Abstract LB030: Building a platform of validated high throughput screening approaches for molecular glue degrader identification

Abstract Over recent years, Induced proximity therapeutics (IPT) and targeted protein degradation (TPD) approaches have become increasingly popular. Whilst significant initial efforts focused upon heterobifunctional degraders, molecular glue degraders (MGDs) represent the next generation proximity-based drugs, but their rational discovery remains challenging. Here, we describe unbiased and broadly applicable approaches for molecular glue hit ID through the development of a suite of orthogonal and highly complementary screening approaches, relying on Homogeneous Time-Resolved Fluorescence (HTRF), Surface Plasmon Resonance (SPR) and High Content Imaging (HCI). To demonstrate the utility of these assays, we have focussed on the therapeutically relevant protein SOS1 as a model target. SOS1 is the major guanine nucleotide exchange factor (GEF) for KRAS, where inhibition of the SOS1/KRAS interaction reduces tumor growth in preclinical settings, and small molecule inhibitors of the SOS1/KRAS interaction are currently undergoing clinical testing for KRAS driven cancers. Our HTRF approach measures the induced interaction between the ubiquitin ligase CRBN/DDB1 and the SOS1 protein. The assay was successfully developed and validated in a miniaturized format utilizing a bifunctional tool compound as a positive control. Complementing our HTRF based approach, we also developed a highly sensitive SPR assay for detecting compounds that generate a ternary complex with CRBN/DDB1. These assay formats are amenable for large scale HTS screening, and as proof-of-concept we have established workflows and screened compound collections from the Sygnature LeadFinder and fragment libraries, in each of these assay platforms. Our combined biochemical/biophysical approach to glue identification using HTRF and SPR demonstrates the successful and scalable use of these methodologies for HTS, with the exciting potential for identification of novel MGD candidates for SOS1. To further complement these strategies for glue identification, we have additionally developed a functional cell-based approach utilizing HCI. HCI offers a medium-high throughput approach to monitor endogenous protein abundance and localization in complex cellular models. One major advantage of imaging assays is that they are amenable to multiplexing, allowing simultaneous detection of target engagement, biomarker analysis and compound toxicity. Here, we showcase HCI as a robust, high throughput method of screening for induced protein degradation at the endogenous level, again focusing on the KRAS pathway. Using this approach we quantified endogenous protein abundance in response to compound treatment, as well as downstream effects in the RAS pathway. Together, these approaches demonstrate multiple, complementary options for identifying molecular glue degraders. Further, the described assays are also applicable for screening and characterization of bifunctional degrader compounds. This platform is poised to identify high quality lead compounds across a range of oncology targets in areas of unmet clinical need. Citation Format: Allan Jordan, Philip Addis, Toby Allen, Tarun Ayra, Sarah Beck, Roslyn Brant, Roland Hjerpe, Benoit Gourdet, Martin Jennings, Rachel Lawrence, Sigrun Campbell-Maurer, Pei Cing Ng, Lyn Parkinson, Eva Rejnowicz, Gonzalo Robles, Elizabeth Rosethorne, Joshua Shaw, Duncan Smith, Denise Swift, Daniel Tait, Stuart Thomson, Chris Tomlinson, Ailsa Townley, Stephanie Ward, Clare Wilson. Building a platform of validated high throughput screening approaches for molecular glue degrader identification abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts) ; 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84 (7Suppl): Abstract nr LB030.