After offering a deep dive last year into the recipe behind its lead lung fibrosis candidate, Insilico Medicine is now giving a tour of its entire molecular kitchen. The artificial-intelligence-powered drug designer is putting forward internal benchmarking data and timelines from 22 therapeutic candidates cooked up by the company’s platforms over the years.
At the same time, Insilico disclosed that it’s begun chasing new therapeutic areas, including obesity and muscle-wasting conditions as well as nonaddictive painkillers.
Insilico first announced in February 2021 that its AI programs had identified a new biological target linked to idiopathic pulmonary fibrosis (IPF), designed a novel molecule to match, and completed several preclinical studies and experiments to ready the potential therapy for its first human trials later that same year—all within the span of 18 months, and for a total price tag of about $2 million.
Then, in March 2024, the company detailed its work step-by-step in a paper published in Nature Biotechnology. To date, the small-molecule TNIK inhibitor dubbed ISM001-055 has gone on to show improvements in lung function among IPF patients in a 12-week, placebo-controlled phase 2a trial, while also demonstrating a favorable safety profile.
Now, the former Fierce 15 and Fierce 50 winner aims to set new standards in the field of drug discovery. Across the nearly two-dozen programs advanced between 2021 and 2024—spanning IPF as well as kidney fibrosis, plus inflammatory conditions and multiple cancers—the company said it has averaged about 13 months to narrow down its lists of potential molecules to a single prime candidate ripe for the next stages of development.
In one case—its early-phase immuno-oncology program being co-developed with Fosun Pharma, which targets QPCTL proteins to help the body engage so-called “cold” tumors that would otherwise evade the immune system—Insilico said it nominated a candidate in just nine months, far faster than the multiple years that traditional discovery approaches can take.
Another program, in inflammatory bowel disease, reached candidate status in 12 months. That molecule, the PHD-targeting ISM012-042, was the subject of a second Nature Biotechnology paper published last December; the company announced positive topline results from two phase 1 studies last month.
“Insilico Medicine remains steadfast in its commitment to transparency throughout the drug discovery process, recognizing the critical role it plays in advancing global innovation,” the company said in a statement. “By openly sharing benchmarks, including developmental candidate timelines and synthesis data, Insilico aims to demonstrate how shedding light on these metrics can drive efficiency across the industry.”
On average, Insilico said it synthesizes about 70 AI-designed molecules before landing on a favorite. They are winnowed down by a series of enzymatic assays to illustrate how well they can bind to their targets as well as through in vitro pharmacokinetic experiments and in vivo toxicology studies in animal models.
According to the company, its success rate for advancing programs from the developmental candidate stage to the IND-enabling stage has been 100%, save for the molecules that were voluntarily discontinued for strategic reasons.
So far, 10 of Insilico’s molecules have received IND clearances, paving the way for human trials.