The influence of renal function on gemcitabine-based chemotherapy for advanced biliary tract cancer: An exploratory subgroup analysis of JCOG1113

368 Background: JCOG1113 is a randomized phase III trial to evaluate gemcitabine (GEM) plus S-1 (GS) versus GEM plus cisplatin (GC) regarding overall survival (OS) for advanced biliary tract cancer (BTC) (UMIN000001685) and the non-inferiority of GS was demonstrated. It is necessary to consider renal function using cisplatin or S-1 because cisplatin has renal toxicity, and the toxicity of S-1 is affected by renal function. Therefore, we evaluated the influence of renal function on the efficacy and safety of GC and GS in JCOG1113. Methods: All enrolled patients (pts) in JCOG1113 (n = 354) were analyzed. Eligibility criteria included chemotherapy-naïve pts with recurrent or unresectable biliary tract adenocarcinoma, ECOG-PS of 0–1, CCr > 50 ml/min, and adequate organ function. Renal function was classified into two groups by creatinine clearance (CCr) as calculated by the Cockcroft-Gault formula; high CCr (CCr ≥ 80 ml/min) or low CCr (80 > CCr ≥ 50 ml/min). The impact of renal function on OS and progression-free survival (PFS) were compared using the Cox regression model. The adverse events (AEs) were compared using Fisher’s exact test. Results: Eighty-eight pts on GC and 88 pts on GS were included in the high CCr group, and 87 pts on GC and 91 pts on GS were included in the low CCr group. There were no differences between the groups regarding, sex, PS, primary site, biliary drainage, operation, or recurrence, except for age. The hazard ratio (HR) of GS to GC for OS was 1.12 (95% CI 0.81–1.56) in the high CCr group and 0.80 (95% CI 0.58–1.11) in the low CCr group. The HR of GS to GC for PFS was 1.06 (95% CI 0.78–1.44) in the high CCr group and 0.69 (95% CI 0.50–0.94) in the low CCr group. Grade 3-4 AEs of white blood cell count decreased (35.3%/23.6%), anemia (29.4%/7.9%) and platelet count decreased (18.8%/10.1%) were more common in GC than GS in the low CCr group. In contrast, the incidence of all grade 3-4 non-hematological AEs was higher (36.0%/11.8%) in GS than GC in the low CCr group ( p = 0.0002). Conclusions: GS was better in terms of OS, PFS, and hematological toxicities than GC in the low CCr group. GS might be recommended for the population with lower renal function in the treatment for advanced BTC.