Drug discovery is an extremely long and expensive process. It can take drug companies up to three years just to screen potential drug candidates — and that’s before animal testing even begins. 90% treatments that then make it into human clinical trials fail before they reach the market.
Enter the AI. A decade ago, scientists saw an opportunity to harness data and computing power to predict which candidates were most likely to treat a target disease. They hoped this approach could shave years off the drug discovery process and increase the success rate of clinical trials.
Investors and big pharma have invested money to make it a reality. European companies applying AI to medical research have raised a total of $2 billion over the past 10 years, according to data from Dealroom. Earlier this year, French pharmaceutical giant Sanofi signed a massive deal worth up to $5.2 billion with artificial intelligence company Exscientia.
“Now there is real acceptance. In the next few years, all drugs will be discovered this way,” says Andrew Hopkins, Founder and CEO of Exscientia. UK-based Exscientia – founded in 2012 – was the first AI drug discovery company. Its $510 million IPO last year was one of the largest ever in the biotech industry.
There are 23 AI-based drug candidates in clinical trials as of August 2022, according to a recent study. However, none of them have hit the market yet. So how close are AI drug discovery companies in Europe to delivering on their promise?
Succeed in clinical trials
Before a drug can be sold on the market, it must go through a long process of preclinical testing followed by a series of human clinical trials that determine if it is safe and effective. The whole process can take 10 to 15 years and cost billions of dollars – yet most candidates fail before they reach the market.
The first drug discovered with AI to enter clinical trials was found by Exscientia through a collaboration with its Japanese pharmaceutical partner Sumitomo Dainippon. The drug began clinical trials in 2020 as a treatment for obsessive-compulsive disorder (OCD), but the study lack and its development was halted a year later.
Hopkins will not comment on these results because Exscientia is no longer involved in the programs it launched with Sumitomo. The Japanese pharmaceutical company is currently conducting a phase 1 trial for a treatment of Alzheimer’s disease with another drug identified by Exscientia.
Some pioneers like Exscientia and BenevolentAI – founded a year after Exscientia, also in the UK – already have early clinical trial results showing their drug candidates are safe in humans. But the technology has yet to prove that the drugs are also effective in treating patients.
“The jury is still out,” says Pierre Socha, a partner at venture capital firm Amadeus Capital.
The next big step will be to see late-stage clinical results for these drugs, he says. “Once we can reliably see results at scale and in a cost-effective way, we will know if AI is going to become one of the main tools for drug discovery.”
We’ll soon start seeing the first results of efficacy from clinical trials, and Socha thinks we might have an answer on whether the technology can deliver on its promise by 2026.
Beyond Drug Discovery
While awaiting trial results, Socha notes that AI is already impacting drug discovery in many other ways. The technology seeps into the entire drug development pipeline.
For example, British startup Ori Biotech is using AI to improve the efficiency of cell and gene therapy manufacturing, which is currently a huge bottleneck in the pharmaceutical supply chain. The company raised $100 million in a Series B funding round in January, backed by Amadeus and Octopus Ventures among other investors.
Artificial intelligence can also enable precision medicine by identifying the patients most likely to benefit from a specific treatment. French Unicorn Owkin is one of the leaders in this field.
Exscientia has proven in a recent clinical test that AI-assisted precision medicine can improve outcomes for cancer patients who have relapsed after at least two previous treatments. The company can examine individual cells from a sample of a patient to find the best medicine for them.
“This is the first time an AI system has improved clinical outcomes in oncology,” Hopkins told Sifted.
Can the pharmaceutical industry really adopt AI?
So far, most AI drug discovery companies are academic spinouts and startups. But for this to really have an impact, incumbent pharmaceutical companies will also need to embrace the technology.
Pharmaceutical companies have started integrating AI into their own drug discovery pipelines. However, they have been much slower to do so – none have yet started clinical trials with drugs discovered using AI.
The problem with pharmaceutical companies is that they apply AI to individual steps of existing drug development methods when it’s the whole process that needs to change, says David Brown, president of the AI startup Healx.
Brown has 50 years of experience in drug discovery and has worked for four of the top pharmaceutical companies. While at Pfizer, he was involved in the development of Viagra – the drug was originally intended to treat heart problems, but it was its side effects that made it a blockbuster.
This inspired him to found Healx. Based in Cambridge, the startup uses AI to sort out drugs that are already safe for humans and reuse them as treatments for rare diseases. Benevolent AI used a similar approach two years ago to identify a drug for rheumatoid arthritis that could also treat Covid-19.
He says one of the main reasons drug discovery fails is human bias when selecting drug targets. For example, a scientist may select a candidate simply because they have worked in that specific area before. Healx aims to remove this bias by entrusting the selection of drugs and their targets to machine learning algorithms. The startup is currently running a clinical trial treating children with fragile X syndrome, a genetic condition that causes intellectual disabilities.
Ultimately, Brown believes pharmaceutical companies need to rethink their entire processes in order to keep up with rapid technological developments in AI.
“A small business with up to 5,000 people could do what AstraZeneca [which has 76k employees] done now,” Brown says. “We need to start increasing efficiency and getting back to the growth rates of the past.”
The market for AI drug discovery continues to grow, and as big pharma signs deals worth billions in this space, new companies are joining the ranks.
One such new contender is Aqemia, a Parisian startup that recently raised €30 million in a Series A funding round led by Bpifrance and Eurazeo. Maximilien Levesque, co-founder and CEO, says his team is developing new AI that can dramatically increase the rate at which promising drug candidates are found using quantum physics.
“We can test virtually half a billion new compounds every day with the same precision as the most accurate method on the market – that is, 10,000 times faster for the same computational cost,” Levesque said. at Sifted.
Aqemia uses quantum physics algorithms to sort through millions of compounds and find those that can best interact with the desired target. While others use “brute force” approaches to test each compound, the company is able to predict interactions by solving a quantum physics equation that no one else knows how to solve, Levesque says.
The company already has a dozen internal drug discovery projects, mostly targeting cancer, and has partnered with major pharmaceutical companies such as Sanofi, Servier and Janssen. The main advantage for the pharmaceutical industry is that this technology can be used without any prior data, which is what all other AI companies need to train their machine learning algorithms. This could allow pharmaceutical companies to quickly catch up with their competitors when developing new drugs.
Other startups apply quantum computing to drug discovery. Two examples are Algorithmiq in Finland and Qubit Pharmaceuticals in France – both raised funds to develop their technology earlier this year.
While the pharmaceutical industry has been slow to embrace digitalization and other new technologies, founders and scientists say the tide is turning. And in every aspect – from test tube to trial – tomorrow’s drugs will look nothing like today’s.
“It’s not about applying a machine learning algorithm to a problem. We want to revamp the whole drug discovery process – that will be our biggest challenge,” says Hopkins.
Clara Rodríguez Fernández is Sifted’s Berlin-based deeptech correspondent. Follow her on LinkedIn.