A final analysis of primary endpoints from the phase III RECOURSE study in patients with metastatic colorectal cancer (mCRC) refractory to standard therapies showed a statistically significant overall survival (OS) and progression-free survival (PFS) benefit with TAS-102 in all prospectively defined subgroup analyses, including prior therapy with regorafenib. The results were presented during the Proffered Paper session in Colorectal Cancer at the ESMO Congress 2014, by Prof. Eric Van Cutsem of the Digestive Oncology Unit, University Hospital Leuven, Leuven, Belgium.
The data from the RECOURSE study was first reported at the ESMO World Congress in Gastrointestinal Cancer earlier this year in Barcelona, Spain. At the ESMO Congress in Madrid, Prof. Van Cutsem presented new data, in particular subgroup analyses of OS and PFS by KRAS status, stratification factors and patient characteristics, and time to worsening of ECOG performance status (PS) to 2 or more.
TAS-102 is a combination of a novel oral nucleoside, trifluridine (FTD) with the thymidine phosphorylase inhibitor, tipiracil hydrochloride (TPI), which prevents the degradation of FTD, enabling sustained and effective FTD levels.
TAS-102 vs placebo in refractory mCRC
RECOURSE was conducted to evaluate the efficacy and safety of TAS-102 in patients with mCRC refractory to standard therapies. Patients with ECOG PS 0-1 who failed two or more prior treatments with fluoropyrimidines, irinotecan, oxaliplatin, bevacizumab and cetuximab/panitumumab (in case of KRAS wild-type disease) were eligible for the study.
Treatment continuation was foreseen until progression, intolerant toxicity or patient refusal. It was a multicenter, randomised, double-blind, placebo-controlled, phase III study with stratification by KRAS status, time from diagnosis to metastatic disease, and geographical region. It was performed in 13 countries around the world. Enrollment started in June 2012 and finished in October 2013.
The study primary endpoint was OS and the key secondary efficacy endpoint was PFS. Other secondary endpoints included safety, tolerability, time to treatment failure (TTF), overall response rate (ORR), disease control rate (DCR), duration of response (DoR), OS and PFS in subgroups determined by KRAS status.
The OS and PFS were evaluated by using univariate and multivariate analyses for prospectively defined subgroups and a retrospectively defined subgroup of patients with prior treatment with regorafenib.
In total 800 patients were randomised to TAS-102 (534 patients) or placebo (266 patients). In intent to treat population, patient demographics and characteristics were balanced between two arms with one third of Asian patients in each group.
The hazard ratios for OS and PFS were 0.68 (p < 0.0001) and 0.48 (p < 0.0001), respectively, both favouring TAS-102. The OS in TAS-102 group was 7.1 month vs 5.3 month in placebo group. Median PFS was 2.0 months in TAS-102 treated patients and 1.7 month in the placebo arm.
The OS and PFS benefit with TAS-102 was consistent across all subgroups. In particular, the HRs for OS in subgroups were 0.58 in patients with KRAS wild-type tumours and 0.80 in KRAS mutated tumours; 0.64 in Western population and 0.75 in Asian patients; 0.73 in patients with PS 0 and 0.61 in patients with PS 1; and 0.69 for patients who have already received or not regorafenib.
In patients with time from diagnosis to first metastasis shorter than 18 months, the HR for OS was 0.84 and 0.64 in those with ≥ 18 months. In patients younger than 65, this HR was 0.74 and 0.62 in the group ≥ 65 years old. In patients who received three prior treatments, the HR was 0.74 and 0.59 in those who received ≥ 4 treatment lines.
Overall survival treatment effect remained similar in the multivariate model
The OS treatment effect remained the same in the multivariate model. No predictive factors were identified. Statistically significant prognostic factors (p < 0.05) in final model based on stepwise selection were:
- time since diagnosis of first metastasis,
- ECOG performance status,
- and number of metastatic sites.
Time to worsening of ECOG PS status to 2 or more was significantly delayed with TAS-102 vs placebo with medians of 5.7 vs 4.0 months (HR = 0.66, p < 0.0001). Post-study treatment was similar between arms (41.2% in TAS-102, 42.5% in placebo).
Safety results were previously presented at the ESMO World Congress on Gastrointestinal Cancer 2014. The most frequently observed toxicities were gastrointestinal and haematologic. Serious adverse events were observed in 29.6% patients in TAS-102 and 33.6% patients in placebo group. Primary reason for treatment discontinuation due to adverse events was 3.6% in TAS-102 and 1.5% in the placebo group. One treatment-related death was observed in TAS-102 group. The rate of febrile neutropenia was 3.8% and frequency of G-CSF usage 9.4% in TAS-102 and 0% in placebo group.
Conclusion
Prof. Van Cutsem concluded that TAS-102 demonstrated a clinically relevant improvement in OS and PFS compared with placebo in patients with mCRC refractory/intolerant to standard therapies. Improved OS benefit was statistically significant or trended favorably for TAS-102 across all stratification factors and predefined subgroups. Consistent with OS results, PFS improvement for TAS-102 was statistically significant across all stratification factors and predefined subgroups. The OS benefit was maintained irrespective of regorafenib use.
Discussant Dr. Christophe Tournigand said that in the RECOURSE study TAS-102 significantly improved OS and PFS in patients with metastatic colorectal cancer, refractory or intolerant to standard therapies. However, questions for the future would be: identifying biomarkers, addressing the question of QoL improvement, and seeing efficacy/tolerance in combination therapy, and efficacy in earlier lines of therapy.
The sponsor of the RECOURSE trial is Taiho Oncology Inc./Taiho Pharmaceutical Co. Ltd.
Note
The study is subsequently published in The New England Journal of Medicine.