Despite the widespread availability of analgesics, pain management remains one of the most important and widespread challenges in modern medicine, with more than 20% of the world’s population suffering from chronic pain, especially chronic neuropathic pain characterized by central sensitization. However, current standard treatments continue to rely heavily on opioids and NSAIDs, which are limited by insufficient efficacy, gastrointestinal toxicity, and serious risks of dependence and addiction, creating a need for novel, effective, and non-addictive brain-penetrating analgesics.
Insilico Medicine (“Insilico”, HKEX:3696), a clinical-stage generative AI-driven drug discovery company, is addressing this critical unmet need by using AI to uncover new biological mechanisms and develop innovative treatments. The company recently announced the nomination of ISM9528 as a preclinical candidate compound (PCC) for pain management. ISM9528, identified by PandaOmics, may be a first-in-class, brain-penetrating, orally available targeted Z inhibitor that provides a novel non-opioid therapeutic approach by acting on a previously uncharacterized biological mechanism. Notably, this is the 31st PCC nominated by Insilico since 2021, further demonstrating the company’s AI-driven R&D capabilities.
The outperformance of a non-opioid candidate over morphine in preclinical models is an unprecedented milestone and perhaps one of the most remarkable and unexpected achievements in our company’s history, the result of frontier AI, human ingenuity and scientific serendipity. Target discovery for highly complex human mechanisms such as chronic pain and migraines has historically taken more than a decade, but PandaOmics has precisely mapped this pathway, leading us to our 31st PCC since 2021. By understanding the biology behind it, we not only address pain management, but also move one step closer to our ultimate goal of deciphering the biology of aging through our dual-purpose strategy. ”
Dr. Alex Zaboronkov, Founder, CEO, CBO, Insilico Medicine
To find new paths to pain management, the team leveraged PandaOmics, Insilico’s AI engine for identifying and prioritizing therapeutic targets. By analyzing large datasets of omics, text, and multimodal data, PandaOmics helped the team uncover new pain-related patterns, highlighted the potential of Target Z in Insilico’s current portfolio, highlighted new targets not previously systematically linked to pain management, and focused the team’s attention on compounds it had already developed.
Subsequent bioinformatics analysis of human and animal model data showed high expression of Target Z in pain-related cell types, strengthening the team’s confidence moving forward with wet lab testing. Migrate quickly alive This study utilized a specialized mouse model to mimic human pain mechanisms and demonstrated potent, dose-dependent pain management effects that may be comparable to or superior to current solutions containing morphine.
Following target validation, the team deployed Insilico’s generative chemistry platform, Chemistry42, aiming for brain-penetrating properties and a balance between efficacy and safety. It encompasses over 40 selection models integrated with several related applications to regenerate compounds with optimized selectivity and ADMET properties for pain indications. After an iterative cycle of AI generation, human feedback, and refinement, ISM9528 was selected as the lead compound for the project.
“The designation of ISM9528 as Insilico’s 31st drug candidate validates the reproducibility and accuracy of our end-to-end AI platform,” said Dr. Feng Ren, co-CEO and CSO of Insilico Medicine. “ISM9528 has both a novel mechanism of action and a highly distinctive chemical structure, with promising brain-penetrating properties, offering excellent efficacy and a favorable safety profile. Looking forward, generative AI and automation are setting new standards, providing solid proof of concept and delivering real-world clinical benefits for patients.”
In preclinical studies, ISM9528, as an oral treatment, demonstrated superior efficacy compared to standard treatments in validated models of acute postoperative pain and neuropathic pain. Furthermore, comprehensive in vitro and alive Safety screening across multiple panels and animal models confirmed excellent safety profiles, with rapid systemic absorption and excellent blood-brain barrier permeability supporting a favorable DMPK profile.
In a spinal nerve ligation (SNL) rat model of neuropathic pain, ISM9528 restored mechanical allodynia in a dose-dependent manner. The intermediate dose produced significant analgesic effects at all time points up to 6 hours after administration, which was comparable to the standard pain treatment, pregabalin, at similar dose levels.
In an acute postoperative pain model, in addition to dose-dependent efficacy, ISM9528 demonstrated immediate efficacy over an equivalent dose of pregabalin at 0.5 hours post-dose.
“There is a tremendous unmet need for non-opioid analgesics, and the preclinical data for ISM9528 are very encouraging,” said Carol Sattler, MD, senior vice president of clinical development at Insilico Medicine. “This is what we call a ‘Eureka’ moment, the type of discovery that completely changes the landscape of how patients are treated. We are rapidly advancing this exciting program towards clinical trials.” In 2027, we aim to deliver medicines that truly transform patient care. ”
As an AI-native biotechnology company, Insilico is redefining efficiency in preclinical drug development and setting new standards for the industry through its advanced AI and automation platform. While traditional early-stage drug discovery typically takes 2.5 to 4 years, Insilico synthesizes and tests just 60 to 200 molecules per program and consistently reaches preclinical candidate (PCC) candidates in an average of just 12 to 18 months. Since 2021, the company has appointed 31 PCCs, 13 of which have received IND approval or clearance.
Insilico continues to improve the performance of its AI platform while expanding the practical application of its technology in drug discovery and life sciences research. The company leveraged its extensive experience and datasets in training platforms to extract thousands of benchmarks and integrate them into MMAI Gym. Serving as both a “trainer and benchmark” for scientific AI, the MMAI Gym allows organizations to train models for domain-specific inference while rigorously evaluating their performance on real-world tasks, advancing the path to pharmaceutical superintelligence. To date, Human Longevity and Liquid AI have collaborated with Insilico and participated as partners in MMAI Gym.
