The Washington PostDemocracy Dies in Darkness

To Ensure Medical Breakthroughs, Give ARPA-H Autonomy

ARPA-H will need to be agile and open-minded. (Photographer: Dan Kitwood/Getty Images)

President Joe Biden’s ambitious plan for a new biomedical agency that will take on breakthrough biomedical research is finally coming to life. Congress this week authorized $1 billion over three years for the Advanced Research Projects Agency for Health, or ARPA-H.

But the fledgling agency still needs a leader and a home — two critical choices that will determine whether it succeeds in the long term or quickly fails.

In making these decisions, officials should mimic what’s best about the Defense Advanced Research Projects Agency, the operation that inspired ARPA-H. And DARPA’s chief asset is its autonomy.

APRA-H is meant to be a space where intriguing, outside-of-the-box ideas for treating and curing cancer, diabetes, Alzheimer’s and other diseases can be tested and, if they don’t work, quickly abandoned in favor of new ones. The agency is meant to support projects that might take too long and be too risky to attract venture capital, but are too goal-oriented to fall under the National Institutes of Health’s basic-science mandate. Rather than explore hypotheses, ARPA-H projects are supposed to solve problems.

Now that the agency has been funded, Xavier Becerra, the secretary of Health and Human Services, has 30 days to decide whether it should be housed inside the NIH or operate as a separate unit within HHS.

To give the agency a fighting chance to realize its ambitions, Becerra should place it outside the NIH and let it operate with autonomy.

Consider DARPA’s track record in developing edgy new technologies for the military. Its work has led to, among other things, the internet, GPS navigation, self-driving cars and Moderna’s mRNA vaccines. This success is due largely to DARPA’s independence (it’s an autonomous unit within the Defense Department), its lack of bureaucracy and its ability to take on projects that might fail. “We had one person between us and the secretary of defense,” says Brad Ringeiser, a former head of DARPA’s Biological Technologies Office.

ARPA-H needs that same freedom to experiment.

The NIH might seem like a natural home, because it stands at the center of America’s biomedical ecosystem, funding research that advances understanding of nearly all diseases. That work is essential to determining how diseases manifest and developing theories about how to treat them. In some cases, it even pushes into finding treatments or vaccines.

It’s easy to imagine that some of ARPA-H’s work will overlap a bit with that of the NIH and its grantees. But NIH research isn’t known for quickly pivoting when the scientific trail veers in new directions. Consider, for example, the meager support the agency has given over the years to Alzheimer’s research that didn’t adhere to the popular hypothesis that the disease is caused by amyloid plaques in the brain.

A key difference is that the NIH typically provides multi-year grants that provide needed latitude for scientific exploration. ARPA-H, in contrast, will likely give out contracts that come with accountability. If DARPA is the guide, those contracts can be ended if an experiment fails or hits a dead end, so that the funding can be quickly redirected to the next good idea.

ARPA-H needs to foster a culture in which it’s possible to quickly test ideas that fall outside the norm. It needs to cultivate wildly promising lines of thinking that might lead to a breakthrough — or not.

The agency will need a leader who can establish that culture, and this will be a tricky hire. Some people think the job should go to a government outsider — perhaps someone with experience in industry or venture capital. But to avoid conflicts of interest, an ARPA-H leader would need to shed any links to industry. The person would also need to be able to fully shift her thinking away from financial return on investment and toward societal benefits.

The former NIH director Francis Collins, now a White House science adviser, has said the ARPA-H leader should have experience working in another government research agency. “Perhaps somebody who was once a program manager at DARPA in their biomedical division,” he suggested in an interview this week with Science.

The decision on where ARPA-H is located could affect the search for a leader. One bill in Congress specifies that the agency be located outside of the DC area. The implication is that it should be near one of the country’s biotech hubs. That could help attract top talent, but Congress should also consider the availability of affordable housing and lab space in places such as Boston and San Francisco.

Congress has so far given ARPA-H much less money than the $6.5 billion Biden had requested. But $1 billion is plenty for now, given that the agency needs time to gear up. It’s probably a year or more away from funding projects.

First, it’s essential to make sure that all its pieces fall into place in a way that maximizes its promise.  

More from other writers at Bloomberg Opinion:

• Moderna to Uncle Sam: My Vaccine, Not Yours: Timothy L. O’Brien

• Where Did $6 Trillion in Covid Funding Actually Go?: Editorial

• Medicare Must Test the New Alzheimer’s Drug: Bach and Redberg

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

Lisa Jarvis, the former executive editor of Chemical & Engineering News, writes about biotech, drug discovery and the pharmaceutical industry for Bloomberg Opinion.

More stories like this are available on

©2022 Bloomberg L.P.