Project IDentif.AI: Harnessing Artificial Intelligence to Rapidly Optimize Combination Therapy Development for...

Area of Science:

• Computational biology
• Artificial intelligence in medicine
• Infectious disease research

Background:

• COVID-19 (coronavirus disease 2019) emerged in 2019/2020, causing high infection and mortality rates.
• Current COVID-19 treatments involve drug repurposing, investigational therapies, and vaccines, with combination therapy being a key focus.
• Traditional combination therapy design relies on drug mechanisms and dose-finding for synergy, not global outcome optimization.

Purpose of the Study:

• To report the results of Project IDentif.AI (Identifying Infectious Disease Combination Therapy with Artificial Intelligence).
• To introduce an AI-based platform for interrogating a large drug/dose parameter space to identify optimal combination therapies.
• To demonstrate the platform's ability to rapidly identify effective treatments for infectious diseases.

Main Methods:

• Utilized an AI-based platform to analyze a 12 drug/dose parameter space.
• Applied the platform to identify optimal combination therapies for A549 lung cell infection by vesicular stomatitis virus.
• Focused on rapid identification of actionable combination therapies within three days.

Main Results:

• The AI platform rapidly identified actionable combination therapies.
• A sevenfold difference in efficacy was observed between optimally and sub-optimally dosed top-ranked combinations.
• Demonstrated the critical importance of precise drug and dose identification for treatment efficacy.

Conclusions:

• The AI platform is disease-agnostic and applicable to N-of-1 and population-wide clinical regimen design.
• This approach can identify highly efficacious and tolerable clinical regimens.
• The platform has potential for addressing COVID-19 and future pandemics, considering healthcare economics and global health policy.