Biomedical Research, Its Funding and Its Practice: Where Do We Go from Here?

In the United States before World War II, support for basic biomedical and clinical research was haphazard, depending somewhat on the kindness of strangers.  For example, that far-seeing politician Franklin D. Roosevelt, who was a victim of polio, established the National Foundation for Infantile Paralysis (March of Dimes) in 1938.  Those of us of a certain age remember the cards filled with dimes in the stores and elementary schools of our youth.  This made a difference, ten cents at a time.  The March of Dimes funded Jonas Salk in his research on the original polio vaccine.  Occasionally universities and hospitals supported biomedical research, such as the University of Toronto and diabetes.  The scientists responsible for insulin sold it to the university for one dollar.  It costs more than that today, even after adjusting for inflation.  The Rockefeller Institute for Medical Research and the Carnegie Institution of Washington were also very productive but relatively small.

World War II demonstrated clearly what public funding of science and technology could accomplish when scientists were given the resources to do their research – radar and modern industrial engineering, for example, plus fresh frozen plasma and penicillin.  Penicillin was discovered by Alexander Fleming in 1928 but was mass produced only during the war.  After WWII, Vannevar Bush of MIT, who had been Director of the Office of Scientific Research and Development from 1941-1947, established the National Science Foundation to support basic research in the sciences.  The National Institutes of Health had its origin in 1887 as the Marine Hospital Service.  The National Institute of Health was inaugurated in 1940 by President Roosevelt (click on the embedded video to hear a Presidential address).  Since then, NIH has grown into an organization with nearly 30 different institutes and centers that cover all biomedical research.  NIH has been remarkably successful by any measure, despite what the current Secretary of Health and Human Services says.

In 2024, the NIH budget was $47B, with about $37B of that provided to scientists and their institutions across the United States and occasionally in other countries.  The Wellcome Trust in the United Kingdom comes in second with a budget about £1.7B (China may dwarf these totals, and that is beginning to show).  There are several reasons NIH has been so successful since 1950.  The people, Republican/Democrat/Independent, through the agency of the US government, have agreed that biomedical research is a public good in both the formal and general senses.  It is no accident the eleven confirmed Directors of NIH since 1950 have been licensed physicians and scientists with an appreciation of the importance of biomedical research.  These men and women published more than 2,100 scientific papers in their medical careers, according to PubMed.

NIH Directors include William H. Sebrell (1950-1955: Truman-Eisenhower) who did his medical internship at the Public Health Service Marine Hospital in New Orleans and was an expert on nutrition who served as Director of the National Nutrition Program during WWII.  Robert Q. Marston (1955-1968: Eisenhower-Kennedy-Johnson) worked with Howard Florey on penicillin.  He later led the desegregation of the University of Mississippi School of Medicine in Jackson and correctly argued that the “War on Cancer” declared by Richard Nixon was bad policy if it conflicted with other research.  Besides, declaring war on an improper noun is a fool’s errand because it cannot he won.

Donald S. Fredrickson (1975-1981: Ford-Carter-Reagan) was an expert on cholesterol metabolism and an author of the foundational text, The Metabolic and Molecular Bases of Inherited Disease, which was last published in hard copy in 2000 in four volumes and more than 6,000 pages; it is now available online.  He was also a member of the National Academy of Sciences.  Bernadine Healy (1991-1993) was a cardiologist and the first woman faculty member at Johns Hopkins University School of Medicine in the Cardiology Division of the Department of Internal Medicine.  Harold Varmus (1993-1999: Clinton) was awarded the Nobel Prize with J. Michael Bishop for their studies showing that oncogenes are the products of normal cellular genes whose mutation leads to cancer.  These are often the target of successful chemotherapy.

Francis Collins (2009-2021: Obama-Trump-Biden) was Director of the National Human Genome Research Institute from 1993-2008 and something of a lightning rod in recent years. [1]  He retired from leading his research laboratory at NIH earlier this month.  Monica Bertagnolli (2023-2025) is a surgical oncologist and the previous director of the National Cancer Institute.  Her research has been on the genetic basis of colorectal carcinogenesis.

NIH is obviously in turmoil in early 2025 as the current Administration takes its whacks at the entire federal government.  But regarding public support for basic and clinical scientific research, what is the better path?  Repairing what must be fixed at NIH or chopping down this oak and regrowing it from a single acorn in an uncertain climate?  The question answers itself.

