An observational study of best supportive care with or without oral capecitabine in patients with metastatic gallbladder cancer.

Abstract

521 Background: Patients with metastatic (M+) gallbladder cancer (GBC) have a median survival between 3 - 6 months and therefore palliation of symptoms without compromising quality of life (QoL) is the principal goal of treatment. This study aimed to compare overall survival (OS), toxicity and QoL in patients receiving best supportive care (BSC) with or without oral capecitabine (X). Methods: Patients with M+ GBC and a KPS of ≥70 were prospectively assigned to BSC + X - arm A or BSC alone - arm B, based on physician discretion. Arm A received BSC + X @1650 mg/m2 d1-14, repeated every 21 days for a maximum of 6 cycles. BSC measures included pain palliation; ascitic fluid drainage; jaundice mitigation and other palliative measures. Overall survival (OS) was computed by the Kaplan-Meier method from the date of histologic confirmation and compared using log-rank test. Qol was measured using FACT-Hep v4 administered at baseline, 3 and 6 months. Mean differences in scores were compared using Mann Whitney / Independent t-test and the linear mixed effects model was used for longitudinal analysis to determine the impact of treatment and time on QoL. Results: Between December 2020 to April 2022, 64 patients were assigned, 32 to each arm; data analysed August 2022. For the arms A vs B, the mean age (years) ± SD was 53.3±10.3 vs 54.5±10.3, p=0.21; females, 72% vs 73%, p=0.21; KPS 70, 9% vs 44%, p=0.008; pre-treatment biliary drainage, 9% vs 31%, p=0.06. In arm A, 78% and 21% completed 3 and 6 cycles of X. Median survival was 3.6 vs 2.0 mo, p=0.001; 26 and 27 patients succumbed (~83%), and at 6 months the survival probability was 28% vs 0% respectively; vomiting grade 1-2, 50% vs 78%, p=0.02; diarrhoea grade 1-2, 59% vs 9%, p=0.00; hand-foot syndrome – only arm A, grade 1, 2 & 3 were 32%, 6.5% & 3.2% p=0.001. Baseline QoL mean scores (Table) were significantly worse for most domains in arm B. However, there was no difference in the longitudinal QoL scores between the two treatment arms – FACTG total score (arm*time p=0.759). Conclusions: Not withstanding better prognosis patients of M+ GBC receiving X with BSC compared to BSC alone, a 2 month difference in median survival in favour of arm A was associated with worse diarrhoea and hand-foot syndrome and no discernable impact on longitudinal QoL scores between the two treatment arms.[Table: see text]