PROTACs are a rapidly evolving modality, currently sparking great excitement within the pharmaceutical industry. Due to limitations of the CRBN and VHL workhorse ligases, there is a desire to identify novel E3 ligase binders for targeted protein degradation applications. This talk will highlight our approach as well as the use of high-throughput chemistry to expedite this effort.

The development of immune checkpoint inhibitors targeting CTLA-4 and PD-1 has revolutionized cancer treatment. However, many patients fail to respond, suggesting the existence of orthogonal checkpoints.  Siglec-7 and -9 are highly expressed on tumor-infiltrating myeloid cells and are known to play suppressive roles in these cells, but with unclear roles in tumor-infiltrating T cells. We found that Siglec-7 and -9 suppress T cell activity by dephosphorylating TCR-related signaling cascades. Using a Siglec-7/9 degrader that targets membrane Siglec-7 and -9 to the lysosome for degradation, we rescued T cell effector function and reprogrammed tumor microenvironment, resulting in productive tumor control.

The human murine double minute 2 (MDM2) protein is a primary, endogenous cellular inhibitor of the tumor suppressor p53 and has been pursued as a cancer therapeutic target in the last 20 years. Although several highly potent and orally active MDM2 inhibitors have been advanced into clinical development, MDM2 inhibitors have major limitations in the clinic, including insufficient efficacy and development of clinical resistance. In this presentation, I will discuss the development of highly potent and orally efficacious MDM2 PROTAC degraders for the treatment of human AML and other types of human cancers.

CDK8 and CDK19 are kinase components associated with transcriptional Mediator complex. To extend the effects of CDK8/19 inhibition and to suppress kinase-independent activities, we have developed three series of PROteolysis TArgeting Chimeras (PROTACs) based on selective inhibitors of CDK8/19 kinases. CDK8/19 PROTACs were 10-fold more potent in the two CCNC-dependent MM lines (IC50 of 20-30 nM) than kinase inhibitors but not in the independent line.

In this study, we discovered PS-10, a molecular glue targeting Bruton's tyrosine kinase (BTK). While PS-10 doesn't bind directly to BTK, it binds to E3 ubiquitin ligase CRBN, forming a ternary complex that leads to efficient BTK degradation. Cryo-EM analysis revealed unique protein interactions, and PS-10's mechanism extends to other kinases, demonstrating broader therapeutic potential.

We present a mechanistic PK/PD modeling framework specifically tailored to Targeted Protein Degraders. Our approach enables a priori predictions to (1) guide compound design & optimization, (2) inform animal study design, and (3) assist in candidate selection. To explore the full potential and requirements we’ll draw on experiences with our in-house degrader pipeline. Impact and potential of the fully model-informed degrader development shall inspire the audience for their own work.

Proteolysis-targeting chimera (PROTAC) protein degraders are heterobifunctional small molecules that recruit a protein of interest to an E3 ubiquitin ligase, leading to proteasomal degradation of the target protein. This presentation will: 1) provide an overview of PROTAC technology, including properties distinguishing PROTACs from other modalities and 2) discuss physicochemical property guidelines and machine learning models for attaining oral absorption in the beyond Rule of 5 space occupied by PROTACs.

Mechanistic modeling approaches have advantages to predict in vivo properties: they are based on physiological relevance and can support additional scalars such as safety margins and drug-drug interaction risk assessment; they are interpretable to guide molecular design; and they are less likely to be influenced by human bias. This work incorporates recent approaches to predict in vivo PK properties and dose projection, and it also validates in vitro to in vivo correlation (IVIVC) analysis to support multiparameter optimization (MPO).

We describe the application of AlphaSeq, a high-throughput, highly sensitive experimental platform for measuring protein-protein interactions, to elucidate >100 novel interactions between therapeutically relevant targets and diverse set of ligases. We further characterize these PPIs through site-directed mutagenesis to prioritize actionable pairs for rational molecular glue discovery. Finally, we depict the systematic AlphaSeq validation and hit-finding approaches we have employed to identify small molecules that enhance these weak ligase-target interactions.

Molecular glues function by binding to a target and reconfiguring its surface to cooperatively engage another target. Motivated by the largely serendipitous nature of molecular glue discovery, we developed a high-throughput chemistry (HTC)-based approach to prospectively discover molecular glues. By systematically installing structural modifications onto a pre-existing ligand of interest, we can discover rare modifications that enable a ligand to function as a molecular glue.

Gain-of-function drug mechanisms are surprisingly common and can often be accessed by minor modification of existing and well-characterized chemical matter that produces new chemically induced proximity events. We have discovered, through chemically tagging or linking together potent inhibitors, multiple such mechanisms. Structural studies demonstrate minimal fortuitous chemically induced protein-protein interactions can drive cooperative binding and generate powerful pharmacology.

Leveraging induced proximity and degradation proteomics, we discovered a novel CRBN-based molecular glue degrader of HuR/ELAVL1, an RNA binding protein abnormally activated in cancer and other diseases. The MGD is moving to the clinics to treat BRAF mutant cancers as a monotherapy, while its efficacy in other disease models, including cancer cachexia, has been validated. The mechanistic studies of HuR degrader in these diseases are going to be discussed.