"We're Scientists Not Alchemists"
Unlike clinical research in AIDS, basic research has not had a powerful constituency to advocate on its behalf. The relevance of basic science to the lives of PWAs is far less apparent than clinical studies which hold the promise of proving a new drug's efficacy. Yet, new treatment options for people with HIV and a vaccine to protect the uninfected largely depend on the success of basic investigation.
-- Gregg Gonsalves, "Basic Research on HIV Infection: A Report From the Front," 1 June 1993
Data are willfully misinterpreted. Time is wasted. Money is wasted. Lobbyists and legislators set research priorities. Hopes are dashed. We are trapped in the outdated clinical research paradigms of the mid-1980s -- not only in vaccine research, but also in antiretroviral research, with its plethora of me-too nucleoside analogues... while pathogenesis research moves on into uncharted territory, its promised clinical payoff cruelly far off to anyone now living with HIV.
-- Mark Harrington, "The Crisis in Clinical AIDS Research," 1 December 1993
For TAG, 1993 began with a rush and a push. The election of Bill Clinton as president released a torrent of legislative initiatives which had been blocked under the first Bush administration. High on the list was legislation reauthorizing the National Institutes of Health (NIH), which had been held up due to right-wing opposition to biomedical research focusing on fetal tissue and stem cells, issues which have once again moved to the center of the political debate under the second Bush. Back in January 1993, Senator Edward Kennedy (D-MA) and Congressman Henry Waxman (D-CA) planned to introduce NIH legislation as the first item of business in the 93rd Congress. Staffers Mike Iskowitz in the Kennedy office and Tim Westmoreland in the Waxman office both planned to introduce TAG's proposed reforms strengthening the NIH Office of AIDS Research (OAR) into the new law.
The NIH bill would strengthen the OAR by giving it a full-time director; the ability to plan, coordinate, and evaluate the AIDS research programs carried out by the 18 NIH institutes; and the ability to propose and administer a centralized AIDS research budget, distributing the money to the various institutes in line with an annual strategic plan developed with extensive input from outside scientists. In addition, the OAR would have an emergency discretionary fund and the ability to move AIDS funds from institute to institute. The plan was designed to eliminate gaps in research, reduce redundancy, address emerging opportunities, and create a more well-rounded and coordinated AIDS research system.
Healy and the institute directors held an emergency meeting in Bethesda. The directors suggested she propose a one-year "study" of the OAR reforms, which would have effectively killed them. Instead she forwarded, without endorsement, a letter from the directors to Shalala. Healy, who wanted to keep her job, told me, "The Institute directors wanted me to walk the plank on this. I told them it was partly their own fault for not having their act together." The bill moved so fast that the NIH bureaucracy had little time to mobilize opposition. The Senate Labor and Human Resources Committee passed the bill unanimously on January 26.
In the House, the Republicans opposed giving the OAR power over the AIDS research budget. Without this budget authority, the OAR would continue to be a paper tiger. This authority was preserved by a single vote on party lines in committee, and the legislation headed for the floor.
Having a hand in legislation was a heady experience for TAG. Early in February 1992 TAG members were arrested conducting civil disobedience at Hoffmann-La Roche headquarters in Nutley, New Jersey, in a protest over sluggish and insensitive drug development programs. Eight days later, Gregg Gonsalves and I were in Kennedy's office watching the debate on S.1 on the Senate floor. For three days, the machinery of government came to a standstill as debate raged over an amendment proposed by Oklahoma Republican Don Nickles, who proposed a rider barring HIV-infected immigrants from the U.S.
The Nickles amendment, an attempt to codify a Bush-era regulation, and to make political points off the plight of thousands of Haitians who were escaping economic and political collapse and landing at Florida shores, was a good early example of how powerful and persistent the opposition would be to Clinton administration efforts to mitigate some of the bad AIDS policies of the previous era.
The Democrats were unable to stop the Nickles amendment from being attached to S.1; it was horrible to watch a racist and xenophobic policy being attached to legislation intended to improve AIDS research. Ultimately, the noxiously amended S.1 passed the Senate by a vote of 93-4.
On February 25 a tearful Bernadine Healy announced she would not stay at NIH: the administration had decided to replace her. The next day, terrorists set off a bomb in the parking lot underneath New York City's World Trade Center. Six people were killed, and hundreds injured.
Congressional passage of the OAR legislation ultimately took several months, but the outcome was clear: once the bill became law, and after Clinton named a new NIH Director, a search for a full-time OAR Director would be undertaken, and NIH AIDS research would be subject to new oversight, planning, coordination, and budget authority.
Gonsalves' report was critical because it was a grim year for AIDS drug development. In April The Lancet published results from the British-French Concorde study comparing AZT with placebo in HIV-infected individuals without symptoms. While a similar study had been stopped in the U.S. in 1989 (when 2% of people on AZT progressed to AIDS, versus 4% on placebo), the Concorde had continued until its results could be generalized to a majority of participants. The results were grim: early AZT was no better, and possibly worse, than placebo. HIV mutated to become resistant to AZT, and the benefits of early AZT were so transient as to be eliminated after three years of therapy. The U.S.-supported guidelines recommending AZT for all HIV-infected individuals with fewer than 500 CD4 cells had been badly discredited.
