FOLFIRI +/- napabucasin in patients with previously treated metastatic colorectal cancer: Overall survival results from the phase 3 CanStem303C study

Abstract

Background Napabucasin is a first-in-class, investigational, orally-administered, reactive oxygen species generator bioactivated by the intracellular antioxidant NAD(P)H:quinone oxidoreductase 1. Phosphorylated signal transducer and activator of transcription 3 (pSTAT3) is a transcription factor hypothesized to predict a treatment effect from napabucasin in metastatic colorectal cancer (mCRC). We present the final efficacy analysis of CanStem303C in patients with previously treated mCRC.

Methods CanStem303C (NCT02753127) was a multi-center, open-label, randomized study of patients aged ≥18 years with histologically confirmed mCRC who failed first-line fluoropyrimidine plus oxaliplatin ± bevacizumab treatment. Patients were randomized 1:1 to FOLFIRI + napabucasin 240 mg oral, twice daily (napabucasin) vs FOLFIRI (control). Bevacizumab could be used in either arm at investigator discretion. Randomization was stratified by geographical region, time to progression from start of first-line therapy, RAS mutation status, bevacizumab as part of protocol treatment, and primary tumor location. Primary endpoints were overall survival (OS) in the general study population (GSP; all randomized patients) and the pSTAT3+ (biomarker-positive [BM+]) population. Key secondary endpoints included progression-free survival (PFS), disease control rate (DCR), and objective response rate (ORR) in the GSP and BM+ population. BM+ status was identified via pSTAT3 immunohistochemistry D3A7 Clinical Trial Assay (Agilent Technologies, Inc.) to detect pSTAT3 protein in cut tissue sections. For the GSP, the study was designed to have 90% power at 1-sided α of 0.025 to detect a 20% reduction in risk of death (hazard ratio [HR] for napabucasin vs control=0.80). For the BM+ population, the study was designed to have ∼88% nominal power at 1-sided α of 0.025 to detect a 30% reduction in risk of death (HR=0.70). OS, summarized using the Kaplan–Meier method, was compared between treatment arms using a stratified (GSP; adjusted for actual stratification) and unstratified (BM+ population) log-rank test.

Results The GSP included 1253 pts (n=624 napabucasin; 629 control); mean (standard deviation) age was 60.1 (11.17) and 59.6 (11.14) years respectively; and 44% of pts were BM+ (n=275 napabucasin; n=272 control). In the GSP, median (m) OS was 14.3 months (napabucasin) vs 13.8 months (control) (HR=0.976, 1-sided p=0.3629). In the BM+ population, mOS was 13.2 months (napabucasin) vs 12.1 months (control) (HR=0.969, 1-sided p=0.3782). PFS, DCR, and ORR were similar between treatment arms in the GSP and BM+ population. Any-grade treatment-emergent adverse events (TEAEs) were reported in 99.5% and 98.7% of patients in the GSP (napabucasin; control), the most common (≥30% of patients) of which were diarrhea (84.6%, 53.9%), nausea (60.5%, 50.5%), vomiting (41.2%, 29.3%), and abdominal pain (41.0%, 25.2%). Grade ≥3 TEAEs were reported in 73.8% and 66.7% of patients in the GSP (napabucasin; control), the most common (≥10% of patients) of which were diarrhea (21.2%, 7.0%), decreased neutrophil count (13.7%, 19.2%), and neutropenia (13.3%, 15.2%). The safety profile in the BM+ population was consistent with that in the GSP.

Conclusions For patients with previously treated mCRC, the addition of napabucasin to FOLFIRI ± bevacizumab did not improve OS, PFS, DCR, or ORR in either the GSP or the BM+ population. TEAEs were consistent with previously published data.

Clinical trial identification Clinicaltrials.gov identifier: NCT02753127.