Artificial DNA and Intelligence
Credit: iStock/agsandrew

A drug for idiopathic pulmonary fibrosis (IPF) designed by the Hong Kong-U.S. player Insilico Medicine with the help of artificial intelligence (AI) has shown a promising safety signal in topline results from a phase I trial.

The drug discovery process is often based on trial-and-error methods that can last for years. Insilico Medicine is one of many biotech companies using AI to cut the time required to design new small molecule drugs and take them to clinical trials.

The company’s AI-designed lead candidate INS018_055 blocks an AI-discovered drug target to treat IPF. The condition involves scarring in the lungs that leads to a drop in lung function, and can result in death. While there are drugs approved to slow the development of IPF, they are unable to stop or reverse the progression of the disease, and can cause unpleasant side effects including nausea and diarrhea.

INS018_055 is intended to provide more therapeutic options to IPF patients with a lower risk of side effects than current drugs. In late 2021, Insilico launched the first exploratory clinical trial of the drug in healthy volunteers in Australia. The latest Phase 1 trial started in February 2022 in New Zealand and tested the drug’s safety, tolerability, pharmacokinetics and the impact of food intake and other medicines on its effects.

According to the trial readout, INS018_055 was generally safe and well tolerated, with no serious adverse events in the volunteers. All treatment-related side effects were mild and the subjects had recovered by the end of the study.

With these safety results in hand, the company plans to test the drug’s ability to help patients with IPF. The company is applying to the U.S. Food and Drug Administration (FDA) for approval of a Phase IIa study that could launch later this year.

“Topline data from our Phase I study of INS018_055 demonstrate the ability of our Pharma.AI platform to discover novel targets and design novel molecules with a high level of translatability to human biology,” said Alex Zhavoronkov, founder and CEO of Insilico Medicine, in a public statement. “This lead program is paving the way for a new era of drug discovery and development that utilizes next-generation AI to identify novel targets and generate novel drugs to treat diseases requiring regular peroral drug administration for the entire lifetime with very high safety requirements.”

Insilico uses neural networks in a machine learning model that crunches reams of biological, chemical and clinical trial data. The company uses a platform called PandaOmics to discover new drug targets; another AI platform called Chemistry42 to design small molecule drugs against a target; and a final platform called InClinico to predict the results of a clinical trial to improve the design of the trial. Other target indications in Insilico’s pipeline include kidney and skin fibrosis, in addition to inflammatory bowel disease, cancer and more.

Insilico’s latest phase I results are a promising start for the field of AI-powered drug discovery. Another AI drug discovery player, the U.K. firm Exscientia, made headlines in 2020 when it launched the first Phase I trial of an AI-designed drug against an established target in treating obsessive compulsive disorder. In contrast, the target of INS018_055 was itself discovered using AI.

The use of AI in discovering new drugs is capturing the imaginations of investors and industrial partners. In early 2022, for example, Insilico formed an alliance with Centogene to tackle rare diseases. At the same time, Exscientia landed a collaboration deal with Sanofi worth up to almost $5.2 billion. And in November 2022, Sanofi sealed a collaboration deal worth up to $1.2 billion with Insilico.

However, the field remains in a very early stage; the potential of AI to push a drug to the market remains to be demonstrated.

Also of Interest