Biotech Rules Part II: Science, Serendipity, and The Web
Biotech Rules Part II: Science, Serendipity, and The Web
Hal Plotkin, Special to SF Gate
Tuesday, March 2, 1999
Thanks to a deluge of recent optimistic press reports, expectations for the biotech industry have reached a fever pitch. But will we, in our lifetime, really see the kinds of revolutionary biotech advances we hear so much about?
No one knows for sure. My real concern is that all this hype is a smokescreen that prevents us from asking whether the right policies are in place to to take advantage of biotech’s many benefits as quickly as possible.
For people suffering with terminal or chronic diseases, this is not a philosophical question. It’s a matter of life and death.
This issue is central to a new report, Unlocking our Future: Toward a New National Science Policy released last fall by the U.S. House of Representatives Committee on Science.
Among the report’s more controversial recommendations is a call for the federal government to take another look at the peer review processes used to make grants to scientists. Its authors say that current peer review procedures don’t go far enough toward enhancing chances for the kind of accidental, serendipitous discoveries that have long been the hallmark of scientific progress. The question the report asks is: what can government agencies do to help scientists get lucky more quickly and more often?
Peer review is a proud tradition within such federal science agencies as the National Institutes of Health (NIH). Under peer review procedures, grant applications are reviewed by the best known scientists working in the same area of study.
Defenders of peer review say only experienced scholars can determine the merits of research proposals that are often quite technically complex. The laudable goal of the peer review process is, of course, to remove politics from the grant-making equation by putting funding decisions in the hands of experts.
On at least some occasions, though, peer review can substitute academic politics for the more common garden-variety kind. Scientists who don’t win federal funds say peer reviewers are sometimes reluctant to award grants for work that might raise questions about experiments being done by the peer reviewers themselves or by their closest colleagues.
What’s more, approaches that run counter to the prevailing wisdom in a given field are rarely favored. Generally speaking, scientists who “think outside the box” are more likely to end up living in a cardboard box than they are to get federal funding.
The history of science, however, demonstrates that experts often don’t have a clue about what is going to happen next. Thousands of crucial discoveries, from Archimedes’ formula for the measurement of volume to the development of vaccines, came about largely by accident, not by design.
For example, when healthy young people started dying mysteriously in the southwestern United States in 1993, it took the Center for Disease Control four weeks to figure out that a new hantavirus, carried by rodents, was the culprit. The discovery was made possible because the NIH, along with the U.S. Army’s Medical Research Institute, had done work, decades earlier, on a problem left over from the Korean War, cataloging the genetic structure of hantaviruses common in Asia.
Who would have guessed that an old catalog of diseases not previously known to have affected anyone in the United States would turn out to be the critical public resource that helped prevent a widespread fatal epidemic?
A similar accidental breakthrough, in cancer research, occurred more recently when scientists looking into the mechanisms of cell division stumbled across the answer to how the drug Taxol retards the spread of leukemia and breast cancer. Armed with that knowledge, cancer treatment researchers now have a promising new avenue to explore.
Whether it’s discovering X-rays, or Teflon, or gravity, or even coming up with the formula for cornflakes, science seems to move forward with greater speed when it’s not exactly sure where it’s going.
Thanks to the Internet, we now have the best chance in all of human history to maximize the role serendipity plays in science. While we can’t plan to get lucky, we can do things that make good fortune more likely. The most successful people I know have one trait in common — they make their own luck. When it comes to science, we may be able to do the same thing.
Take, for example, the modest start made by the National Center for Biotechnology Information, a federally funded agency whose mission is to develop and disseminate scientific data. NCBI’s databases, which contain information useful to scientists and researchers, are offered free of charge to anyone with a modem.
Those databases include GenBank, a DNA sequence archive, a Molecular Modeling tool and records from the Online Mendelian Inheritance in Man project. Few, if any, scientists could ever have hoped to amass such resources working on their own.
NCBI’s website is now getting about 4.5 million hits per day. Not bad for an agency whose relatively tiny $21 million annual budget — about one-third of which supports its website-related activities — is 1/20th the cost of a single Stealth bomber.
Thanks to NCBI, researchers fighting genetic diseases, such as Duchenne muscular dystrophy, are now making progress that previously would have been unimaginable. It’s impossible to predict what additional uses scientists will make of these data repositories in the years ahead. But one thing is clear: the chances of serendipitous breakthroughs have been increased because they exist.
However, there remains much to be done. For example, we still don’t have something as simple as a single good, comprehensive, easy-to-use, free database of the activities and areas of expertise of all federally funded researchers.
Taken together, the federal government currently spends about $35 billion each year on research and development, not counting funds spent by the Department of Defense on weapons development activities. It’s possible, even highly likely, that some of these researchers already have answers other researchers are looking for.
A for-profit company, called the Community of Science (COS), is taking up some of the slack. Research institutions can pay to join COS and supply information about individual scientists working at their sites. The information is then made available to other members whose employers pay membership fees to COS.
However, scientists working on their own or other interested individuals, can’t buy access to the records assembled by COS, which does not sell individual memberships. In addition, only the federal government, which has the full roster of all the research money it spends, has all the data needed to create the kind of searchable interactive archive that could lead to more cooperation between researchers, particularly across disciplines.
If scientists don’t know, and can’t easily find out, who else is working on questions that might be related to the problems they are trying to solve, then what are the chances they will find those individuals?
A recent $8.5 million five-year federal grant made to New Mexico State University to develop and assemble more online biotech-related information resources is a tiny step in the right direction. The recipients of that money say they will spend it on building more free public biotech information services.
An added bonus is the pledge of this new center’s director to do what he can to bring a more diverse group of scientists into the biotech fold by working more closely with historically black colleges and universities and other minority institutions.
Meanwhile, Dr. Harold Varmus, the former professor at UCSF who now leads the NIH, has promised to invest a big chunk of his agency’s 1999 $15 billion budget — which grew by 14.4 percent this year — toward developing more of the bioinformatics technologies needed to catalog and share scientific data.
The debate about what steps the government can take to improve the scientific process is not as sexy or eye-catching as the current surplus of speculative news stories about exciting, and usually unverified, miracle cures for cancer, AIDS, or other diseases.
Federal research databases rarely, if ever, make headlines. Nor do the mechanisms used to award research grants. But advances we make in those areas today could, 100 years from now, turn out to have been the biggest breakthrough of them all.
As the recent congressional report suggests, perhaps the time has come for us to spend less time fixated on where biotechnology might take us, and more time thinking about exactly how we are going to get there.