Analysis of early tumor shrinkage and depth of response in patients with advanced biliary tract cancer treated with gemcitabine plus cisplatin or gemcitabine plus S-1: An exploratory analysis of JCOG1113.
301 Background: JCOG1113 is a randomized phase III trial to confirm the non-inferiority of gemcitabine (GEM) plus S-1 (GS) compared with GEM plus cisplatin (GC) regarding overall survival (OS) in patients with advanced biliary tract cancer (BTC). Although the non-inferiority of GS to GC was demonstrated, the difference in the nature of tumor shrinkage effects between GC and GS is not clear. Early tumor shrinkage (ETS) and depth of response (DpR) are considered as on-treatment markers that reflect the anti-tumor effect to chemotherapy and have been reported to be associated with survival in metastatic colorectal cancer. However, there are few studies assessing ETS or DpR in advanced BTC. Therefore, we evaluated the association between ETS, DpR, and clinical outcomes in JCOG1113. Methods: We conducted an exploratory analysis of JCOG1113, which included chemotherapy-naïve patients with recurrent or unresectable BTC. ETS was defined as tumor reduction in the sum of the longest diameters of the target lesions at week 6 when compared with that at baseline. DpR was defined as the maximum tumor shrinkage observed until 12 weeks from enrollment. Survival curves were estimated using the Kaplan–Meier method. Progression-free survival (PFS) and OS for ETS and DpR were estimated from week 6 and 12 (landmarks) after enrollment, respectively. Multivariable analyses for PFS and OS, adjusted for baseline factors, were performed using a stratified Cox regression model. Results: Of the 354 registered patients in JCOG1113, 277 patients in the ETS group and 230 patients in the DpR group were included in this study. Seventy-seven patients (27.8%) achieved ETS ≥ 20% (ETS high group) and 52 patients (22.6%) achieved DpR ≥ 40% (DpR high group). The proportion of ETS high group (GC, 25.4%; GS, 30.4%) and DpR high group (GC, 21.2%; GS, 24.1%) was similar between the arms. The patient characteristics of ETS high group were not different between GC and GS. The hazard ratio (HR) of the ETS high group compared with the ETS low group for PFS and OS was 0.76 (95% confidence interval [CI] 0.58–1.00) and 0.80 (95% CI 0.60–1.07), respectively. The impact of ETS was higher in GC (HR 0.64, 95% CI 0.43–0.95) than GS (HR 0.88, 95% CI 0.60–1.28) in PFS. The HR of DpR high group compared with DpR low group for PFS and OS was 0.75 (95% CI 0.55–1.03) and 0.79 (95% CI 0.57–1.09), respectively. The impact of DpR was higher in GC (HR 0.63, 95% CI 0.40–0.998) than GS (HR 0.88, 95% CI 0.57–1.37) in PFS. ETS and DpR were significantly associated with both PFS and OS in the multivariable analyses. Conclusions: ETS and DpR may be useful as on-treatment markers associated with PFS and OS in patients with advanced BTC, especially in those treated with GC. Clinical trial information: UMIN000010667.