Randomized phase 3 study of S-1 versus capecitabine in the first-line treatment of metastatic colorectal cancer (mCRC): The SALTO study of the Dutch Colorectal Cancer Group.

Abstract

3640 Background: Hand-foot syndrome (HFS) is a common side effect of capecitabine. S-1 is an oral fluoropyrimidine with comparable efficacy to capecitabine in gastrointestinal cancers but associated with a lower incidence of HFS in Asian patients (pts). The Dutch SALTO study (NCT01918852) was designed to compare the incidence of HFS between S-1 and capecitabine as 1stline treatment in mCRC pts. Methods: Previously untreated mCRC pts, WHO performance status (PS) 0-2, and planned treatment with fluoropyrimidine monochemotherapy were randomized between capecitabine 1250 mg/m2 for pts < 70 years of age, and 1000 mg/m2 for pts ≥ 70 years of age, b.i.d. day 1-14 (arm A) and S-1 30 mg/m2 b.i.d.day 1-14 (arm B). Cycles were administered q 3 wks. Cotreatment with bevacizumab 7.5 mg/kg/3wks iv was optional. Pts were stratified for use of bevacizumab, PS, serum LDH, and institution.Toxicity was assessed prior to each cycle by the investigator using NCI CTC 4.0 and by patients using a diary. Tumor response was assessed every 9 wks. Primary endpoint was the incidence of HFS. A total of 150 pts was required to demonstrate a difference in HFS of ≥ 20% with 90% power (α = 0.05, 2-tailed test). Secondary endpoints were grade 3 HFS, other toxicities, progression-free survival (PFS), response rate (RR), overall survival (OS). Results: A total of 161 pts were randomized between January 2014 - July 2015, 81 in arm A and 80 in arm B. Median age is 73 yrs (50-86), 10% of pts have PS2, 58% were planned to receive bevacizumab. Median follow-up is 9.4 months, 39 pts (24%) are still on study. Pts received a median of 6 cycles. In arm A and B the investigator-assessed all grade HFS is 70% and 41% (p .00025), and grade 3 HFS 20% and 4% (p .0025), resp. Other grade ≥ 3 toxicity only differed for anorexia (2.5% and 12.5%, p .018). All grade HFS assessed by 91 pts (preliminary results) is 82% and 57% (p .014), and grade 3 16% and 4% (p .08). Median PFS is 8.3 months and not significantly different between treatment arms (p .88, HR 0.97 [0.67-1.41]). Data on OS are not mature. Conclusions: Treatment with S-1 results in a significantly lower incidence of HFS compared to capecitabine in pts with mCRC, without compromising efficacy. Clinical trial information: NCT01918852.