News Release

Insilico Medicine nominates novel CDK12/13 small molecule inhibitor as preclinical candidate for multiple tumors

Business Announcement

InSilico Medicine

  • Based on AI-driven indication explorations supported by PandaOmics, CDK12 ranks high on the list of potential targets for multiple tumors.
  • ISM9274 is a highly selective covalent inhibitor of CDK12/13. 
  • In preclinical studies, ISM9274 demonstrated potent antiproliferative activity in more than 60 cancer cell lines in 13 tumor categories, particularly in triple-negative breast cancer, pancreatic cancer, and other areas of urgent clinical need with limited therapies. 

Insilico Medicine (“Insilico”), a generative artificial intelligence (AI)-driven clinical-stage drug discovery company, today announced the nomination of ISM9274, a highly selective covalent dual-inhibitor of CDK12/13 as a preclinical candidate (PCC) for cancer treatment. 

Cyclin-dependent kinase 12 (CDK12) is an important transcription-associated member of the CDK family. It shows versatile roles in regulating gene transcription, RNA splicing, translation, and intronic polyadenylation. Increasing evidence has demonstrated the critical role of CDK12 in various human cancers as both a biomarker of cancer and a potential target for cancer therapy.

Insilico’s research team utilized AI-driven indication explorations supported by PandaOmics, compared the omics data between tumor and normal tissues in over 90 disease-specific datasets across different cancers, and merged the consideration from text-based analysis tools containing clinical trials, grants, and publications. Subsequently, the target identification function integrated in PandaOmics ranked target prioritization associated with indications by the feature scores based on different algorithms like matrix factorization, causal inference, and de novo pathway reconstruction, pointing to a strong causal association between CDK12 and a wide of cancer indications including triple-negative breast cancer (TNBC), lung cancer, colorectal cancer, and pancreatic cancer.

“PandaOmics played an important role in advancing the CDK program development,” said Alex Zhavoronkov, PhD, founder and co-CEO of Insilico Medicine. “PandaOmics is built on more than 20 selected models and 10 trillion data points and is still continuously evolving. It includes a natural language Q&A system for life science information, and a knowledge graph linking diseases, genes, and drugs. We will continue to explore and enrich the functions of this target discovery engine and expand new applications in early drug discovery.”

After determining the target, Insilico researchers designed a series of compounds for CDK12 using  the company’s generative AI small molecule engine Chemistry42, and nominated ISM9274 as the preclinical candidate at the end of August 2023. In in vitro studies, ISM9274 has demonstrated potent antiproliferative activities in more than 60 cancer cell lines in 13 tumor categories, especially in triple-negative breast cancer, pancreatic cancer, and other urgent clinical needs with limited therapies. ISM9274 showed robust in vivo efficacy in multiple xenograft models for both monotherapy and combo-therapy. Moreover, the molecule has a favorable safety profile and ADME properties.

“As potential targets, CDK12/13 poses not only opportunities, but also challenges,” said Feng Ren, PhD, co-CEO and Chief Scientific Officer of Insilico Medicine. “With its robust efficacy in a variety of indications as well as satisfactory safety margins in preclinical studies, we believe that ISM9274, a novel CDK12/13 inhibitor, offers potential breakthroughs to unmet clinical needs in oncology. ISM9274 is now being progressed to IND-enabling studies and we plan to submit an IND application in the second half of 2024.”

Powered by Pharma.AI, an integrated generative AI-driven drug discovery platform spanning biology, chemistry, and clinical development, Insilico has developed a comprehensive portfolio of more than 30 pipelines with 4 programs currently in the clinical stages. The Company’s leading anti-fibrosis program is currently in Phase II clinical trials in both the United States and China. Insilico’s potentially best-in-class small molecule inhibitor of USP1 was recently licensed to Exelixis with an upfront payment of $80 million and additional milestones and royalties.


About Insilico Medicine

Insilico Medicine, a global clinical stage biotechnology company powered by generative AI, is connecting biology, chemistry, and clinical trials analysis using next-generation AI systems. The company has developed AI platforms that utilize deep generative models, reinforcement learning, transformers, and other modern machine learning techniques for novel target discovery and the generation of novel molecular structures with desired properties. Insilico Medicine is developing breakthrough solutions to discover and develop innovative drugs for cancer, fibrosis, immunity, central nervous system diseases, infectious diseases, autoimmune diseases, and aging-related diseases.


About PandaOmics

PandaOmics is a powerful generative AI software specifically designed for therapeutic target and biomarker discovery. It comprises over 20 models produced using generative AI techniques and human expert validation. PandaOmics employs multi-omics and text data with its generative AI methodologies to identify novel targets. Additionally, PandaOmics features a natural language Q&A system for life science information built on knowledge graphs. These graphs are created using transformer-based natural language processing models applied to scientific publications, clinical trials, and grant applications related to target-disease associations. With the capability to score and rank the targets based on novelty, credibility, drug-likeness, safety, and other key target selection drivers, the platform is designed to aid researchers in identifying potential therapeutic targets and biomarkers more effectively and conveniently.

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.