Command Palette

Search for a command to run...

Preprint WatchModerateNovember 5th, 2026

Fragment Covalent Bond Formation at SOS1 Tyr-884 Inhibits KRAS Nucleotide Exchange

Bum-Erdene, K.; Gonzalez-Gutierrez, G.; Ghozayel, M. K.; Meroueh, S.

A covalent fragment forming an adduct at SOS1 Tyr-884 inhibits SOS1-mediated KRAS nucleotide exchange, offering a covalent alternative to non-covalent SOS1 antagonists.

Moderate contradiction

3 prior failures

Two documented clinical failures match this mechanism, or a single Phase 3 failure is on record.

This preprint reports a covalent fragment, 11 (SOF-599), that forms an adduct predominantly at SOS1 Tyr-884 and inhibits KRAS nucleotide exchange with a half-maximal inactivation concentration of 12 micromolar and a reaction half-life of 6 hours, proposed as a starting point for covalent SOS1 inhibitors. The Claidex graph carries three KRAS programs that stopped early, and none of them stopped for lack of efficacy. BGB-53038 was shelved at Phase 1 when the sponsor pivoted to a degrader (bgb-53038-kras-solid-tumors-phase1-strategic-reprioritization), the KontRASt-03 combination platform closed before its combinations reported (opnurasib-kras-g12c-combination-platform-kontrast-03-phase1-2-termination), and a ctDNA-selected stage III trial of sotorasib ran out of an eligible population (sotorasib-durvalumab-kras-g12c-ctdna-stage3-nsclc-phase2-enrollment-collapse). The documented attrition at this node is therefore about portfolio choices and recruitable populations rather than about target biology, so a new chemical starting point upstream of KRAS does not speak to the recorded failure modes. A covalent SOS1 series would still have to clear the same operational constraints that closed the three programs on file.

Abstract excerpt

The activation of the KRAS oncogene from its inactive GDP-bound state to its active GTP-bound state is mediated by Son of Sevenless (SOS) exchange factors. Small-molecule inhibition of SOS1 has been achieved, with three compounds currently in clinical trials. These compounds bind to SOS1 and disrupt a crucial interaction between SOS1 Tyr-884 and KRAS Arg-73. Here, we screen libraries of aryl sulfonyl fluorides to explore covalent bond formation at Tyr-884 as a strategy for covalent inhibition of SOS1-KRAS binding. Initially, ten fragments were tested and several inhibited KRAS nucleotide exchange by SOS1. X-ray crystal structures revealed that one reaction site for the fragments was Tyr-884, but mass spectrometry and nucleotide exchange assays with wild-type and mutant SOS1 indicated additional reaction sites. To identify candidates specific to Tyr-884, we conducted a larger screen and identified fragment 11 (SOF-599). The fragment formed an adduct predominantly at Tyr-884 and inhibited KRAS nucleotide exchange by SOS1 in a time- and concentration-dependent manner. Fragment 11 (SOF-599) exhibited a half-maximal inactivation concentration (KI) of 12 micromolar and a reaction half-life of 6 h. These findings establish 11 (SOF-599) as a promising starting point for the development of covalent SOS1 inhibitors as an alternative strategy to existing non-covalent antagonists.

Matching Claidex post-mortems

3 of 3 indexed

This is an automated contradiction flag, not an editorial judgment on the preprint's quality. Flags identify where the preclinical literature and the clinical failure record diverge.