Insilico Medicine Names AI-Driven Pain Candidate, Publishes Rare Cancer Study

Insilico Medicine, a clinical-stage generative artificial intelligence (AI)-driven drug discovery company, has nominated ISM9528 as a preclinical candidate compound (PCC) for pain management and published a study in npj Precision Oncology showing how AI can identify therapeutic targets for a rare sinonasal cancer. The company has also entered a collaborative research agreement with Memorial Sloan Kettering Cancer Center (MSK) to discover new therapeutic targets for gastroesophageal malignancies.

On July 22, 2026, Insilico announced the nomination of ISM9528 as a PCC for pain management. ISM9528 is a potential first-in-class, brain-penetrant, orally available inhibitor of Target Z that offers a novel, non-opioid treatment approach by acting on a previously unexplored biological mechanism. Identified by PandaOmics, Insilico's AI engine for target identification, ISM9528 demonstrated superior efficacy compared to the standard of care in preclinical studies, along with a favorable safety profile and promising DMPK properties. The research utilized specialized mice models that imitate human pain mechanisms, demonstrating robust, dose-dependent pain management efficacy, with potential comparable or surpassing current solutions, including morphine. This marks the 31st PCC nominated by Insilico since 2021.

The company's AI-driven pain program began with PandaOmics, which analyzed massive datasets of omics, textual and multimodal data and uncovered a new pain-related pattern, highlighting Target Z in Insilico's current portfolio. Following target validation, the team deployed Chemistry42, Insilico's generative chemistry platform, to re-generate compounds with optimized selectivity and ADMET properties for pain indication. After iterative cycles of AI generation, human feedback and refinement, ISM9528 was selected as the lead compound. The company's founder and CEO stated that seeing a non-opioid candidate outperform morphine in preclinical models is an unprecedented milestone, and that target discovery for chronic pain and migraine historically takes over a decade, but PandaOmics precisely mapped this pathway.

In the peer-reviewed study published in npj Precision Oncology, researchers performed the first comprehensive molecular characterization of inverted papilloma-associated sinonasal squamous cell carcinoma (IP-SNSCC), a rare and aggressive subset of head and neck cancers. Conducted with investigators from the University of Chicago and Johns Hopkins University, the study integrated whole-exome sequencing, RNA sequencing, mitochondrial DNA sequencing, and AI-driven target prioritization. The analysis revealed a coordinated cascade of biological changes accompanying the transition to invasive cancer, including cell-cycle regulation alterations, extracellular matrix remodeling, immune signaling disruptions, and metabolic reprogramming. Using PandaOmics' TargetID algorithms, the team prioritized clinically actionable targets with existing FDA-approved inhibitors and novel targets for future drug discovery programs.

In February 2025, Insilico and MSK entered into a collaborative research agreement aimed at discovering new therapeutic targets for gastroesophageal malignancies. MSK contributes high-quality patient genomic, proteomic, and transcriptomic data and deeply annotated clinical cohorts, which are integrated into PandaOmics. The collaboration is currently centered on data gathering, quality control, and integration, with subsequent phases anticipating AI-driven hypothesis generation, target ranking, and detailed biological investigation. The project also supports the evaluation and advancement of identified targets across a range of therapeutic modalities, including biologics and small molecules.

Insilico has also partnered with Takeda to use the Pharma.AI suite for early-stage drug discovery across multiple therapeutic areas. The collaboration gives Takeda exclusive worldwide rights to develop, manufacture and commercialize candidates identified through the program. The Pharma.AI suite includes PandaOmics for target identification, Chemistry42 for de novo molecule design and InClinico for clinical trial prediction. The agreement includes approximately US$60 million in upfront and near-term payments and could reach around US$600 million through preclinical, clinical, commercial and sales milestones, with tiered royalties.

The company is presenting at the AACR 2026 conference in San Diego, where it plans to discuss novel oncology pipelines, partnership and collaboration opportunities, and the broader role of AI in therapeutics.

Founded in 2014, Insilico Medicine uses AI across the development cycle. Its pipeline includes Rentosertib (ISM001-055), an AI-designed candidate for fibrosis that is in phase II trials as of 2025, and ISM3312, an immunomodulatory candidate for COVID-19 and other viral infections that completed phase I in 2024. Another oncology candidate, ISM3091, was admitted to patient testing. While traditional early-stage drug discovery typically requires an average of 4.5 years, Insilico has nominated 20 preclinical candidates from 2021 to 2024, with an average timeline of 12 to 18 months per program and only 60 to 200 molecules synthesized and tested per program. The company was listed on the Main Board of the Hong Kong Stock Exchange on December 30, 2025, under the stock code 3696.HK, and focuses on fibrosis, oncology, immunology, pain, and obesity and metabolic disorders.