For two decades Harold Varmus spent nearly all of his time in the laboratory, avoiding most administrative duties, even a customary turn as department chairman. In the early 1990s though, Varmus decided it was time to contribute more to the institutional framework of scientific research at a time when a growing federal budget deficit threatened research funding. In taking on important administrative responsibilities in science, he followed the example of several other faculty members at the University of California San Francisco, who had recently occupied prominent positions in health and science policy administration. Most notably, his colleague in the department of biochemistry and biophysics, Bruce Alberts, was appointed president of the National Academy of Sciences in 1993, and involved Varmus in policy discussions within the Academy. Varmus served on an advisory committee that studied the future of research funding and joined a group of scientists who supported Bill Clinton in his bid for the presidency in 1992. These efforts brought him to the attention of Donna Shalala, the Secretary of Health and Human Services, who proposed Varmus for the directorship of the National Institutes of Health (NIH) despite his lack of administrative experience. Clinton nominated him in early August 1993 on the strength of his scientific reputation, in the hope that he would revitalize a research institution that had suffered recent setbacks.
Although he was easily confirmed by the U.S. Senate, the choice of Varmus to head the nation's foremost biomedical research and research funding agency drew a measure of opposition. Biomedical scientists and advocates of basic research hailed the selection of one of their own. Health activists, however, feared that his choice would lead to a diversion of funds from targeted research and drug development in the fields of AIDS, breast cancer, and women's and minority health.
Varmus took over the directorship at a critical juncture. During the 1990s NIH employed over thirteen thousand scientists in 25 institutes and centers, most of them located on a 322-acre campus in Bethesda, a suburb of Washington, D.C. The previous director, Bernadine Healy, had become embroiled in disputes with NIH researchers over her attempts to reform the agency's administration. Her proposal to grant patents for genes to the investigators who first discovered them had alienated influential scientists like James Watson, the co-discoverer of the double helical structure of DNA and the first director of the Human Genome Project, the newly launched international research project to determine the nucleotide sequence of all human genes. Promising researchers were choosing appointments in universities, hospitals, and private industry over a career at a fractured and politicized NIH. Moreover, recent federal budget deficits limited increases in the NIH budget for intramural (internal) science programs to less than the rate of inflation in the cost of scientific research--in effect a budget cut--and forced a reduction in the number of extramural research grants that strained the relationship between NIH and scientists in academia and industry (85 percent of its budget is spent on such grants, which makes NIH the leading source of biomedical research funding in the United States and the world). Surrounded by conflict and limited by budget constraints, Healy served for only two years.
Soon after his arrival in November 1993, Varmus embarked on a program of far-reaching administrative and personnel reforms. He inaugurated NIH-wide lectures and symposia to bring NIH scientists together and celebrate their achievements. He recruited ten new institute and center directors, trading on his standing as a leading scientist to attract well-known researchers for these positions, including Richard Klausner, director of the National Cancer Institute, and Gerald Fischbach, director of the National Institute for Neurological Disorders and Stroke. Varmus initiated an institute-by-institute evaluation of intramural programs. At the same time, he established the Center for Scientific Review as a way of restructuring the grant review process and making it more receptive to new disciplines and innovative but risky research. During his tenure, the Human Genome Project was nearly completed (a preliminary map of the forty-six human chromosomes was published in 2000, and the full sequence in April 2003).
Even though he had no legislative or public-relations experience, Varmus proved an adept backroom politician and public promoter of basic scientific research. The diplomatic skills he gained from leading the effort to settle on a uniform nomenclature for the AIDS virus, his facility with language in explaining complex matters of science to Congress and other lay audiences, and his willingness to address the concerns of lawmakers served him well in his primary task as NIH director: to secure generous funding for the agency and for biomedical research. Through public speeches, effective Congressional lobbying, and informal meetings with politicians, he built bipartisan support for an increase in the NIH budget from just under eleven to almost sixteen billion dollars during his tenure, and for a plan to expand it further to twenty-six billion dollars by 2002.
At the same time, Varmus resisted attempts to earmark NIH funds for the fight against specific diseases if the state of scientific knowledge did not warrant hope for a quick cure. He considered "wars" on cancer and other high-profile but complex diseases, including AIDS, an unproductive diversion of personnel and resources. (Varmus contrasted these latter crusades with the fight against polio in the 1950s, which was waged against a known pathogen and with readily available immunological methods.) Instead, Varmus publicly extolled the promise of basic biomedical research, which strives to uncover the underlying biochemical and genetic mechanisms of disease at the cellular level. As he wrote in a New York Times editorial in 1993, he was "convinced that the most effective long-term approach to improving health lies in fostering the research that increases understanding of genes and tissues."
Varmus's research experience taught him lessons in the importance of basic research, freedom of scientific inquiry, and serendipity that he sought to apply as NIH director. When he and Michael Bishop first began their collaboration, they intended to add experimental detail to recent findings about the biochemistry and the life-cycle of cancer-causing viruses. Only after several years of research and accumulating data did they realize that they were in fact making a discovery of much broader scientific and medical significance: the genetic origin of cancer.
