The AI drug developer nominated ISM1354 for obesity, and says early lab and animal data point to better exposure and a wider safety margin.
At the same dose, one of Insilico Medicine’s new molecules delivered at least 18 times more drug into the bloodstream than a clinical-stage comparison compound. That’s the headline number behind ISM1354, a small-molecule preclinical candidate the Hong Kong-listed AI drug developer (HKEX: 3696) has just nominated for obesity and related conditions such as type 2 diabetes [1].
ISM1354 goes after GIPR, the receptor for a hormone called GIP. Insilico’s reasoning is that pairing a GIPR antagonist, a drug that blocks the receptor, with GLP-1 therapies could help people lose fat while keeping muscle, a combination the industry has proposed. The company flags a catch, though. Several GIPR candidates have recently stumbled in clinical trials because of toxicity and tolerability problems, so the real test is whether a molecule can be both effective and safe.
This is a different direction from some of the field’s biggest names. Lilly’s retatrutide, for instance, is described as an agonist of GIP and other receptors, meaning it switches them on. Insilico is betting on blocking this one instead.
In January, Insilico nominated an earlier GIPR antagonist, ISM0676, which produced up to 31.3% weight loss when combined with semaglutide in humanized mice. The company says ISM1354 keeps that kind of weight-loss efficacy while improving how the body handles the drug and how safe it looks.
Starting with absorption, ISM1354 showed consistent pharmacokinetics (how a drug moves through the body) in mice, rats, dogs and monkeys, with oral bioavailability of 75% to 104%. Put simply, most of a swallowed dose appears to make it into the bloodstream. Next to the benchmark, it produced at least 18-fold greater plasma exposure at the same dose.
Safety is where the company leans in hardest. In lab tests, it showed substantially weaker inhibition of a liver transporter called OATP1B1, a marker of transporter-related safety. In human liver cells, ISM1354 had minimal effects on cell viability at concentrations up to 200 μM, while the comparison compound showed toxicity at about 25 μM in human liver cells and about 20 μM in monkey liver cells. In a non-GLP monkey toxicology study (an early-stage version of formal safety testing), it was generally well tolerated with no significant toxicity findings, and an exposure-based analysis suggested a margin of safety of around 45-fold. Insilico says that supports moving into formal GLP toxicology studies.
The comparison compound isn’t named, and all of this is lab and animal work run by the company. No human has taken ISM1354 yet.
Speed, AI and a changing balance sheet
Co-CEO and CSO Feng Ren, PhD, said the team “successfully overcame these bottlenecks in a very short time.” The process leaned on Chemistry42, Insilico’s generative chemistry platform, along with a liver injury prediction model to filter out risky candidates early, and several rounds of design, make, test and analyze cycles.
The company says it has nominated 10 preclinical candidates in the first nine months of 2026, a record for a year, bringing the total since 2021 to 34. It also reported revenue of about $106 million for the first half of 2026, up 287%, with adjusted net profit above $51 million, its first profitable half-year since listing. Partners named include Eli Lilly, Servier and Takeda. It cites roughly $7.3 billion in disclosed deal value for 2026 collaborations. That figure is the headline value of agreements, which is not the same as cash received.
Beyond obesity, the release leans on the company’s wider record. Its lead program, rentosertib (ISM001-055), is described as the world’s first potential first-in-class drug candidate enabled by generative AI, and has entered Phase III trials for idiopathic pulmonary fibrosis. Insilico says it had reached 8 clinical milestones across its own and partnered pipelines by mid-August. For ISM1354 itself, it sees room to expand beyond obesity and diabetes into cardiovascular disease, including obesity-associated heart failure.
The longevity argument
Insilico’s founder and co-CEO Dr Alex Zhavoronkov is direct about where he thinks this is heading. “I believe that GLP-1 therapies are poised to become the first ‘longevity drugs’ that can be popularized on a large scale,” he said, adding, “However, we absolutely cannot trade the decline in body weight at the expense of physical vitality and muscle mass, and this has been a long-standing pain point in the industry.”
The muscle point is the one to hold on to. Losing muscle as we age is a real concern, and the release itself notes the rising risk of sarcopenia in aging populations. Still, nothing reported for ISM1354 measures muscle, so that benefit is an aspiration. If the compound clears formal safety studies and reaches people, the muscle question is where its story will be won or lost.
Image courtesy of Insilico Medicine
[1] https://insilico.com/news/prfc6c0d56cc3e2211e1d0-ism1354-gipr-antagonist
