P-15 Histology classification highlights the difference in the effectiveness of S-1 over capecitabine when combined with cisplatin in patients with HER2-negative unresectable advanced or recurrent gastric cancer with measurable disease

Abstract

As there is no clear preference between S-1 plus cisplatin (SP) and capecitabine plus cisplatin (XP) as first-line therapy for patients with HER2-negative unresectable advanced or recurrent gastric cancer with measurable disease, we performed an integrated analysis of three phase II randomized trials (HERBIS-2, HERBIS-4A, and XParTS II) with the use of individual participant data (IPD) from each trial. The aim of this trial was to investigate any differences in therapeutic efficacy between SP and XP for this subset, by focusing on the differences in histology. IPD from three randomized phase II trials were collected where patients received either SP [S-1 (40–60 mg twice daily for 21 days) plus cisplatin (60 mg/m2 on day 8), every 5 weeks] or XP [capecitabine (1000 mg/m2 twice daily for 14 days) plus cisplatin (80 mg/m2 on day 1), every 3 weeks]. Overall, SP (n=79) vs. XP (n=83) showed significantly better overall survival (median OS, 14.2 vs. 11.0 months; hazard ratio [HR], 0.704; P = 0.048) and time to treatment failure (median TTF, 4.7 vs. 3.8 months; HR, 0.664; P= 0.011) and a trend toward better progression-free survival (median PFS, 5.9 vs. 5.1 months; HR, 0.717; P = 0.052), whereas no difference in overall response rate (ORR, 47.5% vs. 50.6%). Despite no difference of ORR by histological classification, the differentiated tumors showed significantly better OS but not PFS or TTF of SP against XP, due to a deeper tumor shrinkage of SP over XPas demonstrated by the fact that cases achieved >60% reduction were significantly more common in SP than in XP (28.2% [11/39] vs. 6.7% [2/30]; P = 0.029, Fisher’s exact test). The undifferentiated tumors showed a consistent better trend of OS, PFS and TTF of SP vs. XP, likely as cases without tumor shrinkage tended to be less in SP than XP (2.6% [1/38] vs. 13.0% [6/46]; P=0.121, Fisher’s exact test). Our subgroup analysis further identified that a significant benefit of SP vs. XP was found in OS for differentiated tumors with a tumor reduction of 30% or more [SP, 23.7 months (95%CI, 13.2-NA) vs. XP, 11.7 months (95%CI, 7.8-19.6); HR of 0.339 (95 % CI 0.163–0.705), interaction P = 0.003], strongly suggesting that deeper tumor shrinkage by SP vs. XP contributed most to OS in the differentiated tumors. Our data showed that SP is superior to XP in this setting, but there is a qualitative difference between the effects of SP and XP, which depends on the histological type of the tumor. For the undifferentiated tumors, SP has fewer treatment failures than XP, reflecting better PFS, OS, and TTF in SP than XP. For the differentiated tumors, SP achieves deeper tumor shrinkage than XP, which contributes to longer OS, but not PFS or TTF. Further study is needed to determine whether these differences of S-1 vs.capecitabine are reproduced by the combination of oxaliplatin and immune-checkpoint inhibitors, the new standard of care for HER2-negative unresectable advanced or recurrent gastric cancer.