The potential of large-molecule drugs and how technology investments can help tap it

During the recent KeyBanc Capital Markets (KBCM) Technology Leadership Forum in Colorado, Scott Schoenhaus, Managing Director and Equity Research Analyst for KBCM, moderated a panel exploring the impact of artificial intelligence (AI) and digital solutions on large-molecule drug discovery. Schoenhaus was joined by Jason Silvers, CFO of Generate Biomedicines, and Shawn O'Connor, CEO of Simulations Plus.

The panel kicked off with an overview of the enormous potential in large-molecule drug discovery, and how technology can help access it.

Less than a drop in the ocean

Large-molecule drugs have several advantages over small-molecule medications, including fewer side effects and improved efficacy in some cases. But a major driver behind the intense interest in biologics is that the potential applications for treating ailments are almost limitless. To illustrate this point, Jason Silvers of Generate Biomedicines offered this analogy: If the total number of possible proteins is equivalent to all the water in the Earth’s oceans, the number of amino acids occurring in nature is less than a drop of water.

“Traditional drug discovery for large molecules has involved immunizing an animal, seeing what comes out of the immune system, and then improving on that over time to get to something that ultimately becomes therapeutic,” explained Silvers. “That is a very long process, and it’s not always successful. AI and machine learning will enable us to explore this vast space that we couldn’t even scratch the surface of with that traditional approach.”

The application of advanced technology doesn’t mean drug discovery is suddenly simple, but the ability to collect, organize, and process terabytes of data should give researchers a steady stream of promising leads.

“We may only see promising results from 1 in every 100 of the molecules we model,” added Silvers. “But when we’re able to generate a million different potential sequences, that gives us 10,000 good starting points.”

Partnerships and consolidation with Big Pharma

With all this potential, one would think large, established pharmaceutical companies would be jumping on the AI and digital solutions bandwagon. But, as the panelists pointed out, the decision is not so straightforward.

“These companies have been doing business for a very long time, and it's not easy for them to shoehorn a technology angle into the idea of drug discovery. For one thing, there are thousands of protein scientists at these companies who think they can do it based on traditional techniques, so there's a lot of resistance to innovation in that area,” said Silvers. “You also need the right people and expertise to make it work. It’s hard to find someone in technology who also understands biology, and there aren't many biologists who understand technology. That intersection really requires the combination of both those skill sets.”

Shawn O’Connor of Simulations Plus previously worked in retail e-commerce, and he sees similarities between the evolution of that sector and adoption of AI and digital solutions in drug discovery.

“Adoption of e-commerce was very slow until Walmart built solutions into its supply chain,” said O’Connor. “We have yet to see our Walmart in drug development, a company that uses technology to bring a drug to market in five years instead of 10. But we’re seeing some rumblings of that.”

For example, O’Connor noted, the U.S. Food and Drug Administration (FDA) and other regulatory bodies around the world have started using modeling and simulation tools in their evaluation processes and are testing the use of AI-enabled solutions. That is accelerating adoption on the pharma side as well.

“Big Pharma has always looked for others to spend the difficult dollar on the discovery side and then acquire candidates into their program from third parties,” said O'Connor. “Down the road, you'll see pharma start to acquire assets that have been developed in a more AI environment.”

Silvers agreed: “Once the technology is truly proven, there will be a flurry of consolidation and acquisition.”

Investors should do their homework

AI and drug discovery is obviously a very hot topic for investors, and there is a lot of opportunity in this area. “You’ve seen private companies go public in this space, and Nvidia has even partnered with some of these companies,” Scott Schoenhaus added. “As these data sets have grown and the capabilities of machine learning have been augmented, there is a clear opportunity in the drug discovery phase [for investment].”

But while the combination of large-molecule pharmaceuticals and advanced technology may seem tantalizing, the panelists stressed that there’s no shortcut to success when it comes to drug discovery.

“Large pharma and biotechs take on average 10 years to develop a drug, and the failure rate is over 90% — so pharma has a lot of interest in using technologies to either shorten the time frame or narrow the funnel,” Schoenhaus said.

“Drug development is a difficult challenge; there's a reason new drugs aren't coming out on a daily basis,” said O’Connor. “It is a very difficult world, and the trial-and-error process is always going to be part of it. But [using] these tools is like the introduction of statistics in baseball — they can improve and make the process more efficient, but it is still a lengthy process and there is no magic bullet.”

While the development and adoption of tools to speed up that process seems like a slam dunk, investors should be wary of companies that rely too heavily on this technology.

“Organizations that are using machine learning to actually create therapeutic success really need multiple legs of a stool, including clinical expertise and target predicting, so that you can find the right targets and execute on those clinical trials,” added Silvers. “As an investor, if this is a space you're getting into, it's really critical that you understand what type of capabilities those companies have and understand if they have picked the right targets.”

For more information on this panel discussion, reach out to Scott Schoenhaus.

To learn more about our industry expertise, visit Key.com/technology.