57.6%) versus locoregional (33.3%). On multivariable analysis, lymph node positivity (HR 2.06, 95% CI 1.17-3.61, p=0.012) and increasing tumor size (HR 1.15 per cm, 95% CI 1.02-1.36, p=0.026) independently predicted recurrence, while margin status was not associated. Among patients with recurrence, median PRS was 18.6 months (95% CI, 12.2-24.9). PRS did not differ by recurrence pattern (locoregional vs distant, p=NS) or prior receipt of adjuvant therapy (20.1 vs 13.9 months, p=0.527). Treatment after recurrence was heterogeneous: 41 % received systemic chemotherapy, 26% underwent locoregional or surgical interventions, 6% received alternative/investigational therapies, and 23% received best supportive care. CONCLUSIONS: In the largest single-institution series of homogeneous pCCA resections without vascular reconstruction, recurrence occurred in over 40% of patients, predominantly at distant sites. Lymph node involvement and tumor size independently predicted recurrence, while resection margin did not. Post-recurrence survival remained poor and was unaffected by recurrence pattern or adjuvant therapy. These findings underscore the urgent need for novel systemic and perioperative strategies to reduce recurrence and improve survival in pCCA.

Volume

33

Issue

1

First Page

S283

Last Page

S284

Find It @ Sladen

Share

COinS