FOCUS4-D: Results of a randomised, placebo controlled trial of AZD8931 (HER1, 2, 3 inhibitor) in advanced/metastatic colorectal cancer tumours that are wildtype for BRAF, PIK3CA, KRAS & NRAS


Year:

Session type:

Theme:

Richard Adams1,Ewan Brown2,Louise Brown3,Rachel Butler4,Stephen Falk5,David Fisher3,Richard Kaplan3,Phillip Quirke6,Susan Richman6,Leslie Samuel7,Jenny Seligman6,Matthew Seymour6,Kaikeen Shiu8,Harpreet Wasan9,Richard Wilson10,Timothy Maughan11,On behalf of The FOCUS4 Trial Investigators12
1Cardiff University and Velindre Cancer Centre,2Edinburgh Cancer Centre,3MRC Clinical Trials Unit at UCL,4All Wales Genetics Laboratory, University of Wales,5UH Bristol NHS Foundation Trust,6University of Leeds,7Aberdeen Royal Infirmary,8University College London Hospitals Trust,9Imperial College London,10Queen's University Belfast,11University of Oxford,12Collaboration

Abstract

Background

FOCUS4 is a phase II/III trial programme testing targeted agents in patients with advanced colorectal cancer (aCRC) in molecularly stratified cohorts. Her 1,2,3 signal inhibition has been hypothesized as a mechanism for tumour control in aCRC potentially sensitive to EGFR inhibition which might delay emergence of resistance in patients responding to standard therapy.

Method

The FOCUS4-D molecular cohort includes patients whose tumour biomarker assessment confirms wildtype status for BRAF, PIK3CA, KRAS & NRAS. Following 16 weeks of first-line therapy, patients with stable or responding tumours were randomised to receive either oral AZD8931 or matching placebo. Patients were followed for progression-free-survival (PFS) as the primary outcome according to RECIST v1.1 criteria. Secondary outcomes included toxicity. Pre-planned interim analyses were agreed using Multi-Arm Multi-Stage (MAMS) trial design methodology triggered by occurrence of PFS events in the placebo arm. Cox regression was used to estimate PFS hazard ratios of AZD8931 relative to placebo adjusted for minimisation factors.

Results

Across 18 UK sites, 32 patients were randomised (16 into each arm) between February 2014 and March 2016. Patients were balanced for baseline characteristics. At the first pre-planned interim analysis, the Independent Data Monitoring Committee recommended closure of FOCUS4-D on grounds of lack of activity. At this analysis, 26 patients had experienced a PFS event (11 in the placebo arm). Analyses by both Intention-to-treat (HR=1.31 [95% CI 0.56-3.08], p=0.54) and per-protocol (HR=1.42 [95% CI 0.58-3.50], p=0.49) did not indicate any benefit for PFS. Toxicity was minimal with the most frequent relating to skin rash; graded at ≥3 in 2 patients (13%).

Conclusion

Despite minimal toxicity, there was no evidence to suggest any efficacy of single agent Her 1,2,3 inhibition in aCRC patients whose tumour is wildtype for BRAF, PIK3CA, KRAS and NRAS after induction therapy in the first-line setting.