The Nobel Prize in Physiology or Medicine is an imperfect measure of scientific importance, but it is awarded for signal advances in the biomedical sciences.  Thus, we will use it as a proxy for “good science” here.  Virtually every Nobel Prize awarded to scientists from the United States has been supported by funding from NIH or a predecessor or a public institution such as the Agricultural Experiment Station at Rutgers (Selman Waksman, 1950, discovery of streptomycin, the first antibiotic active against tuberculosis).  For NIH funding since 1985, the data are available (so far)  in NIH RePORTER and are shown in Table 1.

Nobel Prizes in Physiology or Medicine Awarded to Scientists Supported by NIH since 1985 – United States.
Year	Scientist(s)	Discovery
1985	MS Brown/JL Goldstein	Regulation of cholesterol metabolism
1986	S Cohen	Discovery of growth factors
1987	S. Tonegawa	Generation of Antibody Diversity
1989	H Varmus/JM Bishop	Oncogenes
1992	Edmond Fischer/Edwin Krebs	Protein phosphorylation in regulation of cell function
1993	RJ Roberts/PA Sharp	Split genes (introns)
1994	AG Gilman/M Rodbell	G-proteins in signal transduction (many are oncogenes)
1995	EB Lewis/E Wieschaus	Molecular control of embryonic development
1996	PC Doherty	Cell-mediated immune response
1997	S Prusiner	Prions: Infectious proteins in disease
1998	R Furchgott/L Ignarro/F Murad	Nitric oxide signaling in the central nervous system
1999	G Blobel	Intracellular protein trafficking
2000	P Greengard/E Kandel	Signal transduction in the nervous system
2001	L Hartwell	Key regulators of the cell division cycle
2002	HR Horvitz	Programmed cell death during development (apoptosis)
2004	R Axel/LB Buck	Odorant receptors and the olfactory system
2006	A Fire/CC Mello	RNA interference
2007	MR Capecchi/O Smithies	Transgenic mice using embryonic stem cells
2009	E Blackburn/C Greider/J Szostack	Telomeres and telomerase
2011	BA Beutler	Innate immunity
2013	J Rothman/R Scheckman/T Südhof	Vesicle trafficking in cells
2015	WC Campbell	Ivermectin as a therapeutic for nematode infections
2017	JG Hall/ M Rosbash/MW Young	Circadian rhythms
2018	JP Allison	Immune checkpoint inhibitors (cancer therapy)
2019	W Kaelin/P Ratcliffe/G Semenza	How cells adapt to oxygen availability
2020	HJ Alter/M Houghton/CM Rice	Discovery of Hepatitis C virus
2021	D Julius/A Patapoutian	Receptors for temperature and touch
2023	K Kariko/D Weisman	RNA modifications for mRNA vaccines
2024	V Ambros/G Ruvkun	MicroRNAs and regulation of gene expression
Notes: RJ Roberts (1993) did his research at New England Biolabs, which to my knowledge remains one of only two independent companies that made the revolution in modern molecular biology possible, but his research at NEB was funded by NIH.  WC Campbell (2015) of Drew University received no NIH support.  His co-awardee is from Japan and received public support.

But the story goes deeper than a mere list and can be illustrated by the interconnections underlying these seemingly disparate results.  One network that intersects with many other networks represented in Table 1 is sketched briefly below.

Peyton Rous of the Rockefeller Institute was awarded the Nobel Prize in 1966 for work that was first published in 1910 on A Transmissible Avian Neoplasm (Sarcoma of the Common Fowl, pdf).  The cancer-causing virus was later called Rous Sarcoma Virus (RSV).  This form of carcinogenesis could not at the time be replicated in mammals and his work was not appreciated. How could birds have any relevance for human biology?  Let us count the ways.  In the 1980s the “oncogene” from RSV was identified as the protein kinase pp60src [2].  Harold Varmus and J. Michael Bishop of the University of California–San Francisco then showed that retroviral oncogenes were originally normal cellular proteins taken up by the virus that when expressed in host cells cause abnormal proliferation leading to cancer: c-onc is normal, v-onc causes cancer.

Beginning in the 1950s Edmond Fischer and Edwin G. Krebs began their research on protein kinases, and by the time pp60src was identified as a protein kinase the importance of this could be placed in a meaningful framework.  The human “kinome” encodes 518 protein kinases.  During the 1960s at Vanderbilt Earl Sutherland figured out how “second messengers” inside cells worked downstream of hormones that bind to the surface of cells, in his case by regulating a major protein kinase that controls metabolism.  In 1987 Paul Nurse identified the human protein kinase that is the master regulator of the cell division by expressing the gene for human cdc2 in a yeast cell deficient in the homologous gene, showing that cell division cycle (cdc) regulation is conserved from the origin of the lineage that includes Fungi and humans, i.e., the human gene works in Schizosaccharomyces pombe.  Well-chosen experimental models do rise to the occasion.

