And now for something completely different.
U.S. President Joe Biden has proposed the creation of the Advanced Research Projects Agency for Health (ARPA-H), a brand new organization that would sit within the National Institutes of Health (NIH). If this proposal and the requested $6.5 billion initial budget can get through Congress this year, ARPA-H would launch sometime in 2022. In Biden’s words, ARPA-H “would have a singular purpose: to develop breakthroughs to prevent, detect, and treat diseases like Alzheimer’s, diabetes, and cancer.”
What would be so different about ARPA-H? Well, it could help fill a big gap that’s existed for quite some time now. Since it began in early 2020, the Covid-19 coronavirus pandemic has been like a gigantic leaf blower, exposing many long-existing problems in our society. Many of these problems have been way bigger than the “we really need more toilet paper” scale. One of the problems has been the need to better facilitate and accelerate more innovation and breakthroughs in health.
For example, from after the original severe acute respiratory syndrome (SARS) outbreak concluded in 2003 up till the Covid-19 pandemic, researchers working on coronavirus vaccines, treatments, diagnostics, and related technologies in many ways struggled to find funding. SARS was not the disease du jour during the 2010’s. As a result anyone working in those areas may have faced the standard “that doesn’t seem very important right now” or “where is the monetary return” or “hmmm, there’s a risk that your idea may not work.”
Yet, when the pandemic slapped our society collectively in the face, suddenly people were saying where the heck are the Covid-19 vaccines, treatments, and diagnostics. Seemingly overnight, questions such as “can we pay for such research” didn’t seem to be quite as important. When it came to finding ways to combat the Covid-19 coronavirus, “risk aversion” and “incremental change” quickly went the way of sweater vests and sleeveless turtle necks, out of style. For example, the NIH quickly set up two new programs to push innovation to help bring Covid-19 tests and vaccines to the market as quickly as possible: the Rapid Acceleration of Diagnostics (RADx) and the Accelerating COVID-19 Therapeutic Interventions and Vaccines (ACTIV).
All of these changes subsequently yielded big dividends. The RADx program led to 32 new technology platforms for Covid-19 tests. And less than one year into the pandemic, there were not only vaccines against Covid-19 but also a new mRNA platform that can potentially transform how other vaccines are developed. Imagine what could happen if the same desire to shed risk aversion and push innovation could be maintained when there isn’t a national emergency. Hence, the need to find new ways of funding and stimulating biomedical and other health-related research.
“I’m very excited about this new [potential] addition to the NIH’s capabilities,” said NIH Director Francis Collins, MD, PhD, about ARPA-H. “It’s the opportunity to take on large, high risk projects quickly in an entrepreneurial way. Personally, this is a big priority, based on our experience with Covid-19. The time is right for a more aggressive approach.”
There’s a reason why this new entity would be called ARPA-H. And it’s not because the name rhymes with sciarpa, which is Italian for scarf, or varpa, a Swedish game where you throw stones. The ARPA-H name is very similar to DARPA. While DARPA may sound like a character played by Jenna Elfman in a circa 2000 sitcom, it has nothing to do with Dharma and Greg. Instead, DARPA stands for the Defense Advanced Research Projects Agency.
DARPA’s been around for over 60 years. U.S. President Dwight D. Eisenhower’s administration created DARPA on February 7, 1958, after the Soviet Union had launched Sputnik 1 in 1957. Sputnik 1 was the first artificial satellite to successfully “reach space” and orbit the Earth. The satellite’s success was a big wake-up call to Americans. The U.S. government realized that they needed to accelerate innovation for the Department of Defense (DoD) or risk falling behind the Soviets in national security and the arms race.
By many metrics, DARPA has been a rousing success, not just for national security but for many different aspects of society. You may have heard of DARPA-enabled breakthroughs such as the Internet, GPS (Global Positioning Systems), and self-driving cars. If you haven’t heard of the Internet, check it out. It’s a great way to share cat videos. Also, without the Internet, you wouldn’t have major companies like Google, Facebook, and Amazon or a website that allow you to click on a button to hear Darth Vader saying, “nooooooooooooooo.” So big business and many others have really benefited multiple times over from DARPA. By the way, DARPA played a key role in the development of messenger RNA vaccines too.
