Between "Science" and "Policy"

A long and divisive two-day meeting of the Antiviral Advisory Committee in July 1991 resulted in the recommended licensing of ddI, though not by unanimous vote. The clincher came on the second day, when the committee considered data that had been obtained, on special FDA consent, from a peek at the ongoing Phase II trial. In the study, patients who had already been on AZT for twelve months or more were randomly assigned either to continue AZT or to receive ddI. The results so far showed a T-cell decline in those who stayed on AZT, compared with a modest rise in those who had been switched to ddI. After what Project Inform referred to as "endless hand-wringing by some committee members," the committee voted in

favor of approving ddI for adult and pediatric AIDS patients no longer responsive to AZT; the vote was five to two.^[47] As it almost always does, the FDA endorsed the recommendations of its advisory panel sometime afterward.

Project Inform's newsletter, commenting on the approval, called it "in many ways the single most important victory in 6 years of AIDS activism" and described the FDA as "courageous."^[48] Kessler himself characterized the vote as "a milestone in drug review." "We're in the midst of a medical emergency," Donald Abrams, one of those who voted for approval, told a reporter for JAMA . "One person dies of AIDS in the United States every 8 minutes," Abrams added, repeating what was in fact a common ACT UP slogan.^[49] Later, when asked about his vote, Abrams described his role in essentially political terms, as one of "representing" doctors and patients from San Francisco, groups that wanted ddI to be approved.^[50]

Normally, as Sheila Jasanoff describes it in a study of regulatory science, the use of independent scientific advisory bodies allows agencies such as the FDA to "harness the authority of science in support of its own policy preferences." The experts on these panels "seem at times painfully aware that what they are doing is not 'science' in any ordinary sense, but a hybrid activity that combines elements of scientific evidence and reasoning with large doses of social and political judgment."^[51] Nonetheless, the experts normally speak in the language of science, and this has the important effect of legitimating the policies that are ratified. In the case of ddI, however, the scientific basis was so contested and the political pressures so extreme that panelists sought to disentangle their separate roles as scientists and policymakers—to make clear that as far as they were concerned, their vote was a scientific endorsement of neither ddI nor the use of CD4 counts as a surrogate marker, but rather a pragmatic policy decision. The panelists were going to allow patients to assume the risks that patients themselves, their activist representatives, and their physicians were demanding that they be allowed to assume.

Some refused to take this step, however. Deborah Cotton, a Harvard Medical School professor who was one of the "no" votes, expressed concern that the approval had set a bad precedent that "creates incentives" for other drug companies to press for approval with inadequate data. She worried, moreover, that it would be "a real challenge to explain to participants in the controlled trials why it is so important for the trials to continue and for them to remain in them."^[52] Now, more than ever, it was necessary to complete the trials

and make sure that the effect of ddI on a surrogate marker would translate into a genuine clinical benefit and decreased mortality. But once the FDA had given its blessing to the drug, based on data that seemed to show that ddI was better than AZT for those who had been on AZT for some time, why would anyone stay in the trial, at the risk of being in the AZT arm? Why not just drop out of the study and have one's doctor prescribe ddI? Belief had now hardened into relative certainty; in such an environment, was a clinical trial possible?

Paul Meier, a statistician from the University of Chicago who said he voted for approval reluctantly, noted that the committee would have preferred to make the approval conditional upon future studies, "yet had to take full approval because there is no such option."^[53] He commented to the New York Times: "I really genuinely worry that we are ratcheting down what has been a good standard for the F.D.A." Some New York activists, while supporting the decision, were also concerned about the inconclusive nature of the data. "This whole situation makes all of us very nervous," David Barr told the New York Times .^[54]