NIH’s Roadmap Plots New Course For Biomedical Science

By Potter Wickware
BioWorld Perspectives Contributing Writer

Elias Zerhouni, director of the NIH, must have felt like an overfed guest turning away from a banquet table by the end of his first year in the job. In 2003 the budget of the world's largest funder of biomedical research was $22 billion and in a five-year doubling mode. The number of research grants was up by nearly 40 percent since 1998, to more than 46,000 a year. To be sure, the period had been marked by clinical and basic research advances of the first order — neuron regeneration, stem cells, steadily falling cancer rates, completed genomes for Drosophila, C. elegans and the Anopheles mosquito (the vector of malaria). But wasn't it time to take a break and digest some of the accomplishments?

Not only was it time to assimilate, said NIH leadership, but new ways of judging and funding research proposals, organizing the scientific enterprise, and training the next generation of scientists were needed. Guided by the agency's Roadmap for Medical Research, a new culture is being dispersed across the biomedical enterprise. New centers to implement changes have already been funded at 12 academic medical sites, with a full component of 60 anticipated by 2012. At $2.1 billion over six years, funding for the roadmap is a trifle more than 1 percent of the agency's total budget, but expectations are for it to cast a long shadow, steering the NIH toward a more pragmatic, results-oriented approach to biomedicine.

As Zerhouni stated in a paper in The New England Journal of Medicine, the problem was that in an era of basic discovery the most important stakeholder — the patient — was being overlooked. Now the neglected "bench to bedside" sector is being reinvigorated with attention to "translational research," as clinical research is now often called. (A bit misleadingly, the name suggests it might have something to do with protein expression, which it does not.)

"Basic science has given us rapid protein identification, sophisticated ways of identifying carbohydrate entities, affordable genotyping," Tony Hayward, NIH's chief of clinical research, told BioWorld Perspectives. "But what we're really looking for is to advance basic discoveries into clinical research. This requires people to gather together the patients, organize the studies, accumulate the data, do the biostatistics and decide which treatment arms work the best."

Increasing Interest With Few Incentives

These are worthy goals for a field that has been in a slump for a generation. Clinical researchers today number only 2 percent of total researchers, down from 6 percent 25 years ago, Zerhouni said in an interview with the NEJM. Why the loss in appeal, and what can be done to reverse the trend?

Bert Shapiro, NIH's director of medical training, explains: "It's harder to get an MD today. It's expensive and med school graduates carry a lot of debt. The incentives to be a clinical researcher are fewer, especially if you're on soft money, where you're taking a lot of risk. A physician seeing patients earns more money, has a more certain income stream, has patient gratitude and status in the community. But in science you may spend months on dead end."

To attract new people into clinical research the NIH proposes to breathe new life into a degree program that has been around for more than 30 years, the MD/PhD. The student's way is paid through both medical and graduate school, and since the training is integrated it doesn't take as long as getting two degrees separately. Shapiro said they have a very high rate of becoming scientists. "More than 80 percent spend most of their time on research and 50 percent see patients regularly. Many become clinical researchers, although a needs gap remains."

Entrenched patterns are certain to present resistance to the changes. There's the inertia of the existing system for crediting publications in impact-factor-ranked journals, said Mike McCune, who leads a new translational research center at the University of California, San Francisco. "The present system rewards individuality, not the collaborative, team-based science that NIH envisions for the future. In multiple authorship papers the author who is not first author or PI is apt to get lost in the shuffle, and this has to change."

Then there are clinical trials bottlenecks caused by things like non-meshing institutional review board guidelines, which determine eligibility of patients and how experiments are conducted. If guidelines from different centers are not in accord, a multi-center trial could be paralyzed. The roadmap has created steering committees aimed at standardizing procedures so different groups can recognize and abide by each others' decisions.

Ethical problems also loom. In an increasingly collaborative biomedical environment, big pharma companies will undoubtedly have an even larger presence than they already enjoy. What can be done to prevent them from inserting confidentiality clauses in research agreements, or suppressing unfavorable data in studies they have paid for? Hayward said, "I think the expertise and depth of the clinical translational science investigators will simply be too great to stifle outcomes." Perhaps. But recent examples of interactions between big pharma and the FDA are not confidence-raising.

An Unstoppable Trend In A Growing Industry

Nevertheless, the trend toward big, team-oriented science is unstoppable. Lars Berglund, who leads the translational science group at the University of California, Davis, told BioWorld Perspectives: "Today we are faced with very complex diseases like heart disease and diabetes: They occur in many people, don't have a single underlying mechanism and have different effects across a range of organs. The basic scientist focused on small systems will continue to occupy a central position in discovery, but the framework of discovery must expand. Different disciplines need to pool their knowledge in ways that have not been done before, and further, they must provide a partnership modality that will work on a variety of projects, rather than having to re-invent the process de novo for each one."

Judged simply by publications the research locomotive is rumbling down the tracks faster than ever. NIH's PubMed service shows nearly 700,000 papers were published last year, with an annual rate of gain near 6 percent. The journals are getting fatter, and the number of titles, now near 7,000, is going up by about 1.5 percent a year. NIH research funding and grant awards have leveled off recently, almost certainly temporarily, and no one believes the feast of biomedical science is anywhere near ending.

In the banquet hall the satiated diner hoists himself to his feet and gazes at the table. "On second thought, maybe I will have a piece of pie."