AI-Driven Drug Design Platform Achieves 70% Success Rate in RORγT Inhibitor Discovery
核心洞察
Simulations Plus (搜索) and the Institute of Medical Biology of the Polish Academy of Sciences (搜索) achieved a 70% success rate in designing active compounds using AI-driven drug design technology.
The collaboration successfully developed novel RORγ/RORγT (搜索) ligands targeting inflammation (搜索) and immune responses, with the lead compound showing potent inverse agonist activity.
Results published in ACS Medical Chemistry Letters demonstrate the potential of AI-driven multi-parameter optimization to accelerate drug discovery for inflammatory diseases.
Simulations Plus (搜索), Inc. and the Institute of Medical Biology of the Polish Academy of Sciences (搜索) (IMB PAS) have achieved a remarkable 70% success rate in designing biologically active compounds using artificial intelligence-driven drug design technology, according to results published in the American Chemical Society (ACS) Medical Chemistry Letters.
The collaboration, launched in 2023, utilized the AIDD module in ADMET Predictor (搜索)® to design novel RORγ/RORγT (搜索) ligands—molecules that impact gene expression related to inflammation (搜索) and immune responses. Within just three months, the research teams developed predictive models, designed optimized compounds, synthesized them, and completed initial testing.
Breakthrough Results in Target Validation
Among 27 compounds tested, 70% demonstrated significant inhibition of RORγT activity, with potency levels that matched or exceeded predictions from ADMET Predictor (搜索). The lead compound exhibited potent inverse agonist activity and featured a novel indolizine scaffold (搜索) not previously reported for this target.
"Importantly, this compound displayed strong efficacy in cellular assays, no significant cytotoxicity, and effectively suppressed the expression of proinflammatory Th17 cytokines (搜索) in human T cells," said Rafal A. Bachorz, Senior Principal Applied Scientist at Simulations Plus (搜索) and lead author of the publication.
AI-Driven Multi-Parameter Optimization
The research demonstrates the power of simultaneously optimizing multiple drug properties including potency, in vivo absorption, synthesizability, and ADMET risk. In vitro ADMET profiling of the most potent compound confirmed favorable drug-like properties as predicted by the AI platform.
"These findings highlight the power of AI-driven, multi-parameter optimization in accelerating drug discovery and underscore the potential of our approach to deliver innovative therapies for patients across the globe," Bachorz noted.
Platform Validation and Future Applications
The successful validation of ADMET Predictor (搜索) models represents a significant milestone for AI-driven drug discovery. Viera Lukacova, Chief Scientific Officer at Simulations Plus (搜索), emphasized the platform's potential to provide clients with a competitive advantage through artificial intelligence and machine learning capabilities.
"ADMET Predictor (搜索) and the AIDD module provide our clients with a first-to-invent advantage by harnessing artificial intelligence and machine learning (AI/ML) to design and optimize compounds for specific targets," Lukacova said.
Therapeutic Implications
The research focuses on RORγ/RORγT (搜索) receptors and their potential role in cancer (搜索) progression and inflammatory diseases. The successful suppression of proinflammatory Th17 cytokines (搜索) in human T cells suggests promising therapeutic applications for conditions involving immune system dysregulation.
The partnership plans to extend collaboration through further rounds of scaffold optimization based on these promising initial results, potentially advancing the development of novel anti-inflammatory therapeutics.