This study aims to validate a novel surgical strategy for locally advanced Bismuth IIIb or IV perihilar cholangiocarcinoma (PHCC) where the tumor involves the right hepatic artery (RHA), rendering the tumor conventionally unresectable due to the high risk of postoperative right hepatic lobe ischemia if the artery is resected without reconstruction. The intervention involves preoperative selective embolization of the tumor-involved RHA using coils. This procedure is intended to induce the development of collateral arterial supply to the right liver lobe from extrahepatic arteries (e.g., right inferior phrenic, adrenal, intercostal arteries) over a 2-4 week period. Following confirmation of collateral circulation via imaging, patients undergo radical left hepatectomy or left trisectionectomy with en-bloc resection of the involved RHA without arterial reconstruction. The study employs a single-arm design. The primary endpoint is the 1-year overall survival (OS) . The study is designed to detect an improvement to 75%, compared to a historical control rate of 54% observed with chemoimmunotherapy for unresectable disease. Secondary endpoints include surgical conversion rate, R0 resection rate, 1-year/3-year recurrence-free survival (RFS), 3-year OS, and safety profiles (incidence of liver abscess, bile leak, post-hepatectomy liver failure, and other Clavien-Dindo ≥ Grade III complications). The study involves multiple phases: screening/preparation (including biliary drainage and portal vein embolization if needed), the PHAE procedure, radical surgery (2-4 weeks post-PHAE), and a 3-year follow-up period. Statistical analysis will be performed on the intention-to-treat (FAS), per-protocol (PPS), and safety (SS) populations.