Does Surgical Approach Influence Benchmark Attainment in Intrahepatic Cholangiocarcinoma?

Delphine Gastineau, Claire Goumard, Alexandre Doussot, Marc‐Antoine Allard, Raphael Venezia, Serena Langella, G. Millet, Ruth Bustamante, Christian Hobeika, Rami Rhaiem, Stylianos Tzedakis, Heithem Jeddou, Bastien Le Floch, Alain Valverde, N. Péru, P Pessaux, Fabio Giannone, Mabrut Jp, Kayvan Mohkam, Antonio Sa Cunha, Chady Salloum, P. Blanc, B Le Roy, Takayuki Kawai, Olivier Farges, S. Hofmeyr, David Fuks, S. Truant, Daniel Cherqui, Alessandro Ferrero, Emmanuel Boleslawski, Jean‐Marc Regimbeau, Eric Vibert, Danièle Sommacale, Olivier Soubrane, Olivier Scatton, Alexis Laurent, Raffaele Brustia, AFC‐ICC‐2009, AFC‐LLR‐2018, PRS‐2019 and FRIES‐ACHBPT‐2024 Study group

BACKGROUND: Intrahepatic cholangiocarcinoma (ICC) is a rare malignancy with rising incidence and poor prognosis. Surgical resection remains the only curative option, with outcomes driven by R0 margins and lymphadenectomy (LND). While open liver resection (OLR) is standard, laparoscopic (LLR) and robotic liver resection (RLR) are increasingly used, with concerns about their oncologic adequacy. OBJECTIVE: To assess the association between composite surgical benchmark attainment (R0 + LND) and short- and long-term outcomes after ICC resection, stratified per surgical approach. METHODS: A retrospective multicenter analysis was conducted including 240 patients treated from 2010 to 2024. The primary endpoint was achievement of both R0 resection and LND. Secondary endpoints included 90-day mortality, overall (OS), and disease-free survival (DFS). RESULTS: Among the 240 included patients 38.8% met surgical benchmarks (R0 + LND), with rates varying by approach: OLR 51.1%, RLR 44.4%, and LLR 14.5% (p < 0.001). 90-day mortality was 5.0% (12/240) overall, with no significant difference between surgical approaches (p = 0.219). One and 3-year overall survival was 79%/30% for OLR, 95%/89% for LLR, and 90%/81% for RLR (p < 0.001). Multivariable analysis identified tumor multiplicity as an independent predictor of mortality (HR = 2.18, p = 0.003), while LLR was independently protective (HR = 0.15, p < 0.001); RLR showed a non-significant protective trend. CONCLUSION: Although OLR and RLR more frequently met oncological benchmark criteria, only LLR was independently associated with improved overall survival, highlighting a potential disconnect between benchmark adherence and long-term outcomes in ICC.