AI-Designed Obesity and Diabetes Drug Candidate Clears Preclinical Hurdle
Insilico Medicine's AI-generated small molecule ISM1354 shows strong oral bioavailability and a favorable safety profile in preclinical studies.
Summary
Insilico Medicine has nominated ISM1354, an AI-designed oral drug candidate targeting the glucose-dependent insulinotropic polypeptide receptor, as a potential treatment for obesity, type 2 diabetes, and related cardiometabolic conditions. Preclinical testing across four animal species showed oral bioavailability of 75–104% and plasma exposure at least 18 times higher than a clinical-stage comparator at the same dose. Safety data indicated significantly weaker liver transporter inhibition, minimal liver cell toxicity, and an estimated 45-fold safety margin in monkeys. The compound was developed using Insilico's Chemistry42 AI platform and is now advancing toward formal toxicology studies. This nomination brings Insilico's total preclinical candidates since 2021 to 34.
Detailed Summary
Insilico Medicine has announced the nomination of ISM1354, an AI-generated small molecule targeting the glucose-dependent insulinotropic polypeptide receptor (GIPR), as a preclinical candidate for obesity, type 2 diabetes, and broader cardiometabolic disease. This development is significant because cardiometabolic conditions are among the leading drivers of age-related functional decline and mortality, and new therapeutic approaches in this space could meaningfully extend healthspan.
The preclinical data package is notably strong on pharmacokinetic grounds. ISM1354 achieved oral bioavailability of 75–104% across mice, rats, dogs, and monkeys — a range rarely seen across multiple species simultaneously — and at the same dose delivered at least 18 times greater plasma exposure than a clinical-stage benchmark compound. High and consistent oral bioavailability is a major practical advantage for any chronic disease drug.
Safety signals also appear favorable. The compound showed substantially weaker inhibition of the hepatic transporter OATP1B1 compared to the benchmark, which matters because OATP1B1 inhibition can cause drug-drug interactions and liver toxicity. Minimal hepatocyte cytotoxicity was observed at concentrations up to 200 μM, versus benchmark IC50 values of roughly 25 μM in human hepatocytes — a meaningful margin. A non-GLP monkey toxicology study estimated a 45-fold safety margin.
The compound was designed using Insilico's Chemistry42 AI platform, which integrates multi-objective molecular generation, free energy perturbation calculations, drug-induced liver injury screening, and iterative design-make-test-analyze cycles. This AI-first pipeline is central to the company's claim of accelerating drug discovery timelines and reducing attrition.
ISM1354 is now advancing to formal GLP toxicology studies, a prerequisite for first-in-human trials. Investors and researchers should note this remains early-stage: all data are preclinical, and human efficacy and safety are unproven. Nonetheless, the pharmacokinetic and safety profile positions ISM1354 as a credible candidate in the competitive metabolic disease space.
Key Findings
- ISM1354 achieved 75–104% oral bioavailability across four animal species, indicating strong cross-species pharmacokinetics.
- At equivalent doses, ISM1354 produced at least 18-fold greater plasma exposure than a clinical-stage GIPR benchmark compound.
- Liver toxicity risk appears low: minimal hepatocyte cytotoxicity up to 200 μM versus benchmark IC50 of ~25 μM in human cells.
- An estimated 45-fold safety margin was observed in a non-GLP monkey toxicology study, supporting progression to formal GLP studies.
- The compound was developed using AI-driven multi-objective optimization, advancing Insilico's total preclinical nominations to 34 since 2021.
Methodology
This is a news report summarizing a preclinical candidate nomination press release from Insilico Medicine, a commercial AI drug discovery company. Evidence is based on internal preclinical data disclosed by the company, not yet published in a peer-reviewed journal. Claims should be verified against full regulatory submissions or peer-reviewed publications when available.
Study Limitations
All data are preclinical and company-disclosed; no peer-reviewed publication has been cited, making independent verification impossible at this stage. Preclinical pharmacokinetics and safety do not reliably predict human outcomes, and many promising candidates fail in clinical trials. The 45-fold safety margin derives from a non-GLP study, which carries less regulatory weight than a full GLP toxicology assessment.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.
Enter your email to subscribe:
