MindRank, a clinical-stage artificial intelligence-powered drug discovery company, has announced that the first patient has been dosed in the Phase III clinical trial of MDR-001, its investigational oral small-molecule GLP-1 receptor agonist (GLP-1RA), for the treatment of obesity and type 2 diabetes in China.

The Phase III study, named MOBILE, is a multicentre, randomised, double-blind, placebo-controlled trial designed to evaluate the long-term efficacy and safety of MDR-001 in approximately 750 patients across 50 clinical centres in China. The trial is led by Professor Linong Ji of Peking University People’s Hospital.

MDR-001 is a biased and selective oral GLP-1 receptor agonist discovered and optimised using MindRank’s proprietary artificial intelligence-enabled drug discovery platform, Molecule Pro. In a previously completed Phase IIb trial, the candidate demonstrated clinically meaningful efficacy, favorable tolerability and improvements in cardiometabolic parametres, positioning it as a potential best-in-class therapy.

The molecule was identified and refined through the company’s end-to-end AI platform, which integrates computational chemistry, structural biology and advanced biological modelling. The programme advanced from project initiation to US Investigational New Drug (IND) clearance within 19 months and progressed to Phase III development in approximately 4.5 years.

According to the company, MDR-001 was developed using its PharmKG biological knowledge graph to integrate scientific literature and biological pathway insights. Advanced modelling tools, including protein dynamic simulation and iterative biological validation, were used to analyse GPCR protein–small molecule binding and signal transduction mechanisms. This enabled precise optimisation of pathway activities such as cAMP agonism and biased ß-arrestin 2 recruitment, guiding the design of a highly selective GLP-1RA molecule.

MindRank stated that its Molecule Pro platform incorporates deep learning, generative artificial intelligence, large language models and ADMET prediction modules, combined with laboratory validation, to optimise structural druggability and safety while minimising toxicity and off-target risks. The platform is designed to accelerate development timelines and improve success rates, particularly for hard-to-drug targets.

The dosing of the first patient in the Phase III MOBILE study marks a significant milestone for MindRank’s AI-driven drug development strategy, as the company seeks to advance innovative therapies for metabolic diseases affecting millions of patients.