RAS/BRAF testing of circulating cell-free DNA (cfDNA) in RAS/BRAF wild-type (RAS/BRAF ^WT ) metastatic colorectal cancer (mCRC) patients: Preliminary results of the phase III LIBImAb study.

Abstract

e15570 Background: The LIBImAb Study is a Phase III, randomized, open-label, comparative, multicenter trial designed to assess the efficacy in terms of progression-free survival (PFS) of bevacizumab versus cetuximab in combination with FOLFIRI regimen in untreated patients (pts) with RAS/BRAF WT mCRC on tumor tissue and RAS/BRAF mutations detected through cfDNA analysis. The study is supported by the Italian Drug Agency (AIFA). We report preliminary data on liquid biopsy testing. Methods: Pts with mCRC, confirmed as RAS/BRAF WT on tumor tissue and RAS/BRAF mutated on cfDNA before systemic therapy (baseline), are randomized to receive either FOLFIRI/cetuximab or FOLFIRI/bevacizumab. Pts with RAS/BRAF WT cfDNA at baseline are treated with FOLFIRI/cetuximab for up to 8 cycles. Pts without clinical progression are retested for RAS/BRAF mutations on cfDNA. RAS/BRAF mutated pts at 8-cycles are randomized to continue FOLFIRI/cetuximab or to switch to FOLFIRI/bevacizumab. Plasma samples from enrolled pts are analyzed for variants in exons 2, 3, and 4 of the KRAS and NRAS genes, and for BRAF V600 mutations, by the fully automated Idylla system (Biocartis). Results: Between August 2021 and January 2025, 410 pts were screened, with 43 randomized. Characteristics of all screened pts included 63.2% males and 34.8% females; 78.6% had left colon tumors, and 17.7% had right colon tumors; 93% had ≤2 metastatic sites at baseline, and 6.6% had >2; liver metastases were present in 34.8%, and lung metastases in 12.2%. Liquid biopsies were performed at baseline and after 8 cycles of treatment on 410 and 230 pts, respectively. The analysis of baseline cfDNA detected RAS mutations in 26 pts (6.3%) and BRAF V600 mutations in 3 pts (0.7%). At the 8-cycles test on cfDNA, RAS mutations were observed in 19 pts (8.2%). A multivariable logistic regression model was employed to assess the effect of pts’ characteristics (age, sex, site of primary tumor, number, and site of metastases) on mutation detection considering pts with results available for baseline and 8-cycles cfDNA tests. The only significant finding for mutation detection in cfDNA was the number of metastatic sites at baseline, with an odds ratio (OR) of 5.22 (95% CI, 1.69-16.10; p=0.0040) for pts with >2 metastatic sites vs ≤2 sites. In pts tested at baseline, the OR for metastatic site number >2 vs ≤2 was 2.58 (95% CI, 0.83-8.00; p=0.1001), while in those who received a test at 8 cycles the OR was 4.91 (95% CI, 1.06-22.80; p=0.0422). Conclusions: Liquid biopsy may more accurately reflect the heterogeneity of mCRC and could be valuable in clinical practice to complement tissue testing. These preliminary results suggest that the probability to detect RAS/BRAF mutations in cfDNA in pts with RAS/BRAF WT mCRC on tissue is higher in pts with >2 metastatic sites, who are more likely to carry a more heterogeneous disease.