AbstractINTRODUCTION Prior data suggest that patients with intrahepatic cholangiocarcinoma (iCCA) have a better prognosis compared to other biliary tract cancers (BTC). However, granularity on outcomes for advanced extrahepatic cholangiocarcinoma (eCCA) (including distal- (dCCA) and perihilar cholangiocarcinoma (pCCA)) and gallbladder cancer (GBC) is lacking. With targeted therapies being developed for these subgroups, benchmark data is of value to understand the natural history and inform future trial design. METHODS The aim of this post-hoc analysis was to provide reference survival data for patients with advanced eCCA treated with first-line cisplatin-gemcitabine (CisGem) chemotherapy within the prospective, randomised Advanced Biliary tract Cancer (ABC)-01, -02 and -03 studies. Individual level data from patients with eCCA and GBC recruited to these studies were retrieved. Survival analysis was performed using Kaplan Meier and univariate and multivariable Cox Regression. Logistic regression was utilised whenever required. RESULTS Of 534 patients recruited into the ABC-01, -02 and -03 studies, eligible patients for this analysis included a total of 117 eCCA treated with CisGem (including 68 with pCCA and 49 with dCCA) and 112 with GBC. pCCA had less prior surgery (4.08%) and had the highest rate of biliary stenting (26.53%). dCCA were predominantly metastatic at study entry (77.94%), while pCCA had a rate of 40.82% and 59.18% for locally advanced and metastatic disease at study entry, respectively. Majority of patients in the GBC cohort were female (62.50%) and predominantly metastatic (81.25%). Estimated median overall survival (OS) for dCCA was 14.25 months (95% CI 8.80-17.44) and 12.18 months (8.31-16.12) for pCCA. Estimated median progression-free survival (PFS) was 8.28 months (95% CI 6.31-9.39) for dCCA and 8.37 months (95% CI 6.53-10.61) for pCCA. PFS and OS for GBC were 6.86 months (95% CI 5.75-8.51) and 10.84 months (95% CI 8.97-12.41), respectively. Objective radiological response (ORR) was 23.93% for eCCA: (dCCA (27.94%), pCCA (18.37%)) and 26.79% for GBC cohort. Prognostic factors of interest for OS included performance status and CA125 for eCCA, and grade of differentiation for GBC. For OS in iCCA, dCCA, pCCA and GBC, multivariable analysis confirmed patients with GBC had the shortest OS (HR 2.01 (95% CI 1.18-3.43); p-value 0.011 (vs iCCA)) when adjusted for other prognostic factors (performance status, stage and tumour markers). CONCLUSIONS Within the BTC subgroups, GBC had the worse survival. Performance status and baseline tumour markers (CA125 and CEA) are strongly associated with worse survival. This subgroup data provides a reference for future studies incorporating immunotherapy and other new treatment strategies. Clinical trial registration number ABC-01/ABC-02 (NCT00262769), ABC-03 (NCT00939848) Impact and implications The data presented in this manuscript provide the benchmark outcome data for subgroups of ECCA and GBC treated with CisGem for future clinical trial design. Despite the absence of immunotherapy being involved, the lack of alternative comprehensive series highlights the value of this series. Future trials could involve novel targeted therapies or immunotherapy.
Lamarca et al. (Sun,) studied this question.