A Phase Ia/b study of MEK1/2 inhibitor binimetinib with MET inhibitor crizotinib in patients with RAS mutant advanced colorectal cancer (MErCuRIC)

Abstract

Background: Targeting RAS mutant (MT) colorectal cancer (CRC) remains a difficult challenge, mainly due to the pervasiveness of RAS/MEK-mediated feedback loops. Preclinical studies identified MET/STAT3 as an important mediator of resistance to KRAS-MEK1/2 blockade in RASMT CRC. This dose escalation/expansion study assessed safety and initial efficacy of the MEK1/2 inhibitor binimetinib with MET inhibitor crizotinib in RASMT advanced CRC patients.

Methods: In the dose escalation phase, patients with advanced solid tumours received binimetinib with crizotinib, using a rolling- 6 design to determine the maximum tolerable dose (MTD) and safety/tolerability. A subsequent dose expansion in RASMT CRC patients assessed treatment response. Blood samples for pharmacokinetics, MET biomarker and ctDNA analyses, and skin/tumour biopsies for pharmacodynamics, c-MET immunohistochemistry (IHC), MET in situ hybridisation (ISH) and MET DNA-ISH analyses were collected.

Results: Twenty patients were recruited in 3 cohorts in the dose escalation. The MTD was binimetinib 30 mg B.D, days 1-21 every 28 days, with crizotinib 250 mg O.D continuously. Dose-limiting toxicities included grade ≥ 3 transaminitis, creatinine phosphokinase increases and fatigue. Thirty-six RASMT metastatic CRC patients were enrolled in the dose expansion. Pharmacokinetic and pharmacodynamic parameters showed evidence of target engagement. Across the entire study, the most frequent treatment-related adverse events (TR-AE) were rash (80.4%), fatigue (53.4%) and diarrhoea (51.8%) with grade ≥ 3 TR-AE occurring in 44.6%. Best clinical response within the RASMT CRC cohort was stable disease in seven patients (24%). Tumour MET super-expression (IHC H-score > 180 and MET ISH + 3) was observed in 7 patients (24.1%), with MET-amplification only present in 1 of these patients. This patient discontinued treatment early during cycle 1 due to toxicity. Patients with high baseline RASMT allele frequency had a significant shorter median overall survival compared with that seen for patients with low baseline KRASMT allele frequency.

Conclusions: Combination binimetinib/crizotinib showed a poor tolerability with no objective responses observed in RASMT advanced CRC patients. EudraCT-Number: 2014-000463 - 40 (20/06/2014: A Sequential Phase I study of MEK1/2 inhibitors PD- 0325901 or Binimetinib combined with cMET inhibitor Crizotinib in RAS Mutant and RAS Wild Type with aberrant c-MET).

Keywords: Binimetinib; Colorectal cancer; Crizotinib; CtDNA; MET biomarker; Pharmacodynamics; Pharmacokinetics; Phase I; RAS mutant.

Fig. 1

Plasma concentrations for crizotinib, binimetinib and AR00426032 for cohort 7, 12 and 13 dose escalation and dose expansion phases. Left: Twenty-four hours PK profiles for crizotinib, obtained at C1D21 (A), for binimetinib obtained at C1D21 (B) and for AR00426032 (C) obtained at C1D21. Star (left) and dotted line (right) indicates that PK samples for dose expansion were analysed at a separate time

Fig. 2

Modulation of pERK1/2^T202/Y204, pMEK1/2^S217/221, pMET^Y1003 and pSTAT3^Y705 expression levels in paired skin and/or tumour biopsies. A-C. Left: pERK1/2-ERK, pMEK1/2-MEK1/2 levels in paired skin biopsies dose expansion Cohort 7: binimetinib 30 mg B.D 1 - 28 d and crizotinib 200 mg B.D continuously (A), Cohort 12: binimetinib 30 mg B.D 1 - 21 d and crizotinib 200 mg B.D continuously (B), Cohort 13: binimetinib 30 mg BD 1 - 21 d and crizotinib 250 mg O.D continuously (C). A-C Right: Densitometry was performed on the WB images using ImageJ software. SC = screening. C1D15: Skin biopsy obtained between 3–6 h following morning dose of binimetinib. D. Left: pERK1/2^T202/Y204 and pMEK1/2^S217/221 levels in paired skin biopsies dose expansion. Right: Densitometry was performed on the WB images shown in A using ImageJ software. SC = screening. C1D15: Skin biopsy obtained between 3–6 h following morning dose of binimetinib. E pcMET^Y1003, pSTAT3^Y705, pERK1/2^T202/Y204 expression and phosphorylation in paired tumour biopsies. Densitometry was performed on the WB images for pcMET^Y1003, pSTAT3^Y705, and pERK1/2^T202/Y204 using ImageJ software. SC = screening. C1D15: Tumour biopsy obtained between 3–6 h following morning dose of binimetinib and crizotinib

Fig. 3

Tumour response, progression-free survival, and overall survival for combined binimetinib with crizotinib. A. Best radiological response observed to treatment as per cohort in dose escalation phase and phase Ib RASMT cohort. B. Kaplan–Meier curves for median progression-free survival (PFS) and median overall survival (OS) in the phase Ib study. Left: Median PFS for all patients is 1.81 months (95% CI 1.51–2.04 months). The survivor function is 0.06 at 6 months (95% CI 0.01–0.17). Right: Median OS for all patients is 5.62 months (95% CI 2.97–7.40 months). The survivor function is 0.44 at 6 months (95% CI 0.27–0.60). C. Kaplan–Meier curves for median overall survival in patients from the dose expansion phase with low (tertile 1), median (tertile 2) and high (tertile 3) baseline RAS mutant allele frequency in circulating tDNA. 1 st tertile: 1.6–10.5; 2nd tertile: 13.57–32.15 and 3rd tertile: 34–17–86.33