Stanley Cohen of Vanderbilt identified epidermal growth factor (EGF) as a hormone that binds to its receptor on the outside of a cell that has a protein kinase domain on the inside and induces the cell to divide.  EGF2 is also known as HER2, which is often amplified in breast and other cancers.  Herceptin is the monoclonal antibody that targets HER2.  By binding to HER2, Herceptin prevents the protein from signaling the cancer cells to proliferate.  These cells eventually die instead of doing their mischief.

This is only a very small sampling of the network effects [3] that are the major product of wide-ranging, investigator-initiated basic research.  None of the discoveries described above would have had much significance in isolation.  It took over 50 years for the work of Peyton Rous to be appreciated, but when it was a floodgate opened.  The research described above was pursued as a public good by (mostly) disinterested scientists who wanted to “figure things out” because as the plaque in Charles B. Huggins’s office at the University of Chicago (Huggins shared the 1966 Nobel Prize with Peyton Rous), “Discovery is our business.” Sometimes discoveries are little, sometimes they are big, but they are all important.. Yes, scientists are only human. Some are unpleasant.  Very few have started out to “win” a Nobel Prize and gotten his or her hands on one [4].

This brings us to a question of today, whither NIH?  The institution can certainly be improved.  Having dealt with NIH and been successful at times, it is still a daunting task to apply for an NIH grant.  Preparation and writing can require a year or more.  Overall success rates are about 20%.  The only way to get that up to the 33% that is reasonable [5] is more funding.  But that is certainly not likely in the foreseeable future.

The typical neoliberal answer to the apparent and real travails of NIH is that Big Pharma can do the research, that business is always more efficient than a university or research institution.  This, of course, leaves out the correct understanding that efficiency, which means doing more with less, has a very tenuous relationship to effectiveness.  Big Pharma generally spends more on marketing, including lobbying, than research.  But that is for another time.

The anticancer drug imatinib (Gleevec/Glivec) is often used as the index case for the notion that Big Pharma (or Little Pharma, initially supported by NIH, to be bought by Big Pharma in a founder cash-out) could supplant NIH and similar organizations.  The development of imatinib was led by Nicholas Lydon of Ciba-Geigy (later merged with Sandoz to become Novartis).  The drug was revolutionary.  It often cured chronic myelogenous leukemia and other cancers outright.  But the target of imatinib had been identified in work that began in the late-1950s by physicians and scientists in what became known later as the Fox Chase Cancer Center, which has depended on NIH fir its existence, before and after it became a part of Temple University.

David Hungerford and Peter Nowell identified an abnormal chromosome in leukemic cells that was a hybrid of Chromosome 9 and Chromosome 22.  This Philadelphia Chromosome was the first genetic abnormality associated with a specific cancer.  Much later the protein produced by the translocation was identified.  This hyperactive mutant protein contains the active part of the Abl protein kinase (again), which regulates cell proliferation, among other processes.  Thus, the cancer connection.  Without the solid foundation outlined above, one that was built using public support, the network effects that made development of imatinib possible could not have existed.  No understanding of protein kinases, no drug target, no drug, no matter how big the budget and how expensive the robot used for screening drug candidates was at Ciba-Geigy.  Moreover, others who assisted in the development of imatinib were funded by NIH (e.g., Tony Hunter, Brian Druker, Charles Sawyers).

The current question is: Can NIH be improved now?  Perhaps.  The incoming NIH Director designee is Jay Bhattacharya of Stanford.  After receiving his MD from Stanford, he went straight to a PhD in Economics, also at Stanford, instead of a residency that would have qualified him to be a physician.  Some of his work on health outcomes has been funded by NIH.  However, Dr. Bhattacharya will be the first Director who is not a physician and biomedical scientist.  This is not disqualifying, but experience matters.  In the meantime, one can only watch and do what can be done.

Finally, much has been written and broadcast recently about the current state of American Universities, with a recurring question being, “Why can’t these universities just use their huge endowments to support biomedical research?”  This is something of a red herring, whatever these well-endowed universities do with their money.  But one of the things both do in my experience is develop the new programs and construct the buildings and infrastructure that make biomedical and other research possible.  State universities do it with support from their people by way of their state legislature.  One needs only to compare the campus of Johns Hopkins University School of Medicine today with that of 1995, or the campus of Georgia Tech today with that 30 years ago.  Those new buildings were not paid for by NIH, but they do make NIH- and NSF-supported research possible and productive.  Research funded by the public in both kinds of institutions should belong to the public, in perpetuity as a public trust, but the Bayh-Dole Act of 1980 does get in the way, in more ways than one.

