22 March 2011. As most researchers are painfully aware, the percentage of grant applications funded by the National Institute on Aging (NIA) has plummeted in recent years. The drop culminated in rumors last fall that the payline—the score needed for a grant to be funded—would be so low that only grants scoring in the top three percentile will get awarded in the next fiscal year (see ARF related news story). Researchers interviewed at the time warned that such a marginal payline would devastate aging research in general and Alzheimer’s disease research in particular. At a time when other institutes of the NIH were holding their own with relatively stable paylines of 12-18 percent, researchers wondered how the situation at the NIA could have gotten so bad so quickly.
The situation this spring looks slightly less abysmal due to austerity measures taken by the NIA. Institute representatives predict that grant requests of less than $500,000 will be funded at the ninth percentile level in the coming fiscal year. The projection is based on appropriations detailed in a temporary continuing resolution adopted by Congress to keep the federal government running. This stop-gap measure expires 28 March 2011. Given the current political climate of austerity, the final payline could well be lower. Only 10 years ago, the NIA was funding almost 30 percent of grant applications (Wadman, 2010).
Researchers are wondering how things became so dire. The plummeting payline prompted Arlan Richardson at the University of Texas Health and Science Center, San Antonio, to do some detective work. With the help of the NIA, Richardson analyzed NIH funding trends from 2005 to 2009. He presented his analysis at the Gerontological Society of America annual meeting last November in Boston. The bottom line is that, within the NIA, there was a huge shift in funding from small grants of less than $300,000, such as R01s, to collaborative program project grants (aka P01s) and large projects (U01s) such as human longitudinal studies and clinical trials, including some initiated by NIA staff. “I expected the large grants to have increased, though not as much as they did,” Richardson told ARF. “But what really surprised me was that the small grants, the key of our science, had decreased almost 50 percent.” As those dropped by half, grants with price tags of $3 million-plus more than tripled.
Alzforum contacted many AD researchers for this story, but few agreed to speak on the record, fearing to be seen as criticizing the funder on whom they depend most. Some researchers questioned whether the NIA realized their funding priorities were shifting in this way, or anticipated what the ramifications would be for AD research. Robin Barr, Director of NIA’s Division of Extramural Activities, told ARF that the drop in R01s reflects both foreseen and unforeseen changes. The former include a deliberate drive among the NIA staff to initiate large U01 projects in aging research. In the 2005-2009 period, three major clinical trials were funded through U01 grants. The Aspirin in Reducing Events in the Elderly (ASPREE) trial, a five-year, 19,000-subject study, looks at whether a daily aspirin does more harm or good in people older than 70. The Testosterone Trial aims to determine if one year on the hormone improves the health of men over 65 who have low plasma testosterone. It plans to recruit 800 volunteers at 12 sites across the U.S. The Lifestyle Interventions and Independence for Elders (LIFE) Study is recruiting 1,600 older adults to test if a moderate intensity exercise program can protect against major mobility disability over 2.7 years. It will also look at cognitive function and cardiovascular and pulmonary health. Other U01s were mostly infrastructural, said Barr. "What caught us a bit by surprise was the rise in applications that crossed the $500,000 boundary and the decline in small grants such as RO3s and R21s," said Barr. He noted that the NIA has no control over the size and scope of applications coming in.
The NIA has since instituted a new, two-pronged strategy to control expensive applications and ensure funding for smaller grants. First, scientists must now obtain prior approval from the relevant division before submitting any application exceeding $500,000. There will also be a cap on the total amount of funds set aside for such applications in a single year. Second, grants over $500,000 will have a stricter funding line than for those below that cutoff, essentially creating a two-tiered funding line, said Barr. "This is important," said Bruce Lamb, Cleveland Clinic, Ohio. "It will definitely help ease the pain for those who are dependent on smaller R01s."
Other new guidelines may be coming on line as well. In 2009, for example, the division of geriatrics stopped accepting any investigator-initiated applications for clinical trial funding. "We certainly have every intention to extend that rule to the Division of Neuroscience, and that is in process now," said Barr. This would apply to all clinical trials funded by the division, including those for AD, that request $2 million or more. But he cautioned that each division will have a total cap for applications requesting $500,000 or more, and if that cap provides substantial relief for the funding line, then it may not be necessary to implement the restriction on large clinical trial funding.
