A study published in Proceedings of the National Academy of Sciences reports preclinical data on GATC-1021, an AI-derived drug candidate for opioid use disorder, showing reduced fentanyl intake and changes in brain plasticity markers.

GATC Health and researchers from the University of California, Irvine said the compound was identified using an AI platform analysing postmortem human brain tissue from people with substance use disorder. The approach aimed to target underlying neurobiology rather than symptoms of dependence, an area where current therapies remain limited.

The paper, titled “AI-Derived Therapeutic Development of a Serotonin Receptor Targeting Drug for the Treatment of Opioid Use Disorder,” describes how the candidate modulates the 5-HT2A and 5-HT6 serotonin receptors. In preclinical models, this dual mechanism was associated with reduced fentanyl intake in both male and female subjects.

Researchers also reported increased markers linked to neuroplasticity, including dendritic spine formation and upregulation of Bdnf in the prefrontal cortex. The study noted that although the compound acts on serotonin pathways, it did not produce hallucinogenic-like effects, which are often associated with similar targets.

Christie Fowler said: “By using AI to analyze human postmortem brain tissue, we were able to design a compound that not only reduces opioid intake, but also directly addresses the underlying biological dysregulation driving addiction—something that has historically been difficult to achieve with conventional drug discovery approaches.”

Opioid use disorder remains a major public health challenge, with millions affected in the US and high rates of overdose deaths linked to synthetic opioids such as fentanyl. Existing treatments, including methadone and buprenorphine, primarily address withdrawal and dependence but can face issues with adherence and long-term outcomes.

The study positions GATC-1021 as a non-opioid approach that targets multiple pathways involved in addiction biology. This “polypharmacy” strategy, as described by the company, aims to restore balance in the brain’s reward system rather than substitute opioid activity.

Ian Jenkins said: “By identifying markers of dysregulation directly from human brain tissue, GATC Health was able to bypass traditional ‘trial and error’ discovery, significantly accelerating the path to clinical potential.”

While the findings highlight the potential of AI-led drug discovery in complex neurological disorders, the results remain limited to preclinical research. Further clinical development will be required to determine safety, efficacy and real-world applicability in patients with opioid use disorder.

The announcement contains elements of a company-led narrative, particularly around platform validation and development speed, suggesting aspects of a sales pitch. However, the peer-reviewed publication and academic collaboration provide a level of scientific grounding that supports its news value.

The research adds to a growing body of work exploring AI in drug discovery, particularly in areas where conventional approaches have struggled to deliver new treatments.