Rare RAS mutations may not undermine cetuximab benefit in colorectal cancer

Metastatic colorectal cancer patients whose tumors carry rare, non-hotspot mutations in the RAS and RAF genes did not appear to lose the benefit of cetuximab therapy in a small retrospective study, according to an analysis published in Pathology and Oncology Research. The finding challenges the assumption that every mutation within the RAS genes’ exon 2 through exon 4 regions automatically predicts resistance to anti-EGFR antibody treatment, an assumption that currently guides drug labeling and clinical testing across Europe.

The study, led by József Tímár of Semmelweis University in Budapest, re-examined stored tumor samples from nineteen patients with metastatic colorectal cancer who had all been treated with cetuximab at the Hungarian Defense Forces Medical Center. Every patient in the cohort had originally been enrolled on the basis of a KRAS exon 2 wild-type result, meaning the standard first-line molecular test had found no mutation at codons 12 and 13, the most common sites of RAS mutation in this cancer. The researchers then applied next-generation sequencing to formalin-fixed, paraffin-embedded tissue samples, all of which had been collected before any anti-EGFR therapy began, to map the full spectrum of RAS and RAF mutations across each tumor.

The sequencing used the Ion AmpliSeq Cancer Hotspot Panel v2, a targeted panel covering K-RAS, N-RAS, H-RAS and BRAF, with a stated variant allele frequency sensitivity of five percent. Average depth of coverage exceeded 300 reads, on-target mapping surpassed 90 percent, and variant annotation drew on the NCBI ClinVar and Franklin databases, with alleles categorized for pathogenicity. Testing was performed blind to clinical information, and tumor-rich areas were macrodissected under light microscopy before DNA extraction. In most cases the primary tumor was available for analysis, but in several instances only metastatic samples passed quality checks; the authors justified using metastatic tissue by noting that large-scale genomic studies have confirmed major driver gene profiles are generally maintained in colorectal cancer metastases.

Out of the nineteen patients, six tumors proved fully wild type for RAS and RAF. The remaining thirteen carried mutations, which the team sorted into two groups. Eight cases harbored hotspot mutations, defined as changes at RAS codons 12, 13, 59, 61, 117 or 146, or at BRAF codon V600, the established sites whose mutation is known to blunt anti-EGFR response. Five cases carried only rare, non-hotspot alterations. In total the researchers detected thirteen rare RAS/RAF mutations, five of which occurred alongside hotspot changes, while the other eight appeared in the five non-hotspot cases. Those rare variants included KRAS E37 and C51Y, NRAS D57N, E62K and G60E together with V14I, HRAS P34L, and BRAF G464R. When any hotspot mutation co-occurred with a rare one, the case was classified as hotspot for the survival analysis.

The clinical picture that emerged from Kaplan-Meier analysis was striking. Progression-free survival curves for the wild-type and non-hotspot groups ran parallel to each other, while the hotspot group separated away, though the three-way comparison did not reach statistical significance, with a p-value of 0.156. Comparing wild-type cases directly against hotspot cases gave a significant difference in progression-free survival, with a p-value of 0.033 and a hazard ratio of 4.161. When the wild-type and non-hotspot groups were combined and compared against the hotspot group, the difference in progression-free survival approached significance at p equals 0.057, with a hazard ratio of 3.501 and a confidence interval of 1.12 to 10.942.

Overall survival told a similar and somewhat stronger story. The survival curve of wild-type cases overlapped that of non-hotspot cases but clearly separated from the hotspot cases, with a p-value of 0.024 for that three-way comparison. Direct comparison showed overall survival of wild-type cases differed significantly from hotspot cases, with a p-value of 0.005 and a hazard ratio of 6.362 and a confidence interval of 1.496 to 27.055, while wild-type versus non-hotspot comparison was not statistically different, with a p-value of 0.092 and a hazard ratio of 0.891 and a confidence interval of 0.248 to 3.205. Combining the wild-type and non-hotspot groups into a single arm produced a significant overall survival difference from the hotspot group, with a p-value of 0.005 and a hazard ratio of 6.693 and a confidence interval of 1.749 to 25.61. When hotspot and non-hotspot cases were combined against wild-type cases, neither progression-free nor overall survival differences reached significance, at p-values of 0.137 and 0.262 respectively.

The authors are careful to frame these results as hypothesis-generating rather than definitive. The cohort is very small, the statistical power of the survival analyses is correspondingly limited, and the groups differ clinically in ways that complicate interpretation. Two wild-type cases and one hotspot case received cetuximab plus FOLFIRI as first-line therapy, whereas all five non-hotspot cases were treated in the second line, mostly after a bevacizumab plus FOLFIRI regimen. Tumor stages and metastatic sites also varied across the groups, with wild-type and non-hotspot cases showing mostly liver metastases while the hotspot group was more heterogeneous in metastatic pattern. These imbalances mean the survival curves cannot be read as a clean experimental contrast.

