Clinico-pathological associations and concomitant mutations of the RAS/RAF pathway in metastatic colorectal cancer

Abstract

Background: Over the past few years, next-generation sequencing (NGS) has become reliable and cost-effective, and its use in clinical practice has become a reality. A relevant role for NGS is the prediction of response to anti-EGFR agents in metastatic colorectal cancer (mCRC), where multiple exons from KRAS, NRAS, and BRAF must be sequenced simultaneously.

Methods: We optimized a 14-amplicon NGS panel to assess, in a consecutive cohort of 219 patients affected by mCRC, the presence and clinico-pathological associations of mutations in the KRAS, NRAS, BRAF, and PIK3CA genes from formalin-fixed, paraffin-embedded specimens collected for diagnostics and research at the time of diagnosis.

Results: We observed a statistically significant association of RAS mutations with sex, young age, and tumor site. We demonstrated that concomitant mutations in the RAS/RAF pathway are not infrequent in mCRC, and as anticipated by whole-genome studies, RAS and PIK3CA tend to be concurrently mutated. We corroborated the association of BRAF mutations in right mCRC tumors with microsatellite instability. We established tumor side as prognostic parameter independently of mutational status.

Conclusions: To our knowledge, this is the first monocentric, consecutively accrued clinical mCRC cancer cohort tested by NGS in a real-world context for KRAS, NRAS, BRAF, and PIK3CA. Our study has highlighted in clinical practice findings such as the concomitance of mutations in the RAS/RAF pathway, the presence of multiple mutations in single gene, the co-occurrence of RAS and PIK3CA mutations, the prognostic value of tumor side and possible associations of sex with specific mutations.

Keywords: Anti-EGFR; Concomitant mutations; Extended RAS; Metastatic colorectal cancer; RAS/RAF pathway.

Fig. 1

Map of type of mutations and their mutual relations. Top: bar plot of mutational frequency for each gene. Right side: bar plot of number of mutations for each patient. Main figure: mutual relationship of mutational events by patients

Fig. 2

Multiple correspondence analysis (MCA) defines underlying the structure of clinical and pathological associations in the KRAS, NRAS, BRAF, and PIK3CA genes in mCRC. x- and y-axes represent the first and second dimension (Dim.1 and Dim.2) of the MCA analysis performed on clinical and pathological data from 219 mCRC patients collected in our center. In particular, we found a statistically significant association of RAS mutations with sex, young age and tumor site (lower left region) and BRAF mutations with anatomical site and old age (upper left region)

Fig. 3

Kaplan–Meier survival analyses and forest plot according to gene mutation status. Correlation between the pathway mutations and progression-free survival (a). wt, wild-type; RAS mt, RAS mutated; RAF mt, RAF mutated. Forest plot of Cox proportional hazards regression with RAS/RAF mutation status and PIK3CA. mutation status (b). p-value < 0.05 is highlighted with one asterisk

Fig. 4

Kaplan–Meier survival analyses according to anatomical site. Correlation between tumor side and progression-free survival