Shortly beforehand, the discovery process and evaluation of CBLB inhibitors with a different scaffold were published in the Journal of Medicinal Chemistry. The paper highlights Compound 10, a potent, orally available lead compound with novel scaffold promising druggability, providing an AI-driven optimization roadmap for CBLB inhibition strategies.
These concurrent achievements in both internal research and development and high-impact academic publications underscore the robust output of Pharma.AI, Insilico's end-to-end AI platform spanning across biology, chemistry, clinical research and scientific research. They demonstrate Pharma.AI's capability to not only accelerate the identification of novel therapeutic candidates but also to produce scientifically rigorous, publishable research that advances the field.
With high expression across multiple tumor types and function as a master negative regulator of immune cell activation, CBLB has emerged as a highly attractive target for cancer immunotherapy, highlighting CBLB inhibition as a potential strategy for immune system activation, both as monotherapy or combination option with anti-PD-1/PD-L1 treatment or other immune checkpoint inhibitors.
As a relatively novel target in immuno-oncology, drug R&D targeting CBLB remains at an early stage compared with well-established targets such as PD-1 and CTLA-4. Available research indicates that most CBLB-targeted candidates are in preclinical or phase I clinical trials, suggesting considerable potential for further molecular optimization of CBLB-targeted agents. In this study, Insilico team aims at systemic improvement on potency and pharmacokinetic profiles, through core replacement and structure-activity relationship (SAR) exploration.
The research started with ligand-based design using Chemistry42, Insilico's generative AI platform for molecule generation and optimization, hoping to obtain novel, improved core structures. Afterwards, scaffold hopping efforts yielded Compound 1, the initial hit compound with relatively weak in vitro activity but significantly improved pharmacokinetic properties.
Following the hit identification, the team proceeded to co-crystal structure analysis and further SAR exploration focused on certain regions of the molecule, leading to the discovery of compound 10, which demonstrated strong in vitro activity and improved metabolic stability in mouse, rat, and dog models.
In in vivo studies using the CT26 colorectal cancer mouse model, Compound 10 was evaluated as both a monotherapy and in combination with an anti-PD-1 antibody. Key findings include:
Safety and Tolerability: All doses of Compound 10 were well-tolerated by the mice, with no significant weight loss observed.
Combination Potential: When combined with an anti-PD-1 antibody, Compound 10 significantly inhibited tumor growth, demonstrating a remarkable synergistic effect.
Potent Anti-tumor Effects: In the medium-dose combination group, three out of six mice achieved complete tumor regression, while in the high-dose group, all six mice achieved complete regression.
Durable Immune Memory: After treatment was discontinued, mice in the high-dose group showed no tumor recurrence, and developed a durable immune memory, protecting them against a rechallenge with the same tumor cells.
Harnessing state-of-the-art AI and automation technologies, Insilico has significantly improved the efficiency of preclinical drug development, setting a benchmark for AI-driven drug R&D. While traditional early-stage drug discovery typically requires 2.5 to 4 years, Insilico has nominated 20 preclinical candidates with an average timeline—from project initiation to preclinical candidate (PCC) nomination—of just 12 to 18 months per program, with only 60 to 200 molecules synthesized and tested in each program.