Earlier this year, Bruce Weniger, a member of the Presidential Advisory Council on HIV and AIDS, and Max Essex, chairman of the Harvard AIDS Institute, wrote an opinion editorial in the New York Times about the current impasse in AIDS vaccine trials. The piece sparked some controversy; here, Richard Marlink, executive director of the Harvard AIDS Institute, explores the topic in more depth.

Wonderful new drugs are restoring health to those living with AIDS. Yet this success contrasts woefully with progress toward a vaccine, where the effort is foundering and bereft of accountable leadership. Our failure to develop a vaccine creates serious consequences, as each year of delay in vaccine development jeopardizes additional lives, adds enormously to human suffering, and incurs staggering health care costs.

Each year, some 40,000 to 80,000 Americans and millions worldwide become infected with HIV. Estimates predict that each person with HIV in the United States will incur medical costs of $119,000 over a life-time, and these costs may increase with the new protease inhibitor cocktails. Undoubtedly, a vaccine is the only long-term solution for the pandemic. And for the developing world, where more than 90 percent of HIV-infected people live, a vaccine is the only solution.

In the early 1980s, after the discovery of HIV and methods to grow it in test tubes, a vaccine was expected to appear soon. But after thirteen years, the National Institutes of Health (NIH) has been unable to bring a single candidate AIDS vaccine into field trials to see if it even works. In response to a critical review, the agency recently restructured its AIDS vaccine program by appointing a part-time committee of external scientists to guide the effort. Although welcome, this reform is unlikely to overcome the fundamental handicaps of the agency itself.

Several reasons account for the agencys problems. NIH, the worlds premier biomedical research agency, is run by sincere and devoted scientists. Yet its principal mission is to advance basic scientific knowledge and to nurture academic research--not to develop products like vaccines. Its ponderously slow bureaucracy hinders efforts to respond quickly to epidemics. In addition, despite the planned revamping, there still will be no senior, full-time official whose sole responsibility is to produce an AIDS vaccine as soon as possible and who has the vaccinology background, budgetary resources, and contracting authority to do so.

History teaches us that a different kind of entity would be better suited to manage an expedited vaccine program in the face of a public health emergency. To illuminate the current dilemma, it may help to recall Jonas Salks struggle to develop a polio vaccine in the early 1950s, when frightening epidemics were sweeping the United States, each year killing up to 3,000 victims and crippling 20,000 more.

By 1949, John Enders and his colleagues had grown the polio virus in test tubes, a critical step in vaccine development and one duly recognized with a Nobel prize. Then Salk applied this discovery by developing a killed whole-virus vaccine. Eminent polio researchers, however, including Enders and Albert Sabin, opposed field trials of the Salk vaccine. At that time, orthodox science dictated that ideal vaccines for viral diseases required living but weakened virus strains. Criticizing Salks product for lacking sufficient scientific rationale, the scientists thought it would either not work, or not work as well as future live vaccines, for which they favored waiting. They sought to postpone the trials.

Fortunately, polio research was then funded by the March of Dimes campaign--an enterprise of the private National Foundation for Infantile Paralysis--whose energetic director, Basil OConnor, possessed the single-minded goal to stop the epidemics as soon as possible. Undeterred by the lack of consensus among scientists and motivated by the fears and concerns of parents, the March of Dimes committed to widespread public distribution of the vaccine, even if it proved to be only 25 percent effective. Thus, Salks vaccine was tested among over half a million children before the summer polio season of 1954.

On April 12, 1955, the results of the vaccine trials were broadcast to an expectant world: the vaccine worked. In comparing the subsequent incidence of paralysis among the children who had received the vaccine with those who had received placebo shots, the rate of paralysis was found to be 72 percent lower among vaccine recipients. A better vaccine--Sabins oral vaccine--was not introduced for another seven years. In that time, Salks vaccine saved tens of thousands of lives and prevented hundreds of thousands of lifelong disabilities. The critical lesson: Do not let a better vaccine of the future preclude a potentially lifesaving one now.

The polio saga illustrates the tension between academic basic science and empirical applied research. The former scrupulously pursues knowledge and elegant solutions, while the latter seeks rapid answers to urgent public health problems. The struggle between these complementary approaches is one of the most unfortunate impediments to rapid AIDS vaccine development. The lack of incentive for industry is another impediment. Drugs that treat are much more profitable than vaccines that prevent.

The current impasse in vaccine development arose in June 1994, when NIH halted plans to conduct field trials of the first generation of bioengineered AIDS vaccines, overruling its own vaccine advisory panel and mid-level staff scientists. These vaccines were called envelope products because they mimic the outer coat of the virus and thus generate antibodies that could potentially cover the virus surface and block infection.

During eight years of research and development, enormous investments of private capital and federal funding were expended for laboratory and animal studies and preliminary human testing of safety and immune response. The new products were found be successful in protecting chimpanzees from HIV. Had these entities been approved, the human trials could have begun in 1994, costing some $20 million annually over three years. And had the vaccines worked in people, the entire expense of the trials would have been recouped by the first 500 cases of AIDS prevented.

All who analyzed the data agreed the envelope vaccines appeared to be safe. Proponents of the field trials said they would accelerate vaccine development by determining definitively if and how well the products worked and perhaps reveal why. Opponents criticized the vaccines according to a fashionable but unproven theory for predicting whether an AIDS vaccine would work. But vaccinology is a trial-and-error science in which progress sometimes requires the courage to conduct careful and expensive human trials to test theory and revise vaccine designs accordingly. Although some experimental vaccines provide clues for efficacy, such as blocking infection in animal models, most do not. As Jonas Salk once said, You have no chance of success unless you are willing to fail.

The decision to cancel field trials sent shock waves through the biotechnology and pharmaceutical industries, which convert basic science findings into vaccine products. Discouraged, Biogen, Bristol-Myers Squibb, and Merck-Repligen dropped their AIDS vaccine programs. Chiron-Biocine, Genentech, and Therion Biologics scaled back the pace of their product development.

So, what is needed to accelerate a successful AIDS vaccine? More committed funding, for one thing. Vaccines have long been the stepchild of AIDS research at NIH, garnering less than 10 percent of the total spent each year. Despite the planned reforms in organization, only 7 percent of the additional $100 million Congress earmarked for AIDS in the current fiscal year will go to vaccines. To succeed, vaccine research must receive a greater portion of NIHs support.

We should also learn from organizational structures with proven track records in delivering successful products. Consider the successful development of the U.S. space program: the mission was given to a newly created NASA, not to the existing, yet grant-making, National Science Foundation. The March of Dimes gave us both the Salk and Sabin polio vaccines. In addition, our militarys applied research system has developed dozens of vaccines--from flu to measles, plague, and typhoid--sometimes in record time to meet war needs.

The mandate for developing an AIDS vaccine, therefore, should go to a goal-oriented, rapidly acting, streamlined, and semi-independent task force or agency to pursue all possible strategies systematically and simultaneously. This entity should propose legislation to overcome economic and legal disincentives for private investment and should work closely with industry and other nations to shepherd vaccine products through field trials and FDA approval. Its director should report to the White House and be an accomplished vaccinologist or pharmaceutical research manager. And the agency should be well funded by the government.

The consensus among scientists today is that an effective AIDS vaccine is possible. In a time of disillusionment over Sputnik, President Kennedy galvanized the nation by setting a deadline to reach the moon. In May of this year, President Clinton declared a goal of protecting the worlds population with an AIDS vaccine within the next ten years. Reinventing the federal AIDS vaccine effort will be an essential step toward such an urgent and inspiring achievement.