We can now return to Vannevar Bush.  His vision was to spread the wealth around the country and have research funded by the public conducted where the people lived.  Since 1950 this has worked very well, at institutions with and without large endowments.  Trump v2.0 is up in arms about indirect costs (“overhead”) associated with this funding.  This averages about 30% overall (click on the “Schools of Medicine” link) for extramural NIH support, which seems reasonable.  The alternative is for the public to build the laboratories directly.  There can be no doubt that on occasion a small increment of this money is spent less than well.  But every dollar I spent from the grants I have received over the years was audited in real time.  This is true of virtually everyone who has ever had a conventional grant from NIH, NSF, DOE, USDA, or any other organization that funds scientific research.  Contrast this with the Department of Defense, which has not passed an audit in living memory.

The problem is not that money is “wasted,” although there are careerists who last a long time as a member of their own funding club.  One former colleague of sorts has more than 500 entries in PubMed, but he is still not the member of the National Academy of Sciences he expected to be in the later stages of his career, primarily because of the conflation of quantity with quality.  The problem is that the only way for scientific research to produce useful knowledge is to essentially let a thousand flowers bloom.  Some seeds may not germinate.  Some blooms will soon wilt.  Some blooms will be perennial flowers.

Peyton Rous (1879-1970) showed 115 years ago that a chicken cancer could be transmitted from one chicken to another by way the virus that became known as Rous Sarcoma Virus.  The oncogene that had been taken up from the chicken DNA by that virus is a normal protein kinase in the chicken (and other animals).  Would oncogenes have been figured out eventually?  Yes.  Still, there are 85 FDA approved protein kinase inhibitors used to treat various cancers that can be traced back to Peyton Rous of the Rockefeller Institute for Medical Research who wanted to understand cancer.  Not chicken cancer, but cancer.  Rous would be very happy with the far-reaching reaching network effects of his research.  These were barely imaginable during his long life.  If American science is to regain its footing, we must go back to his future and that of those who stood on the shoulders of giants, even if they were not recognized at the time.

Notes

[1] The NHGRI strategy for sequencing the human genome was supplanted rather rapidly by Craig Venter’s shotgun technique more twenty-five years ago, and the federal research establishment has not covered itself in glory over the past five years.  This has contributed to current dissatisfaction with science and scientists.

[2] For our purposes here, a protein kinase adds a phosphoryl group from ATP to another protein and “turns this other protein on” to make things go.  This includes cell division, metabolism, and cellular signaling from outside to inside.  Without protein kinases no eukaryotic cell can function.  “pp60src” stands for phosphoprotein of 60,000 molecular weight that produces a sarcoma due to RSV infection.

[3] From the Harvard Business School link: “The term network effect refers to any situation in which the value of a product, service, or platform depends on the number of buyers, sellers, or users who leverage it. Typically, the greater the number of buyers, sellers, or users, the greater the network effect—and the greater the value created by the offering.”  We can ignore the use of “leverage” as a verb while substituting “scientific results, techniques, infrastructure, and scientists” for “buyers, sellers, or users.”  Network effects make science go.  Without network effects there can be no scientific progress.

[4] One James Dewey Watson may be the exception that proves this rule, but his collaborators demurred.

[5] I have been writing (35 years) and reviewing grant applications (25 years), including service as a review panel leader, so I think my anecdata are data.  What I have learned is that one-third of applications should get funded because there are no substantive differences among the top 33% of any given pool.  With success rates often in the single digits the process resembles a lottery.  Scientists who will spend as much as a year on a grant application while running a laboratory and teaching are serious minded people who know how to make discoveries and meet a payroll.  Another third should get funded after revision. The final third will probably remain hopeless forever.  My first mentor in biochemistry told of the time when review panels met to decide which applications not to fund.  Imagine the opportunity costs of funding less than 10% of grant applications to the National Cancer Institute.  Or only 20% in the other institutes.  What I have learned, to paraphrase George Carlin, is that in some fields those who are funded are a “relatively small club and you ain’t in it.”  NSFW!  But this is very fixable.  Also, George was more on point about everything than even he could have imagined twenty years ago.  Alas.