There’s at least one proposal left over from the Trump administration that President Biden is set on reviving: the creation of the Advanced Research Projects Agency for Health (ARPA-H).

Under the administration’s debut budget proposal, the National Institutes of Health would receive $6.5 billion to launch an agency modeled after the military’s Defense Advanced Research Projects Agency (DARPA). ARPA-H would accelerate the development of medical treatments for Alzheimer’s disease, cancer, cardiovascular disease and more.

For Biden, whose son Beau died of brain cancer in 2015, the potential creation of the agency is personal. Biden and his wife, Jill, founded the Biden Cancer Initiative in 2017 to continue the Obama administration’s “cancer moonshot,” which Biden led as vice president. The nonprofit indefinitely suspended operations during Biden’s presidential campaign, but his attention to the issue has continued into the presidency.

“Folks, diseases like Alzheimer’s, diabetes, cancer — they’re all on the cusp of being able to be dealt with,” Biden said in a speech last month where he touted the ARPA-H. “You know, if we don’t do something about Alzheimer’s in America, every single, solitary hospital bed that exists in America — as the nurses can tell you — every single one will be occupied in the next 15 years with an Alzheimer’s patient — every one — costing us in excess of a trillion dollars.”

But there’s a battle brewing over where exactly the agency should be housed — and how it should be structured to have the most impact. Some medical experts and lawmakers say that for the agency to successfully innovate, it should be a stand-alone entity within the Department of Health and Human Services, and free of what many experts view as NIH’s bureaucratic and time-consuming approach to innovation and research.

“By and large, it’s a conservative type of evaluation of innovation,” David Walt, a chemical biologist at Harvard University and former chair of DARPA’s advisory council, said of NIH practices. “It’s a process that requires a lot of consensus from people with very different backgrounds, and there’s regression toward the mean with respect to things that are incredibly innovative — there’s almost always a naysayer in the room.”

The White House and NIH Director Francis Collins, however, are making a full-court press to house the entity inside NIH and are lobbying lawmakers to get behind their approach. Collins and Eric Lander, Biden’s new director of the White House Office of Science and Technology Policy, published an abstract in Science Magazine on Tuesday morning outlining how a “DARPA-like culture at NIH can drive biomedical and health advances.”

Although Collins and Lander acknowledge the proposal “is radically different from NIH’s standard mechanisms of operation and will require a new way of thinking,” they argue that ARPA-H should still be housed in NIH because of a common mission and “the vast range of biomedical and health knowledge, expertise, and activities at NIH.”

“I think it would be a big mistake to put this outside of NIH,” Collins told The Washington Post in an interview. “You’d immediately create all manner of administrative duplications. There would be potentially a sense of competition about who is doing which part of which project, [and it would] lose the synergy with [an] already deep bench of scientific capabilities in NIH.”

The proposal has already garnered bipartisan support on Capitol Hill, but there is some uncertainty about the proposed agency’s location. Reps. Diana DeGette (D-Colo.) and Fred Upton (R-Mich.) on Tuesday announced a draft provision to create ARPA-H in their “Cures 2.0” bill and have been working closely with the White House.

DeGette and Upton’s legislation, which broadly seeks to modernize health-care access and fund medical research, mirrors the White House proposal for the new biomedical research agency to “revolutionize how we prevent, treat, or cure a range of diseases.” But DeGette said in an interview they have yet to decide on the structure and agency’s home.

“Congressman Upton and I will take comments from everywhere and then we’ll decide where the best place for [ARPA-H] to be housed is. But the administration has a lot of say since it’s their initiative,” DeGette said. “But the key isn’t where [the agency] is housed — it has to do with the independence and nimbleness it has.”

DeGette and Upton are meeting with patient advocacy groups that have quietly expressed some resistance to the administration’s idea of an NIH-based agency. Collins, however, downplayed DeGette’s openness to housing the agency elsewhere and claimed DeGette and Upton are “already very much leaning” in the direction of NIH.

“It will be pretty difficult to make the counterproposal at this point, that placing it somewhere else is going to succeed — it would take a very long time to even get off the ground,” Collins added.

Walt, however, surmised that once approved by Congress, a stand-alone agency could get “up and running pretty quickly.” The former chair of DARPA’s advisory council also expressed skepticism that NIH could successfully replicate the DARPA model, in which projects are not subject to peer review and program managers directly make funding decisions based on performance.

Liz Feld, the president of the Suzanne Wright Foundation, which focuses on pancreatic cancer research advocacy, is a vocal supporter of ARPA-H’s creation as a stand-alone agency. Feld was part of the group that developed and pitched the original proposal to then-President Donald Trump in 2017 and warns against doing “business as usual” with NIH.

Referring to Sen. Edward M. Kennedy (D-Mass.), Sen. John McCain (R-Ariz.) and Rep. John Lewis (D-Ga.), Feld said, “The three of them have died from pancreatic cancer or glioblastoma, and we haven’t made one advancement in early detection? These were their colleagues.”

The Health Advanced Research Projects Agency, or HARPA, was initially pitched to the Trump administration to improve the mortality rate of pancreatic cancer. There was also an additional proposal to develop a way to identify early signs of changes in people with mental illness that could lead to violent behavior.

Collins was previously against the creation of HARPA under Trump, according to people familiar with the discussions at the time. The NIH director said the previous proposal had a “very heavy focus on pancreatic cancer, as opposed to a much broader set of possible targets” and “didn’t seem to have been very well thought through.” He said NIH’s handling of the coronavirus pandemic changed his mind.

“I think the covid-19 experience that we’ve had at NIH gave me greater confidence that this approach really can work for biomedical research,” Collins added.

Lander said it’s ultimately Biden’s choice whether to base ARPA-H inside NIH.

“This project is the president’s project,” he said. “He’s been interested in this since the campaign and he has a deep understanding of the biomedical ecosystem. … This isn’t really my choice or Francis Collins’s choice. This is the president’s choice.”

But a House Republican aide working on the issue said that where the agency lives is “100 percent Congress’s call at the end of the day,” adding there’s a lot of “excitement” from Republicans about the prospects of a stand-alone agency to catalyze disease-specific work and the private sector. GOP leadership has expressed interest in the agency, too: House Minority Leader Kevin McCarthy (R-Calif.) has been briefed on its creation, and his office has been tracking its progress.

“Congressional Republicans are excited about the prospects of an independent agency outside of the bureaucratic structure of NIH that could engender the type of innovation that a bipartisan group of members of Congress would certainly support,” the aide said.