Hepion Pharmaceuticals Introduces Proprietary A.I. Platform for Drug Development

Hepion Pharmaceuticals today announced the rollout of a proprietary artificial intelligence ("AI"), machine learning platform called, AI-POWR™.

AI-POWR™ is Hepion's acronym for Artificial Intelligence - Precision Medicine; Omics that include genomics, proteomics, metabolomics, transcriptomics, and lipidomics; World database access; and Response and clinical outcomes. AI-POWR™ allows for the selection of novel drug targets, biomarkers, and appropriate patient populations. AI-POWR™ is used to identify responders from big data sources using Hepion's multi-omics approach, while modelling inputs and scenarios to increase response rates. The components of AI-POWR™ include access to publicly available databases, and in-house genomic and multi-omic big data, processed via machine learning algorithms. AI outputs allow for improved response outcomes through enhanced patient selection, biomarker selection and drug target selection. The aim of AI-POWR™ is to help to identify responders a priori and reduce the need for large sample sizes through study design enrichment.

Hepion intends to use AI-POWR™ to help identify which patients that will best respond to CRV431, its lead drug candidate for treatment of NASH patients, currently in a phase 2a clinical trial. It is anticipated that applying this proprietary platform to Hepion's drug development program will ultimately save time, resources and money. In so doing, Hepion believes that AI-POWR™ is a risk-mitigation strategy that should reap benefits all the way through from clinical trials to commercialization.

This new business line will be led by Dr. Patrick Mayo, Hepion's current Senior Vice-President, Clinical Pharmacology, who will now fill the expanded role of Senior Vice-President, Clinical Pharmacology and Analytics. Dr. Mayo has been building the technology and will continue to build a team to meet the Company's needs. Dr. Mayo's experience spans three decades of Clinical Pharmacology and AI, going back to an award received for AI in 1997 for Application of Neural Networks,1 co-authored by Dr. Brian Corrigan, currently Global Head, Clinical Pharmacology at Pfizer.

"Many years ago, Dr. Corrigan introduced me to the idea of neural nets, but it was Dr. William Baker who worked on the lunar model with NASA that emphasized the use of multivariate and big data analytics for real world problems," commented Dr. Mayo. "One significant limitation of conducting traditional, large sample size, longer-term clinical trials is the financial burden and human cost associated with them as they can take months or years to complete. Indeed, trials in NASH patients often take an excruciatingly long time and the responders comprise only a small number of patients relative to all the patients enrolled. After many months, with a large number of patients, companies are left wading through mountains of data where the delta, or effect size, between responders and placebos can be quite small. This raises questions about whether it is appropriate to conduct post-hoc sub-group analyses and also about commercial viability and future third-party reimbursement. It further raises questions about how to identify patients that may be best suited to receive a drug that is not associated with high responder rates. Time and time again, we see companies working in NASH and other therapeutic areas that complete trials only to be left trying to explain why their drugs performed sub-optimally."

Dr. Mayo Says: Our analyses confirm that NASH is a very heterogenous disease and we need to have a better understanding of interactions between changes to proteins, genes, lipids, and metabolites, to name a few, induced by both drugs and disease. All of this is further complicated by variable drug concentrations, patient traits and temporal factors. AI-POWR™ is designed to address many of these typical challenges, as we believe we can use our proprietary platform to shorten development timelines and increase the delta between placebo and treatment groups. In NASH, we have gathered literature data on 2,918 patients, generating 149 million discrete data points which computationally allows us to apply principles of Precision Medicine to this disease. Building out our clinical pharmacology group with this proprietary AI should allow for outcomes optimization in our clinical trials, and we believe ultimately give us a distinct advantage with our future commercial launch activities.

Dr. Foster, Hepion's CEO Says: Dr. Mayo has been developing AI in pharmaceutical research for close to three decades now and has been able to mine many revealing characteristics of CRV431 from our in vitro and in vivo preclinical and clinical activities. He will continue with these activities as he builds his team and utilizes AI-POWR™ to both drive our ongoing Phase 2a NASH program and identify additional potential indications for CRV431 to expand our footprint in the cyclophilin inhibition therapeutic space. From a business development perspective, this platform can also help us identify additional therapeutic products for repurposing that can complement and expand our company's pipeline beyond cyclophilin inhibition.

SOURCE: Hepion Pharmaceuticals, Inc.

Related Posts

Subscribe Our Newsletter