The NIA is also addressing what had become an inequality in the way applications were assigned percentile scores. Large P01 grants did not get percentiles based on how they compared with each other, but on how they compared to other grants that come under the purview of the NIH's Center for Scientific Review. This is the body that organizes the study sections that review most of the grant applications (including R01s). But the comparison was skewed in favor of the P01s. The explanation is complex, but, in essence, if 10 percent of P01s got the highest score possible, they would all reach the first percentile, whereas under the same circumstances, R01s would be scored at the fifth percentile in an averaging process. Compounding this, though P01s are supposed to be ranked on a scale of 1 to 9, there now seems to be a general feeling among reviewers that scoring is binary, with a 1 meaning the application will be funded and a 2 meaning it will not. This equates to many applications getting the best possible score, as happened last year. "We realized at that point that the process was unfair," said Barr. The NIA has since revised its policy on ranking P01s so that they are now more fairly compared to other grants.
For his part, Richardson believes that the ranking inequity goes even deeper, into the scoring process itself. Whereas most smaller grant applications, including R01s, are ranked by study sections that review tens or hundreds of submissions at a time, P01s are reviewed individually by ad hoc committees. "No one is looking at the totality, at how the different P01s compare to each other," he said.
Barr agreed that this could be an issue. "It is possible that a committee gives an application a higher score than if they were looking at it in context," he told ARF. In talking with ARF, some researchers questioned whether too many incentives were stacked up in favor of larger grants. Even the structure of the NIA itself may lean toward funding big-science, translational projects. Since divisions compete against each other for funding, landing a large clinical trial or cooperative project means getting a larger slice of the pie.
Lary Walker, Emory University, Atlanta, Georgia, said that the push toward translational research was appropriate, but he was worried about long-term effects on basic research. "There has to be something left to translate," he told ARF. Lamb agreed. "There is an important balance as to how much money is spent looking at disease mechanisms versus translational research versus clinical trials," he said. "The questions are, How do we reach that balance?, How should these decisions be made within individual institutes?, and What are the longer-term consequences?" Todd Golde, University of Florida, Gainesville, noted that prioritizing will be tough. "There may be some fat that can be trimmed, but getting rid of some of those large program project grants is not going to be easy," he said. He noted that some of them, such as the Alzheimer's Disease Research Centers, provide a special return on investment that smaller grants simply cannot.
Some researchers have called for a separate institute to fund Alzheimer's research—a view Golde echoed. "Is it time for a National Institute on Alzheimer's Disease and Related Dementias?" he asked. "Having AD research within the National Institute on Aging compromises both general aging research and Alzheimer's disease research, because you end up with competing interests," he said. He noted that once the word "Alzheimer's" appears in a grant, it is almost guaranteed to be funneled to the NIA for review, whereas aging-related grants can often end up at other institutes. Other researchers would not be so quick to separate aging from AD. "I think there are widely different opinions on that," said Lamb. "Historically, new institutes within the NIH have brought attention to particular areas, and there are clearly some advantages to that. But there is also something to be said for the biology of aging being critically important for all kinds of age-related diseases. If you take one out of the mix, what about all the others?"
The bottom line may be just that. "The problem is not so much how to slice the pie, but that the pie is not big enough," said Golde. He claimed there is a mismatch between the societal impact of AD and the amount of federal funds spent on AD research. The NIH spends about $3 billion a year on HIV-AIDS, whereas the total NIH expenditure on Alzheimer's, even though it affects many more people, is around $500 million (with potentially another $200 million spent on related conditions such as mild cognitive impairment). "I don’t want to single out HIV-AIDS, but we believe that AD is a relatively defined entity as well, so why aren’t we doing the same thing for that?" asked Golde. "Clearly, patient advocacy led to substantial funding at a critical juncture, paving the way for research advances and the development of therapies that proved to be quite successful, though, of course, not yet curative,” he added.
Most researchers contacted for this article seemed resigned to the fact that in the current economic and political climate, increased government funding for AD research is unlikely. "That would require extensive lobbying, and unfortunately our lobby does not seem effective," said Golde. Several bills have been sponsored in Congress to support AD research in principle, including the National Alzheimer’s Project Act (see ARF related news story), which mandates the Department of Health and Social Services to devise a national plan to tackle research and treatment of AD. However, no new appropriations have been made. Most recently, the MIND act, which would establish bonds that would fund research, was passed (see related news). And there are indications that other institutes, including the National Institute of Diabetes and Digestive and Kidney Diseases and the National Institute of Mental Health have also funded more large grants. This would reflect a more general push toward big science projects. In fact, NIA Director Richard Hodes pointed out that the latest figures from the National Institute of Arthritis and Musculoskeletal and Skin Diseases, and the National Institute of Allergy and Infectious Diseases, project paylines for R01 grants of about 8 percent (12 percent for both for new researchers). "It could turn out that the NIA was the canary in the coal mine," suggested Richardson.—Tom Fagan.