Conversion therapy of tislelizumab plus lenvatinib and GEMOX in unresectable locally advanced biliary tract cancer (ZSAB-TransGOLP): a multicentre, prospective, phase 2 study
Mené sur 41 patients atteints d'un cancer des voies biliaires de stade localement avancé et non résécable (âge médian : 58 ans), cet essai multicentrique de phase II évalue l'efficacité, du point de vue du taux de résection R0, et la toxicité d'un traitement néoadjuvant combinant tislélizumab, lenvatinib et chimiothérapie de type GEMOX
Background: The optimal conversion regimen that allows more patients with unresectable biliary tract cancer to access surgery remains unclear; there is currently no standard conversion therapy for biliary tract cancer in China, with commonly used regimens including immunotherapy-based combinations and local therapy. The ZSAB-TransGOLP study aimed to assess the efficacy and safety of tislelizumab plus lenvatinib and GEMOX (gemcitabine plus oxaliplatin) chemotherapy (GOLP) in patients with this disease.
Methods: This single-arm, phase 2 study was conducted at two centres in China. Eligible patients aged 18–70 years with previously untreated locally advanced unresectable biliary tract cancer (intrahepatic cholangiocarcinoma, perihilar bile duct cancer, and gallbladder cancer), and an Eastern Cooperative Oncology Group performance status of 0 or 1, Child–Pugh score of A, and at least 3 months' life expectancy were enrolled. Patients received 200 mg intravenous tislelizumab on day 1 and intravenous GEMOX (0·5 h of 1000 mg/m2 gemcitabine on days 1 and 8; and 2 h of 85 mg/m2 oxaliplatin on day 1) in a 21-day cycle for three cycles, and 8 mg oral lenvatinib once daily. Tumour resectability was determined by the multidisciplinary team every 3 cycles of conversion therapy; patients who were ineligible for R0 resection and did not require surgery after six cycles received maintenance therapy with tislelizumab plus lenvatinib at the same dose as used in conversion therapy until completing 1 year of treatment, disease progression, intolerable toxicity, death, consent withdrawal, or investigators' decisions. The primary endpoint was the R0 resection rate. All treated patients were evaluable for safety and primary endpoint. The trial is registered with ClinicalTrials.gov (NCT05156788) and is ongoing but closed for recruitment.
Findings: Between Dec 27, 2021 and July 3, 2023, 52 patients were screened, 11 were excluded for ineligiblity, and 41 patients were enrolled and received the GOLP regimen. All patients were Chinese, with median age of 58 years (IQR 54–65); 21 patients (51%) were male and 20 patients (49%) were female. Median duration of GOLP treatment was 3 cycles (IQR 3–6). 28 (68%) of 41 patients underwent surgery. At a median follow-up of 19·5 months (IQR 14·6–25·0) by data cutoff on Jan 20, 2025, the R0 resection rate was 63% (26 of 41 [95% CI 47–78]). All patients had at least one any-grade treatment-related adverse event (TRAE); grade 3–4 TRAEs occurred in 20 (49%) of 41 patients, with neutropenia (14 [34%] of 41) being most common. Serious TRAEs occurred in 4 (10%) of patients and included neutropenia (three [7%]) and decreased platelet count (one [2%]). No TRAE-related deaths occurred.
Interpretation: With promising efficacy and manageable safety, GOLP represents a potentially feasible and high-efficiency conversion regimen for unresectable locally advanced biliary tract cancer.
The Lancet Oncology , résumé, 2025