TwoXAR merges artificial intelligence, drug discovery and… clones?

This article was originally published here

Artificial intelligence (AI) is steadily reshaping healthcare from all sides, introducing technologies we wouldn’t have thought possible five or 10 years ago.

It’s happening in the clinic (see HealthTap’s Doctor A.I.), it’s happening in diagnostics (see IBM Watson), and now it’s moving into earlier-stage drug discovery with Palo Alto, California-based twoXAR.

“In the couple years that we have been around, we’ve been told hundreds of times that computers cannot do this; that biology is too complex; that this will never work,” said Andrew A. Radin, CEO of the AI-driven biopharmaceutical company. “Yet, in every single disease program where we have run proof-of-concept studies on our novel AI-identified candidates, we have generated efficacious results across standard end points.”

Using a custom-built computational platform, twoXAR works to identify what it calls “unanticipated associations between drug and disease.” With the compounds of interest in hand, the team runs a series of preclinical studies to ‘de-risk’ them. The ultimate aim is to advance the candidates into the clinic through industry and investor partnerships.

How do clones fit into this? They don’t really, but it just so happens that the two cofounders share the same relatively uncommon name, Andrew Radin. Andrew A. Radin is the CEO and Andrew M. Radin is the chief marketing officer. Together they formed twoXAR (two times Andrew Radin) in 2014 with a $3.4 million seed round led by Andreessen Horowitz’s Biofund and Stanford Start X Fund.

One of the selling points is the agnostic approach a computer can take to drug discovery. There’s no human bias, no restrictions on what disease areas can be targeted or what kind of science needs to be done. The platform can sift through both small and large molecule libraries.

It’s an interesting concept, given the high rate of failure (approximately 90 percent) in clinical drug development.

With the rise of AI, many different interpretations are coming to light. In an email forwarded by a company representative, Andrew M. Radin described what AI means to the company.

“In our case, leveraging real-world big biomedical data to build predictive algorithms that can make predictions that can be used to drive rational decision making in drug discovery,” said Andrew M. Radin. “AI is the term that best encapsulates and most succinctly describes what we do in a way that can be understood by our various audiences including investors and potential biopharma partners.”

In February, twoXAR announced a partnership with Osaka, Japan-based Santen Pharmaceutical. Under the agreement, twoXAR will use its AI platform to discover, screen, and prioritize novel drug candidates that are most likely to be able to treat ocular indications, specifically glaucoma.

An earlier project saw the platform applied to hepatocellular carcinoma (HCC or liver cancer). The company screened a library of more than 25,000 potential drug candidates, identifying the 10 top candidates for HCC. Proof-of-concept studies were then performed by The Asian Liver Center at Stanford University. 

“The objective of these experiments was to establish which of the original 10 candidates we identified might be promising liver cancer treatments and generate preliminary preclinical data,” Andrew A. Radin explained. “While we had a few promising results among the 10 candidates, TXR-311 stood out as it killed liver cancer cells with high selectivity. Specifically, very low doses of TXR-311 killed five different liver cancer cell lines. But in healthy liver cells, 500 times as much TXR-311 was needed to cause cell death. In contrast, treating healthy cells with only 3 times the dose of sorafenib [an FDA-approved therapy] needed to kill liver cancer cells is enough to kill healthy cells.”

This recipe for drug discovery hits on an emerging theme within artificial intelligence. Done well, AI doesn’t replace humans and biology, it enhances it. TwoXAR’s computation platform is only one piece of the puzzle, as the company then looks to vet and advance a ‘derisked’ drug candidate into clinical trials.

Another central theme: AI companies have to forge their own path. Computers have never reached these frontiers before.

“Being entrepreneurs, we thrive in hearing ‘no’ as an answer, but interpreting it as ‘not yet,’” Andrew A. Radin said. “We believe our biggest success to date is doing that which people have told us could not be done. Whether that’s discovering novel drug candidates with novel biology using our AI-driven platform or generating efficacious results in proof-of-concept studies or sharing IP on our discoveries with a leading biopharma company.”

Expect more biopharma partnerships in the coming years. In terms of scalability, a quick search on LinkedIn reveals at least eight more Andrew Radins if the company wants to go three or four XAR.

In terms of scalability, a quick search on LinkedIn reveals at least eight more Andrew Radins if the company wants to go three or four XAR.

Photo: ANDRZEJ WOJCICKI, Getty Images

0 replies

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply