Another milestone has been reached in Insilico Medicine’s AI-powered drug discovery pipeline – positive results from the phase 2a clinical trial testing ISM001-055, a first-in-class antifibrotic inhibitor, in patients with idiopathic pulmonary fibrosis (IPF). This is the first time an AI-designed drug for an AI-discovered disease-associated target has been tested in the clinic, definitively marking Insilico’s transformation of drug discovery and demonstrating the power of Pharma.AI, the end-to-end generative AI platform.

The traditional drug discovery process is a time-consuming, financially demanding, and risk-filled endeavor that often takes more than 10 years to complete and costs more than $2.0 billion. From its inception, Insilico Medicine’s goal has been to use advanced generative artificial intelligence (AI)-powered approaches to streamline and improve this pipeline, expediting the rate at which novel therapies are safely brought into the clinic.

One of the Company’s flagship programs has shown immense potential in treating idiopathic pulmonary fibrosis (IPF), a debilitating and age-related lung disease that lacks a curative treatment. ISM001-055 is a novel, generative AI-designed small molecule inhibitor that targets TNIK (TRAF2 and NCK-interacting kinase), a protein kinase that orchestrates several pro-fibrotic pathways that drive IPF pathology.

The generative artificial intelligence platform Pharma.AI is at the center of ISM001-055’s history. The team at Insilico chose to point its multi-pronged Pharma.AI toolset at IPF because of the enormous unmet need for patients: current approved therapies for IPF do not reverse disease progression, only slow it, with patients typically dying 2-5 years after diagnosis, and the number of IPF patients continue to grow from today’s estimate of 30,000-40,000 new cases per year in the United States as the population ages.

Fibrotic disease is a broad indication that is estimated to be the cause of death in as much as 18% of all deaths globally. Fibrosis is the result of complicated biological processes that may involve many types of cells in many types of tissues throughout the body, and the ultimate cause is not always identifiable, such as in IPF. Identifying a disease-associated molecular target is an important step in tackling this poorly understood age-related disease and would enable development of a safe and effective treatment to resolve fibrosis at the cellular level.

Insilico first identified TNIK as a priority molecular target for treating IPF in 2019 using the Biology AI module of the Pharma.AI platform. The AI module integrated huge repositories of published data – “omics” data sets like gene expression and genome profiles, publication texts, patents, grants, and more – to identify genes most likely to be responsible for IPF pathology and disease resolution when treated with a targeted therapy.

With a target in sight, the Insilico team went to work developing a small-molecule inhibitor that could interrupt TNIK’s functions. Within a year, the Chemistry AI module of the Pharma.AI platform aided our medicinal chemists and biologists in designing, optimizing, and synthesizing ISM001-055. Our group rigorously tested ISM001-055’s safety, toxicity, and efficacy, resulting in its preclinical candidate nomination in February 2021, just 18 months following its initial identification by Pharma.AI. This rapid timeline is a testament to the power of generative AI in streamlining the drug discovery process.

Following its preclinical candidate nomination, ISM001-055 was tested in multiple preclinical animal models of lung, kidney, and skin fibrosis. ISM001-055 exhibited favorable safety, favorable toxicity, and robust efficacy across all tested models, justifying in-human clinical testing beginning in 2021. After completing these investigational new drug-enabling studies, Insilico Medicine carried out a Phase 0 microdose trial and two independent Phase 1 clinical trials in New Zealand and China to translate these studies into humans. Indeed, ISM001-055 was found to be safe and well-tolerated by healthy volunteers across all 3 trials. Investigators observed no significant accumulation after 7 days, and ISM001_055 exhibited a favorable pharmacokinetic profile. In March 2024, this collective body of work from target identification to Phase 0/1 clinical trial completion was peer-reviewed and published in Nature Biotechnology as ISM001-055 progressed into Phase 2 clinical testing.

Today, we report a first look into data from one of ISM001-055’s Phase 2a clinical trials (NCT05938920). This study was a randomized, double-blind, placebo-controlled trial that enrolled 71 IPF patients across 21 sites in China. Patients were randomized to receive either placebo, 30 mg once-daily, 30 mg twice-daily, or 60 mg once-daily for 12 weeks. Patient enrollment was initiated in April 2023, and the last subject’s follow-up visit was completed in August 2024.

In this 12-week study, ISM001-055 met its primary endpoint of safety and tolerability across all dose levels. ISM001-055 demonstrated a dose-dependent improvement in forced vital capacity (FVC), the amount of air a person can exhale after a full inhalation, which is a critical lung function measure in IPF patients. While patients receiving a placebo experienced an average decrease of FVC of –62.3 mL, patients receiving 60 mg of ISM001-055 exhibited the largest improvement in FVC of +98.4 mL, indicating not only a slowdown of disease progression but an improvement in lung function. Complete Phase 2a data from this study will be presented at an upcoming medical conference and published in reputable medical journals.. A separate Phase 2a (NCT05975983) clinical trial in the U.S. is also ongoing and actively enrolling patients.

Insilico Medicine will engage regulatory authorities to discuss the design of a Phase 2b study. The company aims to explore extended treatment durations and larger patient cohorts to further investigate ISM001-055’s therapeutic potential in IPF. These exciting findings underscore the power of generative AI in bringing new therapies in areas of unmet clinical need safely and quickly to patients.