What’s made DARPA so successful? Well, DARPA is very different from the various organizations that currently fund and support biomedical research. DARPA is flat. Not flat like the Earth, but flat as in not very hierarchical. The 100 or so program managers (PMs) and office directors at DARPA have a fair amount of autonomy to make funding decisions. PMs have term limits of three to five years and therefore may not have the motivation to spend their time engrossed in politics. Instead, their incentive is to pursue bolder and riskier ideas that may have higher payoffs. They may be more likely to shoot for the moon.
“DARPA has been risk embracing,” explained Tara Schwetz, PhD, Assistant Director for Biomedical Science Initiatives for The White House Office of Science and Technology Policy (OSTP). “They have this innate sense of urgency in what they are trying to do. They are also independent and autonomous, held accountable through metric-driven approaches.” She added that DARPA supports “multiple approaches to solving the same problem and has a lot of flexibility and options. There’s a sense of normalizing failure.”
Contrast this with the funding options that are currently available for biomedical and other health-related research and development. Schwetz emphasized how biomedical and health research in general are often “too safe in thinking.” Sure, venture capitalists and other investors claim that they want bold ideas. But ultimately, they want a fairly high probability of getting a big monetary return on their investment in the near term, as in as soon as possible. The emphasis here is on the word “monetary” and not “lives saved”, “suffering alleviated”, or “all the feels.”
People and organizations including University endowments and pension funds give venture capitalists and other professional investors money to invest to see preferably double-digit returns each year and not in five, ten, or twenty years. Investors can’t go back to those who provided the funds and say, “our investments haven’t returned you much money but here are a bunch of smiles.”
Plus, imagine going on the TV show like Shark Tank with the following pitch: “this would be a really important scientific advance. It may work. It may not. If it does work, it’s not completely clear what the impact may be. But there’s a good chance that it could be eventually big.” Chances are this won’t get an “I’m in” from Mr. Wonderful. In fact, Mark Cuban did once tell a team,”Follow the green, not the dream.”
Thus, most venture capital and other such investment vehicles are more oriented towards things that are already clear products and services. They favor those that are ready to be sold and can offer near-term monetary return. They also frequently want a piece of the action, part ownership of your venture.
But real science doesn’t proceed like Tony Stark as Iron Man in The Avengers: End Game. It’s not like baking a bunch of brownies. Scientists don’t always first decide what they want the end product to be, disappear in a lab for a few days, and then emerge with the solution. Instead, before getting to concrete products and services, there may be plenty of trial and error, a fair amount of “we don’t quite know where this going,” and lots of surprises and unexpected impact. Remember the Internet started in the 1960s as a platform for government researchers to communicate and share information. They probably didn’t say back then, “let’s start this thing that will allow people to start all sorts of businesses such as selling goggles for dogs, having people swipe right and left on dating profiles, and collecting everyone’s data to sell for a lot of money.”
Similarly, when large corporations such as pharmaceutical companies fund research, there typically either has to be clear short-term monetary gain or at least alignment with and advancement of the company’s core business. A company that sells teddy bears is probably not going to fund research on stopping mucormycosis unless it is planning a new “fungal bear” line of products. Corporations are also more likely to support work that has a clear near-term purpose and monetary return. Ultimately, corporations are beholden to their shareholders, who in turn usually want to see the value of the company’s stock increase.
Additionally, many corporations may want to own what they fund and thus control how any of the resulting technologies are used. This may end up limiting the potential impact of breakthroughs. In fact, it’s not unusual for companies to acquire technology that may be potential competition with the intention of simply squirreling it away or even sinking it. Again, this is not a Seinfeld “you very, very bad” finger wag at companies. It is simply acknowledging the incentive structures that companies work under.