In frequent congressional testimony, Varmus argued that without fundamental research, finding cures for diseases was a matter of conjecture and luck. By contrast, with such knowledge, a range of new therapies could be systematically developed. At the same time, he insisted that progress in basic science could not be planned, but was the result of inspired, persistent, and often unforeseen explorations by researchers who were free to follow their curiosity, instincts, and findings over many years. He convinced lawmakers to adopt and invest in his vision for long-term basic scientific study.
Varmus faced his greatest political challenge in navigating the intense controversy over stem cell research and the federal government's ban on the use of human embryos in such research. Scientists prize stem cells for their unique generative power. As it divides into two, a stem cell produces one of a variety of specialized cells, and one copy of itself, which can then divide again in a similar manner, and so on, over and over. Stem cells thus hold the promise that they can one day be used to regenerate tissue, organ, or nerve cells damaged by disease or the effects of aging. But the therapeutic promise of stem cell research was tempered by the ethical and political dilemmas it raises because stem cells were initially harvested from embryos that were discarded by fertility clinics or aborted. Under pressure from anti-abortion and conservative groups, Congress in late 1999 banned NIH from conducting or funding research that involved the deliberate creation or destruction of human embryos, whether to derive stem cell lines or for other purposes. In addition to facing the challenge of negotiating the contentious moral and political issues raised by this controversy, Varmus was concerned that it threatened the continued growth of the NIH budget.
Balancing the moral claims of patients who might benefit from stem cell research with the moral claims of potential life tested the limits of Varmus's political acumen. While he vowed to adhere to the ban on federal funding for stem cell research that involved the destruction of embryos, Varmus pointed out in congressional hearings that the ban, which did not apply to privately-financed laboratories, would give the public no oversight power, no role in setting research protocols, and no means of determining the uses of research findings. He endorsed an opinion by the legal office of the Department of Health and Human Services that the ban allowed NIH to fund embryonic stem cell research as long as such funding was not used to procure or destroy embryos, which would instead be donated by fertility clinics. Even so, Varmus proposed to proceed only after an NIH advisory panel had drawn up guidelines for stem cell research. The controversy remained unsettled during Varmus's tenure.
In November 1999, Varmus left NIH to become President and Chief Executive Officer of Memorial Sloan-Kettering Cancer Center in New York City. From the time of his confirmation in November 1993, he had considered a term of six years to be an appropriate and effective tenure for the NIH director, long enough to implement institutional changes and set new research agendas, but not long enough to produce fatigue and a bureaucratic mindset. Moreover, Varmus wanted to separate the process of choosing a new NIH director from the upcoming presidential election, so that the choice would be made in the best interest of science, not be clouded by political and electoral considerations.
When he was still at NIH, Varmus had become intrigued by the Internet and its potential to provide unprecedented access to scientific and medical information. During his last year at NIH he proposed E-Biomed, a system for electronic publication of new results and ideas in biomedical science that would enable researchers to make scientific and medical information accessible instantly, worldwide, in one place, and without the expense to subscribers incurred by the traditional system of publishing in journals. The cost of peer review and online publishing would be paid not by readers or their institutions, but by authors from their research grants. Scientific reports would be submitted either through editorial boards (as in traditional publishing) or, for less formal reports, directly through the general repository, approved by two editors with appropriate credentials. The NIH would build and maintain the electronic platform and software for publishing and archiving the digital articles, including those published in participating print journals.
The proposal stirred much interest but also much criticism and debate, particularly about peer review and publishers' revenues. Over several hundred years, traditional scientific publishing practices had evolved to ensure the quality and integrity of new research reports, and judge their importance, via editorial and peer review. In the culture of science (as in other scholarly fields) publication is a key component of career advancement, and where one publishes is as important as how much one publishes. Journals with a reputation for higher standards, as acknowledged by the scientific community, often provide a researcher with higher visibility and status. It seemed unlikely to many scientists that an untested venue such as E-Biomed could do the same. Scientific publishing had also become a very profitable enterprise in the twentieth century, as research in all fields rapidly expanded. Varmus's new model proposed to make that enterprise obsolete. At best, the open-access archive model seemed to require that publishers forego subscription revenue and copyright ownership, in effect giving away their product. Likewise, professional scientific organizations that published journals often depended on subscriptions to fund their operations, and could not embrace a model that seemed to threaten their livelihood. Despite this widespread skepticism, Varmus forged ahead with the repository component of E-Biomed, and worked with David Lipman (director of the National Center for Biotechnology Information at NIH) to develop the freely available online archive of scientific journal literature now called PubMed Central. In 2000, he became co-founder of the Public Library of Science (PLoS), a non-profit online publisher which makes full-text, freely searchable original articles in biomedical science available without cost to readers or restrictions on further distribution, except for a requirement that proper citations be provided. His continuing campaign for open-access publishing remains one of his main professional occupations.