Combining cutting edge computational and experimental methods for targeting KRAS mutations in non-small cell lung cancer.

Historically, KRAS mutations have been notoriously difficult to target despite their status as the most commonly mutated oncogene in the RAS gene family. Pioneering work by Shokat and colleagues has led to the discovery of KRAS G12C-GDP mutant-specific inhibitors, with two such inhibitors adagrasib and sotorasib now FDA approved for treatment of non-small cell lung cancer (NSCLC). Unfortunately, several patients did not achieve full treatment response. Further drug discovery is urgently needed to identify compounds capable of synergizing with available KRAS G12C inhibitors to prevent drug resistance, pan-KRAS inhibitors capable of binding multiple KRAS mutations, and KRAS-GTP inhibitors. This review encompasses the development of the first KRAS G12C inhibitors to recent advances in precision oncology utilizing artificial intelligence (AI) to identify compounds capable of targeting KRAS G12C, D, and V individually, as well as pan-KRAS and SOS1 inhibitors. Recent studies support the view that integration of AI algorithms with experimental methods is a key aspect in stream-lining the drug discovery process and identifying molecules with greater structural diversity, less off-target effects than traditional screening methods. Furthermore, the authors believe that AI will eventually become standardized in drug discovery for aggressive driver oncogenes across multiple cancers.