Why the rare mutations matter clinically comes down to how regulatory labels are written. Anti-EGFR antibodies entered colorectal cancer management nearly two decades ago, and initial observational studies showed that patients whose tumors carried KRAS exon 2 mutations at codons 12 and 13 were less likely to respond. Later work identified additional mutations in exons 2, 3 and 4 of KRAS and NRAS that also influenced responsiveness. In July 2013 the European Medicines Agency restricted panitumumab to patients whose tumors were RAS wild type across exons 2, 3 and 4, including codons 12, 13, 59, 61, 117 and 146, and in November 2013 it ruled that cetuximab should likewise be given only to metastatic colorectal cancer patients without RAS mutations, determined by validated testing methods. The BRAF V600 mutation was subsequently established, largely through retrospective clinical studies, as another negative predictive marker for these antibodies.

The biology of why hotspot mutations confer resistance is well mapped. RAS proteins act as molecular switches that must be toggled off by GTPase activating proteins, or GAPs, which trigger hydrolysis of GTP to GDP; the proteins’ own intrinsic hydrolytic activity is too slow to do the job alone. Hotspot mutations at codon 12, which sits in the GAP-binding region, and at codon 61, which is central to GTP hydrolysis, render the proteins insensitive to GAP-mediated inactivation and drastically diminish their intrinsic GTPase activity. Notably, the KRAS G12C variant is an exception, retaining roughly 75 percent of wild-type GTPase activity, and G13D, A59T and K117 show elevated nucleotide exchange rates while reducing GAP sensitivity or hydrolytic activity. Mutations at A146 enhance nucleotide exchange without affecting GAP sensitivity. BRAF, another member of the same signaling pathway, is mutated in roughly ten percent of colorectal cancers, predominantly at V600 in exon 15, which constitutively elevates kinase activity; class-I V600 mutations confer resistance to anti-EGFR therapy, class-II mutations moderately raise kinase activity, and class-III mutations lower it.

What the rare variants in this cohort actually do functionally remains largely unresolved. The NRAS D57N and G60E variants had been observed before, and functional assays indicated D57N does not change NRAS activity while G60E produces a moderate increase; G60E has also been detected in Noonan syndrome, a RASopathy. The functional activity of KRAS E37 and C51Y, NRAS V14I and E62K, and HRAS P34L is not known, though their positions fall in key structural regions such as the P-loop for GTP and GDP binding and the switch I and II regions for protein-protein interactions. The non-hotspot BRAF G464R case was a class-II mutant. A recent biochemical and bioinformatic study covering 935 hotspot and non-hotspot RAS variants performed functional analysis on only a small portion of them, none of which matched the variants found here. Earlier retrospective work on 4,122 anti-EGFR-treated colorectal cancer patients identified 54 rare mutant cases with lower response rate and disease control rate, but that cohort contained fifteen well-established hotspot mutations that may have influenced the results, and no progression-free or overall survival analysis was performed. A preclinical study of rare KRAS mutations and cetuximab found effects ranging from resistance through neutrality to preserved sensitivity, but none of the mutations tested there appeared in this cohort. The authors conclude that without careful evaluation of the functional consequences of individual mutations, clinical outcome cannot be properly predicted, and they call for similar retrospective analyses on much larger cancer series to test whether the blanket exclusion of all exon 2 through 4 RAS mutations from anti-EGFR therapy is justified.

Subject of Research: Predictive value of rare non-hotspot RAS and RAF mutations for cetuximab efficacy in metastatic colorectal cancer

Article Title: A retrospective analysis of cetuximab efficacy based on the comprehensive RAS/RAF mutation profiles of colorectal cancer

Article References: Tímár, J., Rásó, E., Barbai, T., Baranyi, M., Kenessey, I., Pápai, Z., & Uhlyarik, A. (2026). A retrospective analysis of cetuximab efficacy based on the comprehensive RAS/RAF mutation profiles of colorectal cancer. Pathology & Oncology Research, 32, Article 1612560. https://doi.org/10.3389/pore.2026.1612560

Image Credits: AI Generated

DOI: 10.3389/pore.2026.1612560

Keywords: colorectal cancer, cetuximab, RAS mutations, RAF mutations, non-hotspot mutations, anti-EGFR therapy, next-generation sequencing, progression-free survival, overall survival, European Medicines Agency, BRAF V600, biomarker testing