KRAS mutant (KRAS MT) and RAS/BRAF wild type (RAS WT) colorectal cancer (CRC) responds differently to targeted agents. A better understanding of early effects on signal transduction following exposure to targeted agents may explain this and inform future therapeutic strategies.
A panel of 25 CRC cell lines (10 KRAS MT, 15 RAS WT) and 13 samples isolated from CRC patients’ (pt) serous effusions were used. Using a multiplex antibody-based platform, simultaneous changes in 60 phosphorylated (p) proteins were quantified following 1-hour exposure to clinically used concentrations of: gefitinib, PI3K inhibitor: pictilisib, AZD5363, everolimus, trametinib and vemurafenib. Statistical analysis included logistic regression, T tests and Spearman’s correlation.
Collectively pMEK was the most commonly upregulated protein in KRAS MT cell lines; 55% KRAS MT vs 18% RAS WT, a difference that was significant (P < 0.05) with pictilisib and gefitinib. Triplicate validation with pictilisib confirmed upregulation of pMEK, as well as pERK, pMSK, pGSK3α in KRAS MT but not RAS WT cell lines. KRAS MT cells had a higher mean pictilisib half maximal inhibitory concentration (P < 0.05), being highest in cells with the greatest pMEK rise. Validation with gefitinib confirmed consistent downregulation of pMEK, pMSK, pGSK3α/β, pPRAS40 and pS6 in sensitive RAS WT but not KRAS MT cells (P < 0.05), despite comparable decrease of pEGFR. Gefitinib sensitivity was related to the degree of downregulation of the aforementioned proteins. In 10 KRAS MT pt samples, pMEK upregulation was most commonly seen with pictilisib (38% of pts). As in cell lines, gefitinib led to upregulation of intracellular effectors in KRAS MT pt samples e.g. mTOR in 40% of pts. In RAS WT cells compensatory activation of receptor tyrosine kinases was the commonest rewiring feature, e.g. pFGFR1 upregulation with gefitinib in 40% of cell lines and 66% of pts.
There are significant differences in the rewiring of signal transduction between KRAS MT and RAS WT CRC cells exposed to clinically relevant concentrations of PI3K and EGFR inhibitors. These differences can help devise/refine future strategies to combine targeted agents in CRC.
Clinical trial identification
Legal entity responsible for the study
Royal Marsden NHS Foundation Trust.
Cancer Research UK.
All authors have declared no conflicts of interest.