AI-designed CDK12/13 inhibitors target resistant cancers

Insilico Medicine, a biotechnology company leveraging generative artificial intelligence (AI), has announced a significant breakthrough in cancer therapeutics. The company’s latest study, published in the Journal of Medicinal Chemistry, details the discovery of a novel series of orally available covalent CDK12/13 dual inhibitors, designed to combat refractory and treatment-resistant cancers.

Harnessing AI for drug discovery

The study highlights compound 12b, a potent, selective, and safe therapy that targets CDK12/13, a critical component in the DNA damage response (DDR) pathway. Using its proprietary AI-driven platforms, PandaOmics and Chemistry42, Insilico Medicine successfully designed and refined these inhibitors to address prior challenges of toxicity and inefficacy observed in earlier covalent and non-covalent inhibitors.

CDK12 and CDK13 belong to the cyclin-dependent kinase family and play a crucial role in maintaining genomic stability by regulating DDR genes. This function is particularly important in oncology, as tumours often develop resistance to antitumour therapies due to genomic instability. Insilico Medicine’s novel inhibitors have demonstrated promise in overcoming these resistance mechanisms.

Targeting a broad range of cancers

Insilico Medicine employed PandaOmics to identify CDK12 as a high-priority target through multiomics and literature-based analyses. Further AI-assisted indication prioritisation revealed several key cancer types where CDK12/13 inhibition could be particularly effective. These include:

• Gastric cancer
• Ovarian cancer
• Prostate cancer
• Non-small cell lung cancer
• Liver cancer
• Triple-negative breast cancer (TNBC)
• Colorectal cancer

AI is revolutionising drug discovery, accelerating the development of novel treatments for cancer, neurological disorders, autoimmune diseases, and other hard-to-treat conditions.

Promising preclinical results

During preclinical in vitro and in vivo testing, compound 12b exhibited nanomolar potencies across multiple cancer cell lines. The compound also demonstrated favourable absorption, distribution, metabolism, and excretion (ADME) properties, along with significant efficacy in models of breast cancer and acute myeloid leukaemia (AML). Importantly, 12b achieved these results without inducing intolerable side effects, marking a substantial advancement in the development of safer cancer therapies.

“At Insilico Medicine, we hope to revolutionise the drug discovery process with cutting-edge artificial intelligence technologies, and the discovery of CDK12/13 dual inhibitors with potential to address treatment-resistant tumours is another step towards the promise,” stated Dr Hongfu Lu, co-lead author of the study and Senior Director of Chemistry at Insilico Medicine.

“The research and development process showcases how AI-guided drug design can achieve both precision and safety in cancer therapeutics, and we are striving to advance the compound in this programme into clinical trials in the near future.”

Clinical trials and beyond

Insilico Medicine’s latest breakthrough highlights the growing influence of AI in drug discovery, offering new hope for patients with treatment-resistant cancers. The company aims to advance compound 12b into clinical trials, supporting a new generation of targeted cancer therapies.

As the biotechnology industry continues to embrace AI-driven innovations, Insilico Medicine’s success in developing CDK12/13 inhibitors marks a significant step forward in the quest for more effective and safer cancer treatments.

This study was published in the Journal of Medicinal Chemistry.