Philanthropies such as the Bill and Melinda Gates Foundation, Bloomberg Philanthropies, and the Robert Wood Johnson Foundation are able to step a bit away from the shackles of near-term monetary return to varying degrees. Indeed, they have been able to fund certain “bold” health initiatives that others haven’t such as eliminating polio and improving road safety. But such philanthropies by necessity have to be somewhat limited in their scope. They can’t just say, “we’ll fund anything interesting,” or “here’s some money, let’s see what can happen in the future.” That’s because their available funding may be more limited and their metrics tend to be noticeable positive change in particular chosen areas. Donors frequently want to see a particular problem solved or at least clearly quantifiable progress towards the solutions fairly soon.
Then there are existing governmental agencies such as the Centers for Disease Control and Prevention (CDC), the National Science Foundation (NSF), and NIH. Since it’s charged with protecting the country’s health, CDC tends to focus more on immediately practical here and now research. Going to CDC with a bold new idea that may have unexpected impact down the road can be like calling 9-1-1 and saying, “I may have a problem in several years so I would like some help now.” As the Zika outbreak, the obesity epidemic, the opioid epidemic, and the Covid-19 pandemic have shown, CDC in many cases doesn’t even have enough money to deal with today’s problems, let alone what may happen in the future. Similarly, the NSF can’t just focus on health-related breakthroughs. It has a fairly broad mission, tackling all sorts of science. WhileoOther government agencies such as the U.S. Department of Agriculture (USDA), the Environmental Protection Agency (EPA), and National Aeronautics and Space Administration (NASA) do have crossovers with health, they have other responsibilities as well.
That leaves NIH to be the government’s primary driver of fundamental biomedical and health research. Fundamental research is research that tries to unearth and better elucidate the mechanisms behind health and disease. Such research can address scientific questions and problems that may eventually lead to an even set of innovations. NIH funding certainly has led to many discoveries and breakthroughs as well as substantial contributions to the U.S. economy, as detailed by an NIH web page. For example, the NIH’s Human Genome Project (HGP) has yielded a 178-fold return on investment or nearly $1 trillion of economic growth. While biomedical and other health research is conducted in many settings across the U.S., ranging from universities to nonprofit organizations to businesses to government labs, a predominant chunk of that money comes from the NIH.
The success of the NIH hasn’t meant that there aren’t holes in the system. Individual researchers have to spend considerable amounts of time writing extensive applications to the NIH to request money to support both their organizations (e.g., professors bring resources to their universities via both direct and indirect funds) and their research endeavors. The cycle for the review of such applications is usually at least eight months, and the vast majority of such applications are unsuccessful. Furthermore, the NIH has not experienced a really big jump in its budget since the 1990’s even though conducting research has gotten more and more expensive and complex. As a result, less than 10% of applications to many of the NIH institutes end up getting funded. And even when a proposal gets funded, the original budget can get cut anywhere along the course of the project.
Budget limitations and the resulting intense competition can mean more risk aversion when choosing which projects to fund. Say you were on a trip and lost all your possessions except for maybe $20. You probably wouldn’t say, “let’s get bold and buy a inflatable sloth float to see what happens.” No, your focus would be on purchasing very practical items that would assure some kind of return.
Another challenge is the peer review process. In theory, peer review, which is having other scientists external to the NIH evaluate and score the proposals, can help ensure that a proposal makes scientific sense and provides value. However, when it comes to more unusual and innovative proposals, peer review can be fraught with land mines. The outcome of the review can depend heavily on which specific peers are assessing your application. Scientists can vary substantially in their willingness to entertain new ideas and take risks. Some peer reviewers may be stuck in particular ways of thinking. So if your proposal doesn’t resonate with even one of the peer reviewers, it could sink like a pair of cinder block water skis.
Then there are the politics. Peer review can become a bit like the set of the movie Mean Girls, or perhaps more often Mean Boys. The scientific community is full of cliques and social clubs, especially among those who had the same advisors or are in the same narrow field or disclipline. There’s the risk of reviewers favoring applications from people whom they know and like. Those from more diverse backgrounds, who corss disciplines, or who don’t have the “proper” connections may run into walls during peer review too. Ultimately traditional peer review may not always bring fresh new ideas and new people to the table.
