Covalent inhibition of KRAS^G12C has emerged as a promising strategy for the treatment of non-small cell lung cancers (NSCLC) that harbor this oncogenic mutation. To date, the identification of structural starting points for covalent KRAS^G12C inhibitors has relied on either the high-throughput screening (HTS) of small molecules bearing cysteine-reactive electrophiles (e.g., diverse acrylamide- or disulfide-containing libraries) or on the attachment of cysteine-reactive warheads to high-affinity non-covalent binders. Although the development of LUMAKRAS^® (sotorasib) from the optimization of leads identified through the screening of purpose-built cysteine-reactive libraries demonstrates the effectiveness of such approaches in identifying useful starting points even for traditionally ‘undruggable’ targets, such screening approaches have historically been limited in the range of chemical space that can be effectively sampled due to limitations on the size and diversity of such libraries.
DNA-encoded library (DEL) technology has emerged as a promising tool for identifying ligands for challenging protein targets, and has allowed for the accelerated preparation and screening of incredibly large chemical libraries with significantly reduced costs relative to HTS approaches. While this technology has most often been used to identify reversible (non-covalent) binders of target proteins, recent progress has led to the discovery of irreversible binders using DELs bearing covalent warheads. Here, we report the successful use of a covalent DEL platform to screen ~16 million compounds against KRAS^G12C, leading to the identification of a structurally distinct class of covalent KRAS^G12C inhibitors. We additionally describe the details of library production and screening, data analysis, and hit validation, as well as the optimization of the resulting screening hits. These efforts have culminated in the identification of structurally novel potent and selective covalent KRAS^G12C inhibitors, whose structural characterization provides additional insight into effective approaches for inhibiting the function of this challenging therapeutic target.

Alex Pickrell

Appearances