"Waiting for the Right Disease"
Samuel Broder, the head of the NCI, had not been putting all his eggs in the suramin basket. In late 1984 he had put out the
word to the big pharmaceutical companies (the ones he considered capable of quickly bringing a drug to market): Send us anything you have on the shelf that might inhibit a retrovirus, and we'll do the assay to see if it halts replication of HTLV-III. Burroughs Wellcome, the North Carolina-based subsidiary of a large British firm called Wellcome PLC, submitted ten compounds, and in February 1985 one of Broder's researchers, Hiroaki Mitsuya, found that one of the compounds was a reverse transcriptase inhibitor with strong antiviral activity: azidothymidine, called 3&0374;-Azido-3&0374;-Deoxythymidine in full or just AZT for short.
AZT had a peculiar history. In the early 1960s, a researcher named Jerome Horwitz at the Michigan Cancer Foundation decided to design a drug that would keep cancer cells from duplicating. With funding from the NCI, and working with such unlikely ingredients as herring sperm, Horwitz and his coworkers synthesized a group of compounds called dideoxythymidines that were designed to look like nucleosides, the building blocks of DNA. In theory, these "nucleoside analogues" would substitute themselves for real nucleosides, thereby interfering with formation of DNA molecules. Without more DNA, the cancer cells would simply stop duplicating. In practice, the treatment was a complete failure. Horwitz gave AZT and the other dideoxythymidines to mice with leukemia, but the drugs showed no effect. "My colleagues and I said that we had a very interesting set of compounds that were waiting for the right disease."
Burroughs Wellcome had tested AZT against animal viruses but had dropped this line of inquiry since it was unrewarding. Now, after getting the good news about AZT from Broder, Burroughs Wellcome filed an "IND" (investigational new drug application) with the FDA. Phase I trials began in July 1985 with nineteen U.S. AIDS patients, under the auspices of the NCI and in collaboration with Duke University. Mitsuya announced the results of the six-week study on the last day of an AIDS conference the following January: AZT kept the virus from replicating in fifteen of the nineteen research subjects, boosting their immune systems (as measured by their T-cell counts) and relieving some of their symptoms. "It's not a dream drug," Mitsuya explained in a television interview, stressing the need for additional testing.
The formal publication of the study in Lancet in March 1986 spelled out more of the details. Researchers recently had discovered that AIDS was frequently accompanied by neurological impairments,
which indicated that the virus was also affecting cells in the brain. An effective therapy, therefore, would have to be capable of crossing a circulatory system defense called the "blood-brain barrier," a feat that many drugs could not accomplish. Fortunately, AZT did appear to cross the blood-brain barrier. In addition, though some subjects had experienced headaches or had developed low white cell counts, the drug could be tolerated relatively well. This was a relief, because AZT "might have been expected to produce intolerable side effects" (in the words of Jean Marx, the reporter for Science who described the trial), given the mechanism of drug action. AZT "fooled" the reverse transcriptase enzyme into using it, in place of the nucleoside it imitated, when transcribing the virus's RNA to DNA. Then, once AZT was added to the growing DNA chain, AZT's structure prevented any additional nucleosides from being added on: reverse transcription simply came to a halt at that point, and the virus stopped replicating. But the problem was that since AZT terminated DNA synthesis, one might logically anticipate that it would have harmful effects on the DNA in healthy cells.
Having shown initial evidence of relative drug safety , the researchers had accomplished the formal objectives of a Phase I trial. But there was nothing to prevent them from reporting the apparent good news about efficacy —the news that attracted media attention. "The results also suggest that at least some immunological reconstitution occurred in most of the patients …, and that a clinical response was obtained in some." However, these findings had to be treated cautiously, since simply being in a trial "may have a strong placebo effect in influencing such factors as appetite and sense of well-being, and it is even possible that improved nutrition may then induce changes in immune function." Only the next step—a so-called "placebo-controlled" Phase II study, conducted in "double-blind" fashion so that neither the subjects nor the researchers would know who was receiving AZT and who was receiving a placebo (a look-alike dummy pill)—could determine whether the observed clinical improvements were truly due to the drug. (This study would be funded by Burroughs Wellcome and conducted at a number of academic centers, including the University of Miami, where it was under the direction of Dr. Margaret Fischl, and the University of California at San Diego, under Dr. Douglas Richman.) The NCI researchers concluded with a summary of the questions that remained to be answered: "We cannot say whether AZT can be tolerated over a long time, whether viral drug resistance will
develop, or ultimately whether AZT will affect disease progression or survival in patients with HTLV-III-induced disease. These are issues which can be resolved only by appropriately controlled long-term studies."