In RAS wild type metastatic colorectal cancer patients (pts), high tumor expression of microRNA miR-31-3p and miR-31-5p has been associated with poorer benefit of anti-EGFR therapy. miR-31-3p expression has been shown to be predictive of treatment effect in the FIRE-3 trial. The PETACC-8 phase III trial assessed the efficacy of cetux addition to FOLFOX compared to FOLFOX only in pts with resected stage III colon cancer (CC). The primary end point was negative, but a trend towards benefit of cetux on disease free survival (DFS) was observed in pts with RAS/BRAF wild type (WT) tumors. The current study aimed at assessing miR-31-3p and 5p tumor levels as prognostic and predictive biomarkers of adjuvant cetux benefit in these pts.
miR-31-3p and 5p levels were measured by RT-qPCR from 477 WT pts tumor RNA. The primary objective was to demonstrate a benefit of cetux on DFS for low miR-31-3p expressers. Cox regression model was used for univariate and multivariate analyses. Optimal cut-off values for low and high expressers were determined post hoc.
In the studied population, cetux benefit was significant for DFS (HR = 0.71 [0.50;1.00] p = 0.05) but not for Overall Survival (OS) (HR = 0.79 [0.53;1.18] p = 0.25). Expression of miR-31-3p and miR-31-5p were highly correlated (R > 0.9). Higher miR-31-3p and 5p expression were associated with shorter DFS and OS (p < 0.01). Pts with low miR-31-3p levels (n = 218/435, 50%) had a non-significant benefit from cetux on DFS (HR 0.61 [0.33; 1.11] p = 0.10) and OS (HR = 0.67 [0.33; 1.35] p = 0.26). Pts with low miR-31-5p levels (n = 233/477, 49%) benefited from cetux on DFS (HR = 0.46 [0.25; 0.84] p = 0.01) and OS (HR = 0.50 [0.25; 0.99], p = 0.047) whereas high expressers did not (DFS: HR = 0.87 [0.56; 1.34] ; OS: HR = 1.01 [0.61; 1.67]). miR-31-5p was predictive of treatment effect on DFS at the 10% level (interaction tests: p = 0.09 for DFS and p = 0.103 for OS). Results were still significant after adjustment for clinical covariates.
In pts with resected stage III WT CC, miR-31-3p and 5p were prognostic of DFS and OS. Pts with low miR-31 expression benefited from cetux addition to FOLFOX for adjuvant therapy, and miR-31-5p was predictive of cetux efficacy.
Clinical trial identification
Legal entity responsible for the study
Y. Gaston Mathe: Ex employee: IntegraGen. S. Martin-Lannerée, C. Vazart, K. Fontaine: Employee: IntegraGen. A. Caumont, F. Montestruc: Consultant: IntegraGen. J. Taieb: Honoraria: Lilly, Celgene, Servier, Amgen, Roche, Merck, Sirtex, Sanofi. P. Laurent-Puig: Honoraria: IntegraGen, Merck-Serono, Amgen, Boehringer-Ingelheim, Roche, Lilly, Sanofi. All other authors have declared no conflicts of interest.