Researchers have developed a new method of looking for disease targets using artificial intelligence, which then forecasts whether or not a treatment will be authorized by the FDA.
In its current condition, drug discovery is wasteful and burdened with increasing failure rates, signaling that the process is uncertain and imprecise. Modeling human diseases as networks simplifies complex multicellular processes, aids in the understanding of patterns in complex data that people are unable to detect, and thus enhances the precision in prediction.
Through thorough experimenting, researchers at the University of California San Diego School of Medicine developed an AI-assisted method that identified a first-in-class gut barrier-protective drug in IBD and predicted that the drug candidate would have success in Phase-III clinical trials. The research identified a new approach of searching for disease targets through artificial intelligence which then makes a prediction regarding the chances the drug will be approved by the FDA.
“Academic labs and pharmaceutical and biotech companies have access to unlimited amounts of ‘big data’ and better tools than ever to analyze such data. However, despite these incredible advances in technology, the success rates in drug discovery are lower today than in the 1970s,” stated the senior author of the study, Dr. Pradipta Ghosh.
He later went on to explain the results stating that, “the drug identified by the AI model not only repaired the broken barriers, but also protected them against the onslaught of pathogenic bacteria that we added to the gut model. These findings imply that the drug could work in both acute flares as well as for maintenance therapy for preventing such flares.”
The researchers employed an inflammatory bowel disease (IBD) disease model, which is a complicated, recurrent autoimmune condition marked by gut lining inflammation. IBD is a priority illness area for drug discovery since it affects people of all ages and diminishes their quality of life. It’s also a difficult condition to treat because no two patients behave the same way.
The Center for Precision Computational System Network, iNetMed’s computational arm, created an artificial intelligence methodology for the first phase, dubbed target identification.
The AI method aids in the modeling of a disease by creating a map of consecutive changes in gene expression at the outset and throughout the disease’s course. The use of mathematical accuracy to recognize and extract all conceivable underlying laws of gene expression patterns, many of which are neglected by present approaches, distinguishes this mapping from other existing models.
“In head-to-head comparisons, we demonstrated the superiority of this approach over existing methodologies to accurately predict ‘winners’ and ‘losers’ in clinical trials,” Ghosh explained.
In conclusion, despite being at the forefront of biomedical research, treatments to restore and/or protect the gut barrier integrity in IBD have yet to develop. The UC San Diego researchers solved this unmet need by developing an AI-guided drug development strategy that varies from current practice in three key ways: First, target identification and prediction modeling guided by a Boolean implication network. Second, target validation in network-rationalized animal models that most accurately mimic human disease. Third target validation in human preclinical organoid co-culture models, inspiring the concept of Phase ‘0′ trials that have the potential to personalize therapy choice. The confluence of these treatments confirms a first-in-class agent for treating IBD’s damaged gut barrier.
“Our approach could provide the predictive horsepower that will help us understand how diseases progress, assess a drug’s potential benefits and strategize how to use a combination of therapies when current treatment is failing,” stated the co-senior author of the study, Dr. Debashis Sahoo.
The study was published in, Nature Communications, on July 12, 2021.
Abstract. Using machine-learning, a path of continuum states was pinpointed, which most effectively predicted disease outcome. This path was enriched in gene-clusters that maintain the integrity of the gut epithelial barrier. We exploit this insight to prioritize one target, choose appropriate pre-clinical murine models for target validation and design patient-derived organoid models. Potential for treatment efficacy is confirmed in patient-derived organoids using multivariate analyses. This AI-assisted approach identifies a first-in-class gut barrier-protective agent in IBD and predicted Phase-III success of candidate agents.
Sahoo, D., Swanson, L., Sayed, I.M. et al. Artificial intelligence guided discovery of a barrier-protective therapy in inflammatory bowel disease. Nat Commun 12, 4246 (2021). https://doi.org/10.1038/s41467-021-24470-5.