Generative AI helps discover ISM3830's novel framework

Generative AI helps discover ISM3830's novel framework

Casitas B-lineage lymphoma-b (CBLB) is an E3 ubiquitin-protein ligase that acts as an intracellular checkpoint and masters the negative regulator of T cell and natural killer (NK) cell activation.

According to previous studies, CBLB is highly expressed in many immune system subgroups of various cancers, indicating its great potential as a therapeutic target for cancer immunotherapy, especially in advanced colorectal cancer, prostate cancer, renal carcinoma, melanoma, etc.

Insilico Medicine (“Insilico”), a clinical-stage generative AI-based drug discovery company, today announced the nomination of ISM3830, a potentially first-in-class, orally available, highly selective CBLB inhibitor with AI-powered novel scaffold, as a preclinical candidate against multiple advanced tumors.

“With repeated real-world evidence, including the nomination of preclinical candidates and positive clinical results, we are even more confident that generative AI has the potential to accelerate drug development and enable true therapeutic innovation,” said Feng Ren, Ph.D., Co-CEO and CSO of Insilico Medicine.

Equally important, CBLB inhibition's fundamental mechanism of action supports indications with low response or resistance to current immune checkpoint inhibitors - an area of ​​significant unmet need that we aim to address with the combined power of AI speed and human intelligence."

Feng Ren, Ph.D., Co-CEO and CSO, Insilico Medicine

It is worth noting that the novel scaffold of ISM3830 with a similarity index to the currently available molecule of 0.42 was discovered using Chemistry42 and its ADMET predictor module. Using a structure-based strategy based on currently available cocrystal structures, new candidate compounds were generated by more than 40 generative AI models integrated into Chemistry42, evaluated using the ADMET predictor module, and further optimized in repeated generation experiments guided by integrated reward pipelines. Subsequently, synthesis and comprehensive bioassay testing finally revealed that ISM3830 is a promising drug candidate, whose specific advantages are listed as follows:

• Powered by Insilico's proprietary generative AI platform, the drug candidate overcomes the current metabolic and absorption bottlenecks of CBLB inhibitory therapies.

• Preclinical safety screening demonstrated a low risk of hypotension, gastrointestinal toxicity and off-target toxicity in the DRF study, as well as excellent selectivity and an improved safety margin.

• Promising drug availability and excellent ADME/PK profilein vitroAndin vivomeasured at the bottomin vivoClearance and higher oral bioavailability in all preclinical species.

Preclinical studies suggest robustnessin vivoEfficacy in mouse models and induction of long-term tumor immunity confirmed by CT26 rechallenge experiments. In addition, there was potential for combination with a wide range of treatment options, including immune checkpoint inhibitors, chemotherapy and other targeted agents.

In addition to the milestone achievement of the PCC nomination, Insilico's R&D team recently published the development process of a CBLB inhibitor of the same series in a peer-reviewed article in the Journal of Medicinal Chemistry entitled "Discovery and Biological Evaluation of Novel, Potent, and Orally Available CBLB Inhibitors", demonstrating how cocrystal structure analysis, iterative structure-activity relationship (SAR) exploration, systematic optimization of potency, selectivity, etc Pharmacokinetic profiles, combined with structure-based AI generation results, could provide a roadmap for further optimization of CBLB inhibitors.

In March 2024, Insilico published in Nature Biotechnology the path of the rentosertib program from launch to Phase 1 clinical trials, along with a portion of the experimental data. In December 2024, Insilico published the AI-assisted preclinical research journey and partial experimental data for the intestinal-restricted PHD1/2 inhibitor ISM5411 in Nature Biotechnology. In January 2025, Insilico, along with partners including the University of Toronto, published research in Nature Biotechnology exploring generative AI using a quantum-classical hybrid model to develop novel KRAS inhibitors. In May 2025, Insilico published collaborative research on AI-assisted development of pan-coronavirus inhibitors in Nature Communications. In the same month, Insilico published another Nature Communications article on AI-assisted development of next-generation ENPP1 inhibitors to modulate the innate immune system.

Notably, Insilico's June publication in Nature Medicine marked the world's first clinical proof-of-concept for AI drug discovery through the success of Rentosertib (ISM001-055), the potential first-in-class drug candidate with a novel target and innovative structure discovered using AI.

By integrating AI and automation technologies and extensive in-house drug discovery capabilities, Insilico delivers innovative drug solutions for unmet needs including fibrosis, oncology, immunology, pain, and obesity and metabolic disorders, nominating 23 preclinical candidates as of 2021, of which 10 molecules have received IND approval.

Sources: