Adjuvant S-1 versus observation in curatively resected biliary tract cancer: A phase III trial (JCOG1202: ASCOT)

382 Background: Capecitabine is usually used for patients with curatively resected biliary tract cancer (BTC) in EU and US, but no clear survival benefit has been shown in phase III trials. S-1, an oral fluoropyrimidine derivative, has shown promising efficacy, with a mild toxicity profile, in patients with advanced BTC. The aim of this trial was to confirm whether adjuvant S-1 therapy might improve the overall survival (OS) in patients with curatively resected BTC. Methods: This open-label, multicenter, randomized phase III trial was conducted in 38 Japanese hospitals. Eligible patients were aged 20 to 80 years old, had undergone R0/R1 resection for histologically confirmed adeno(squamous) carcinoma of the extrahepatic bile duct, gallbladder or ampulla of Vater (T2-4, N0, M0 or T1-4, N1, M0) or the intrahepatic bile duct (T1-4, N0-1, M0) (7th UICC classification), and had an ECOG performance status (PS) of 0 or 1.The calculated sample size was 440 to detect hazard ratio for OS of 0.74 with one-sided alpha of 5% and a power of 80%. Patients in surgery-alone arm received no further anti-cancer treatment, while those in adjuvant S-1 arm received 4 cycles of oral S-1 chemotherapy at the dose of 40 mg/m2 twice daily for 4 weeks, followed by 2 weeks of rest. Primary endpoint was OS, and secondary endpoints were relapse-free survival (RFS), incidence of adverse events, and proportion of treatment completion. Results: A total of 440 patients (surgery-alone, n = 222; adjuvant S-1, n = 218) were enrolled from September 2013 to June 2018. The data cutoff date was June 23, 2021, and the median follow-up duration was 45.4 months. Of all randomized patients, OS was significantly longer with adjuvant S-1 than surgery-alone (hazard ratio [HR] 0.694, 95%CI, 0.514-0.935; one-sided p = 0.008; the 3-year OS, 67.6% [surgery-alone; 95%CI, 61.0-73.3%] and 77.1% [adjuvant S-1; 95%CI, 70.9-82.1%]). Adjuvant S-1 was also better for RFS (HR 0.797 [95%CI, 0.613-1.035], 3-year RFS, 50.9% [surgery-alone; 95%CI, 44.1-57.2%] and 62.4% [adjuvant S-1; 95%CI, 55.6-68.4%]). All preplanned subgroup analyses (PS, age, cancer type, cancer stage, R factor, and serum CA19-9) revealed favorable OS and RFS for adjuvant S-1 arm. The main grade 3-4 adverse events in adjuvant S-1 arm were biliary tract infection (7.2%), diarrhea (2.9%), appetite loss (2.9%), fatigue (2.9%), and the treatment was well-tolerated. Conclusions: Adjuvant S-1 therapy led to significantly longer survival than surgery alone and becomes the standard of care for resected BTC. Clinical trial information: UMIN000011688.