The traditional drug development and approval process in the United States typically spans 10 to 12 years, during which millions of Americans may succumb to diseases like incurable cancers. While the FDA has implemented programs like accelerated approval pathways and expanded access to mitigate delays, many patients still face lengthy waits for new treatments.
Artificial intelligence (AI) and machine learning (ML) are increasingly recognized for their potential to significantly shorten the drug approval process and overcome barriers that prevent promising treatments from reaching the market. In response, the FDA released a discussion paper titled “Using Artificial Intelligence and Machine Learning in the Development of Drug and Biological Products” to integrate AI and ML more systematically into drug development.
The vast and rapidly growing amount of data in healthcare, projected to reach 175 zettabytes by 2025, is crucial for drug development. AI’s ability to swiftly analyze this data, including omics datasets, can expedite the discovery of effective treatments. For instance, studies like Fumagalli et al. (2023) have shown that AI can efficiently process complex omics data, aiding in selecting therapeutic compounds.
AI and ML can also enhance the precision and personalization of medicine by analyzing diverse data sources, helping identify the most effective therapeutics for different disease stages and treatment resistance levels. They offer the potential to build models predicting drug efficacy and toxicity in patient cohorts, thus improving clinical trial success rates.
Additionally, AI and ML can reduce research and development risks, increasing the likelihood of FDA approval and market availability of novel therapeutics. For many Americans awaiting lifesaving treatments, integrating AI and ML in drug development could be crucial in improving and saving lives.
To read more, click here.
[Source: HealthData Management, December 18th, 2023]