Nature and AI in drug discovery: a solution to high costs

Millions of patients across the world struggle to afford necessary medications due to the high cost of prescription drugs. In fact, an astounding 33% of Americans find it difficult to afford prescription drugs due to high out-of-pocket costs.

The negative repercussions of this reality stretch far beyond the patient and impact the economy and healthcare systems. When patients forgo medications, their conditions are likely to worsen, causing them to seek emergency or urgent care and hospitalizations, which are often more expensive than preventative or routine care and further challenge already strained hospital systems. This also contributes to the combined $500-billion cost to the US healthcare system due to medication nonadherence.

Further strains on the economy occur when employees grow sicker due to their inability to afford required medications, leading to decreased worker productivity and absenteeism, which costs businesses in the US $225.8 billion annually and in the UK £100 billion annually.

The high cost of medications has a negative impact on the global economy, and solutions are required along the entire healthcare value chain to decrease prescription drug prices.

What factors contribute to the high cost of prescription drugs? While there are factors along the entire drug development and commercial value chain, we will hone in on drug development.

Two significant contributors are the high costs associated with drug development, a long and expensive endeavour, and the high rate of clinical trial failure for drugs that have advanced to human testing.

It can take over 12 years and more than $2 billion for a drug to move from preclinical testing to final approval, and drug development companies may seek to recoup these costs when a drug successfully enters the market. In addition, less than 12% of drugs that enter clinical trials achieve approval from the US Food and Drug Administration, and companies can spend up to $375 million on clinical trials per drug. Therefore, drug development companies are also driving up costs on approved drugs to make up for the capital lost on failed clinical trial programmes, resulting in higher prescription drug prices passed on to the patients and payers.

These challenges present a new mission for the drug development industry: to discover drugs that are more likely to work in patients, resulting in less money lost to failed clinical trials. While there are many factors that contribute to drug pricing apart from development costs, companies that achieve this mission may have the ability to offer cheaper drugs to patients due to their operational and capital efficiencies in the drug development process.

Perhaps ironically, there is a source of medicines that have a higher chance of success in clinical trials, but few if any of the top 25 pharmaceutical companies leverage this resource for drug development: nature.

Why nature? Humans have used nature as medicine for millennia and scientists have long turned to nature to discover new drugs. For example, many antibiotics are derived from microorganisms and one of the world’s most used over-the-counter medications, aspirin, is derived from the willow tree. Even today, over 30% of FDA-approved new molecular entities are derived directly or inspired by microbial or plant species.

A recent analysis also showed that drugs derived from or inspired by nature are more likely to survive clinical trials. Despite the obvious value of nature as source material for new medicines, nature is underutilized in modern drug discovery. Part of the reason why is that traditionally, to identify a novel molecule from nature, you had to first isolate it from every other molecule in that organism, creating a high-purity sample that could be tested using complex scientific experiments and instruments, such as Nuclear Magnetic Resonance, or NMR, to identify its composition and structure. However, as in many scientific disciplines, artificial intelligence (AI) and machine learning (ML) are turning this paradigm on its head.

Scientists in industry are using AI, ML and metabolomics (the study of metabolites or small molecules in biological samples) to overcome traditional bottlenecks that have limited progress in nature-driven drug discovery. Implementing AI/ML into the drug discovery and development process has led to new technologies with many downstream advantages. These advantages include the discovery of drugs that may be more likely to work in humans and shortened R&D timelines that may enable drugs to reach patients faster, ideally resulting in lower costs to patients due to the decreased R&D costs incurred.

At Enveda, we are a clinical-stage biotechnology company that is demonstrating the utility of AI/ML in metabolomics for accelerating the discovery and development of drugs from nature to bring drugs to patients faster. We have developed novel algorithms that predict the structures and properties of all metabolites in complex biological mixtures, leading to the identification of the molecules with the greatest therapeutic promise. By integrating AI in this process, we have maximized capital and operational efficiency and generated a large portfolio of 10 development candidates in just four years since seed financing, four times faster than the industry average. The end result of integrating AI/ML into nature-inspired drug discovery is developing exciting candidate medicines faster, cheaper and better than industry averages, so that new medicines can reach patients more quickly and at a lower cost.

Another company that is applying AI/ML to discover medicines from nature is UK-based biotechnology company Basecamp Research. Basecamp Research collects data from the natural world, in locations ranging from ice caps to hidden jungle caves, with the goal of identifying proteins with beneficial properties. Their data set seeks to comprehend and recreate the collective intelligence of nature to uncover previously unseen relationships between different protein biological structures.

As scientists across industry continue to integrate AI/ML into their research, we can overcome traditional bottlenecks that have limited the discovery of drugs from nature.

By using AI/ML to speed drug discovery from pools of molecules found in nature that are more likely to survive in clinical trials, we can lower drug pricing and lift the burden of expensive healthcare costs from millions of patients.