Clinical utility of the Idylla CDx MSI test in identifying MSI-H mCRC patients eligible for nivolumab monotherapy or in combination with ipilimumab therapy (CheckMate 8HW phase III trial).

Abstract

28 Background: CheckMate 8HW (CA209-8HW), a Phase 3, randomized, 3-arm open-label study, evaluated nivolumab monotherapy (nivo: Arm A), nivolumab plus ipilimumab combination therapy (nivo+ipi: Arm B) or investigator’s choice chemotherapy (chemo: Arm C) for the treatment of subjects with deficient mismatch repair/microsatellite instability- high (dMMR/MSI-H) metastatic colorectal cancer (mCRC). The Idylla companion diagnostic (CDx) MSI Test was used as an investigational method, next to an IHC MMR method, to centrally and retrospectively confirm MSI-H status in trial participants enrolled by local MSI-H/dMMR assays. Methods: Clinical utility of the Idylla CDx MSI Test (Idylla) in the intended use population was assessed by evaluating: (1) progression-free survival (PFS) and the corresponding hazard ratio (HR) of nivo+ipi vs. chemotherapy in the 1st line (1L) therapy of mCRC, (2) PFS HR of nivo+ipi vs. nivo in all lines, and (3) objective response rates (ORR) of nivo+ipi and nivo in all lines. Enrollment of patients in the trial was based on local dMMR/MSI-H status per at least one of the three methods (IHC, PCR, NGS), referred as the Clinical Trial Assay positive (CTA+) population. The Idylla test was used at the central laboratory to retrospectively evaluate the patients’ MSI status by testing the submitted tumor specimens. A bridging study between the efficacies of the trial randomized population (CTA+) and the intended use population of the Idylla CDx MSI Test was conducted to demonstrate the clinical utility of the test. Results: Of the 837 CTA+ subjects randomized to all arms, 725 subjects had valid Idylla results by central testing. From the remaining 112 subjects, 7 were tested by the Idylla but resulted in an invalid result and 105 were unevaluable by Idylla as they did not meet the test or other quality requirements. In all CTA+ subjects in 1L, the PFS HR for nivo+ipi (n=200) vs chemo (n=101) was 0.32 (95% CI of 0.22 - 0.45). In comparison, in the CTA+ subjects centrally determined as MSI-H by Idylla, the PFS HR for nivo+ipi (n=147) vs chemo (n=71), in 1L was 0.20 (95% CI: 0.12–0.31). In addition, in CTA+ subjects centrally determined to be MSS by Idylla, the PFS HR of nivo+ipi (n=30) vs chemo (n=15) in 1L was 1.51 (95%CI: 0.68-3.33). In the nivo+ipi vs nivo comparison in all lines of therapy, the PFS HR for centrally determined Idylla MSI-H group (n=504) was 0.60 (95% CI: 0.45–0.80 ) compared to PFS HR of 0.63 (95% CI: 0.51–0.79) in CTA+ (n=705). The ORRs for nivo monotherapy and for nivo+ipi in centrally determined Idylla MSI-H group were 58.7% (95% CI, 52.3, 64.9) and 73.5% (95% CI: 67.7, 78.8), respectively, in all lines of therapy. Conclusions: The Idylla CDx MSI Test can effectively identify MSI-H mCRC patients who may benefit from treatment with nivolumab as a monotherapy and/or treatment with nivolumab in combination with ipilimumab.