“ARPA-H would be echoing many aspects of the DARPA model,” related Schwetz. “We learned a lot from Covid. The question has been how can apply we apply to some of the tenets that DARPA holds dear.”
Like DARPA, ARPA-H would not put proposals through such lengthy peer-review processes. Instead, the PMs would have considerable authority to make funding decisions. The right PMs, ones who are more willing to chance failure when seeking bigger wins, could then choose projects that would otherwise be deemed too high risk, too costly, too long-term, too applied, too complex, too broad, or too little “show me the money” by other existing funding mechansims. According to Collins, ARPA-H would be primarily for a “mix of start-ups, small businesses, and academics.” When I asked Schwetz how long typical funding will last for a given project, she indicated, “depending on the questions, it may range from one year to five, six, or seven years.”
Of course, a photocopy of DARPA would not be the right solution for biomedical and health research. That would be like creating *NSYNC when The Backstreet Boys are already around. Moreover, DARPA served one and only one customer, the DoD, focusing specifically on national security. Advances like the Internet and mRNA vaccines happened to be side benefits. By contrast, there is no single customer when it comes to health. Health spans so many different sectors and involves a really complex system of product developers, regulatory agencies, payers like insurance companies, patients, hospitals, physicians, and other healthcare professionals. “While the customer base of DARPA is the DoD, the customer base for ARPA-H would be the full American population,” Schwetz emphasized. “It will be working in an environment where patients will have to use what’s developed.”
Collins emphasized that ARPA-H will have a fairly broad focus. All sorts of science and products can fall under its umbrella, ranging from diagnostics to medications to computer approaches such as artificial intelligence, affectionately known as AI. There also will be attention to and an emphasis on improving health equity and reducing health disparities.
As mentioned earlier, ARPA-H will not be a completely stand-alone organization. As Collins described, “it would be within NIH but a new Division.” Being within NIH will help ARPA-H interact with the various parts of NIH. “This will offer a real opportunity to ensure strong connections,” said Schwetz. “We have to make sure that it can get the full complement of synergies. It will have to have strong partnerships across the whole federal government.” However, as mentioned earlier, ARPA-H will still be quite different from the rest of NIH when it comes to its operations, culture, and personnel.
The success of ARPA-H will depend heavily on recruiting the right people. “It will be led by a Director with a lot of authority and autonomy,” said Collins. A commentary in Science by Collins, Schwetz, NIH Deputy Director Lawrence A. Tabak, PhD, and Director of the White House’s Office of Science and Technology Policy (OSTP) Eric S. Lander, PhD, mentioned that the ARPA-H Director would serve “a single term of 5 years, with the possibility of a single extension in rare cases.” Schwetz spoke of “Trying to attract people who are really enthusiastic thinkers. People who think outside the box, who can developed a pathway for trying to address a problem.” In fact, according to Schwetz, part of the recruitment process may be seeing what the potential PMs could pitch to ARPA-H in terms of project areas and initiatives.
Of course, proposing something doesn’t mean that Congress will pass it. Getting Congress to agree can be like getting a bunch of marmots to put on an episode of Dancing with the Stars. Nevertheless, it is hard to argue against the need for more innovation in health. If ARPA-H turns out to be anything like DARPA, many, many segments of society would benefit. “We need to be sure that Congress is going to approve this,” said Collins. “We have sent fact sheets to the Hill and have gotten positive response.” Schwetz indicated that “From what we are hearing, this will be well received. We are optimistic that this can get this started by fiscal year 2022.”
Coincidentally, there is an “aha” in the name ARPA-H. Could ARPA-H indeed be an “aha” moment for health-related research and development in the U.S.? The scientific talent in the U.S. has the potential to do a lot more innovation but may now be lacking the funding to take more risks and be bolder. “This is one of those moments where you can see how the landscape has evolved,” said Collins. “This would be a way of capturing that potential without having things linger for any longer. I’ve been at NIH for 28 years. This is exactly the right time for this to happen.”