Phaseout of animal testing offers moment of truth for "organs-on-chips"

• In one instance, an "organ-on-a-chip" was shown to save the industry billions annually.

Driving the news: Last week, the FDA announced plans to phase out animal testing requirements for antibody therapies and other drugs and said companies that use other methods may receive streamlined product reviews.

• It comes more than two years since Congress removed a mandate that companies conduct animal testing before human clinical trials.
• However, FDA never updated regulations to reflect the change and left outdated rules in place, often calling for companies to provide animal testing anyway, industry experts say.
• The new approach is one of the first initiatives launched by FDA commissioner Marty Makary. Bipartisan legislation recently introduced in Congress would also codify the change.

Between the lines: Alternatives to animal testing include methods such as organ-on-a-chip technology, AI models and human cell-based assays.

• Companies that might have previously used rabbits to test if a drug irritated the skin could instead test it on human skin cells grown on a microchip.
• The National Institutes of Health and the European Commission are funding more research projects to minimize animal use in the drug approval process.

A 2022 study of an organ-on-a-chip used to test for drug-induced liver injury found it could improve safety and efficacy testing and generate more than $3 billion annually for the pharmaceutical industry.

• But many companies still opt to use animal-testing models to cover their bets, said Donald Ingber, founding director of the Wyss Institute for Biologically Inspired Engineering at Harvard University.
• This announcement by FDA could ultimately not just make it clear companies can use alternative testing, but incentivize it.
• "The fact the FDA put out this announcement that 'We'll accelerate your drug review' gives a pull we haven't had before," Ingber told Axios.

Yes, but: If the ultimate goal is to predict drug efficacy using AI models, more data is needed, especially in areas like oncology and neurology, experts said.

• "For some organs we have more knowledge than for others," said Sergiu Pasca, a neurobiologist at Stanford University. "We just don't have enough information about how most of the [brain's] circuits are functioning, how some of the genes are expressed, where they're expressed, and how they're expressed dynamically in diseases."
• Pasca and others have developed organoids — 3D clumps of cells that begin in the lab as stem cells and are then are chemically coaxed into becoming neurons, cardiac cells and other tissues.
• The solution, at least in the near-term, may be a combination of both animal and non-animal methods, he says.

AI models for drug development also face familiar challenges.

• They are "black-box models that are often poorly understood, despite labels such as 'explainable.'" Reinhard Laubenbacher, professor of systems medicine at the University of Florida, told Axios in an email.
• "And in many cases in medicine, we are still in a data-poor
  situation, an additional challenge for reliability," he said, adding that overall he's upbeat about long-term prospects.
• He said "a major research program to study the issue, develop
  representative test cases, and develop standards, can bear fruit in a short period of time," adding that the Advanced Research Projects Agency for Health could be very helpful.

Bottom line: Many of these new approaches still are at a point where dynamic processes like disease progression can't be adequately studied, Naomi Charalambakis, director of communications and science policy for Americans for Medical Progress, told Axios.

• "We want to be able to get to a day where we don't need to rely on animal models. When that day will be, we're not sure because they provide such a valuable set of information because they are a whole living body system," Charalambakis said.

What to watch: The FDA plans to launch a pilot program allowing select monoclonal antibody developers to use a primarily non-animal-based testing strategy.

• On the science side, Pasca says the new methods need to be benchmarked.
• "It is a great opportunity to push the boundary of what is possible ... [and] to take up this challenge, and really try to do it, rather than let it happen naturally," he says.