P-258 - HGCSG1503: A retrospective cohort study evaluating the safety and efficacy of TAS-102 in patients with metastatic colorectal cancer: Analysis of cases of prior regorafenib

Abstract

Introduction: The J003 trial and RECOURSE trial revealed the safety and efficacy of TAS-102 for patients with metastatic colorectal cancer (mCRC). In March 2014, TAS-102 was approved in Japan. However, in these pivotal trials, there were few cases in which regorafenib was administered as prior treatments, and also there were few reports on the effectiveness and safety of TAS-102 after administration of regorafenib. Methods: We retrospectively analyzed the clinical data of 411 patients who received TAS-102 in the multi-institutional retrospective study (HGCSG1503). This study was analyzed by CTCAE v4.0 for adverse events (AEs), RECIST v1.1 for response rate (RR)/disease control rate (DCR). To compare patients who received regorafenib before TAS-102 (Prior REG) and those did not receive regorafenib (No prior REG), Fisher’s exact test was used in terms of patient characteristics, AE, RR/DCR, and Log-rank test was used in terms of TTF, PFS and OS. Results: No prior REG and Prior REG were 285 and 126, respectively. The patient’ characteristics between No prior REG and Prior REG were generally balanced except for lung metastasis (56.8% in No prior REG, 67.5% in Prior REG; p = 0.049), prior oxaliplatin administration (95.1% in No prior REG, 99.2% in Prior REG; p = 0.045), prior irinotecan administration (90.5% in No prior REG, 100% in Prior REG; p < 0.001) and patients over 18 months since diagnosis of metastasis (63.2% in No prior REG, 84.1% in Prior REG; p < 0.001). The AEs between No prior REG and Prior REG were also generally balanced. RR/DCR were 0.8/37.5% in No prior REG and 0/36.5% in Prior REG (p = 1.000/0.908). Median PFS was 2.3 months in No prior REG and 2.1 months in Prior REG (HR 1.157, p = 0.185). Median OS was 8.1 months in No prior REG and 5.7 months in Prior REG (HR 1.355, p = 0.007). Conclusion: In this analysis, No prior REG population contained lung metastasis, and short interval from diagnosis of metastasis. The adverse events, detail of administration, RR/DCR, and PFS were no significant difference regardless of the administration history of REG. The OS was significantly extended with No prior REG population. This is presumed because 30% of No prior REG population received REG for post TAS-102 therapy.