ACTG 155 was a particularly egregious example. At the time, based on laboratory results, researchers believed the best combination therapy approach was dual therapy with AZT and ddC. ACTG 155 took people who had been on AZT for more than six months and randomly assigned them to continue AZT, add ddC, or switch to ddC monotherapy. The results showed that people did no better if they added or switched to ddC than if they simply stayed on AZT. Instead of admitting that combination therapy failed in this AZT-experienced population, the ACTG 155 protocol team conducted a post hoc subset analysis and tried to show that combination therapy benefitted those who entered the study with over 150 CD4 cells. Unfortunately, the study was not designed to look at subsets in this way, and there were too few endpoints in this stratum to reliably demonstrate anything. In Berlin the investigators presented their post hoc subset analysis, and the activists from TAG and GMHC angrily denounced them for presenting an "intention-to-cheat" analysis. A customarily mild mannered Derek Link sprang to his feet in the stadium sized lecture hall and demanded of 155 presentor Margaret Fischl, "How much is Roche paying you to say this?"
Berlin was the most depressing AIDS conference ever. The bad news abroad was only highlighted by the contrast with events in Washington, where President Clinton signed the NIH Revitalization Act of 1993 on the same day that researchers presented Concorde and 155 in Berlin.
Four days later TAG founding member, TAGline co-editor, and composer Chris DeBlasio died. On September 23, Project Inform's Jesse Dobson, who had founded the Immune Restoration Think Tank, died in Oakland. On October 12, ACT UP/San Francisco's Andy Zysman, who had focused on accelerating research on AIDS related cancers such as Kaposi's sarcoma and non-Hodgkins lymphoma, died in San Francisco.
Researchers had previously ignored this area, but in the early 1990s it began to attract serious interest. NIAID held a workshop on the issue in summer 1993. Later that year TAG held a forum on the issue. In December the Institute of Medicine held a meeting on long-term survivors and those who were exposed to HIV, but uninfected. Slowly, NIH began to fund innovative research designed to elucidate the factors responsible for long-term non-progression and survival. This research would ultimately lead to discoveries such as the importance of HIV-specific CD4 and CD8 cells in protecting from infection and disease; the existence of genetic mutations which protected from HIV infection; and the existence of the HIV co-receptors CXCR4 and CCR5. These discoveries would lead to new drug candidates and new approaches to vaccine development.
Gregg's report on the basic science of HIV infection made a number of other recommendations which remain timely. He urged researchers to focus on physiologically relevant models of infection such as primate and human retroviral infection, rather than simply looking at the virus and T cells in laboratory dishes. He called for major increases in funding for basic, as opposed to NIH-directed applied, research grants. He called for an increased focus on immunology, new collaborative efforts, and a national network of tissue, cell, and specimen repositories to provide clinically relevant samples for interested researchers. He also cited the need for faster application of new ideas from laboratory research to the clinical setting.
In August, President Clinton nominated Harold E. Varmus as NIH Director. Varmus, who won a Nobel Prize for helping to discover oncogenes (genes associated with cancer), had opposed the OAR reforms in the spring. When he arrived at the NIH, however, he discovered a lack of coordination within the AIDS research program, and was obliged to support the administration's reform policy. He began a series of meetings with leading researchers from around the country to discuss what to do about AIDS, and convened a search committee to find someone to direct the OAR.
In 1992, however, I had moved over to the Primary Infection Committee, which dealt with anti-HIV drugs. Progress here was much slower. The pipeline was full of "me-too" AZT-like drugs such as ddI and ddC. Newer compounds such as non-nucleoside reverse transcriptase inhibitors (NNRTIs) and protease inhibitors were still in phase I trials. Other new approaches such as recombinant soluble CD4 (rsCD4) and a tat inhibitor had failed to live up to their promise.
The Primary Infection Committee was the powerhouse committee of the ACTG, and less susceptible to pressure or influence from activists. In addition, HIV was more difficult to study and treat than many of the OI microbes. Moreover, some researchers still had sore feelings from earlier activist campaigns which had overstated their rhetoric, such as ACT UP's campaign against "the Gang of Five" in 1990. "We're scientists, not alchemists," ACTG chair Larry Corey had told me, somewhat plaintively -- but correctly, in 1990 after a particularly contentious discussion about the failure of soluble CD4.
In 1992, the Primary Infection committee approved ACTG 229, a phase II study of Roche's protease inhibitor saquinavir, even though Roche had failed to define a maximum tolerated dose. I felt frustrated because my efforts to get the ACTG to address such issues were met with inaction. Eventually the failure to define the proper dose of saquinavir would lead to a lot of problems, and many people with HIV would develop resistance to other protease inhibitors because they had been given suboptimal doses of saquinavir.
In December 1993 I released a report on The Crisis in Clinical AIDS Research at the 17th ACTG meeting in Washington. It was a blistering critique of several poorly designed, conducted, and interpreted AIDS treatment trials, including the Army's phase I study of recombinant gp160, the ACTG 155 study of AZT/ddC in AZT-experienced persons, and the laboratory studies at Harvard which had led to the flawed ACTG 241 study of AZT, ddI and nevirapine in people who were AZT-experienced.
Nevirapine was the first non-nucleoside RTI to enter the clinic. Used as monotherapy, it had potent antiviral activity for a couple of weeks, until HIV mutated a single amino acid and became a hundred-fold less susceptible to nevirapine. In the test tube, however, combinations of AZT, ddI, and nevirapine appeared to suppress HIV replication completely: the now notorious "convergent combination" strategy. This led the ACTG to design the 241 study. Unfortunately, there were problems both with the lab work and with the study design. The lab work had been based on constructing mutations which inadvertently rendered HIV non-infectious: this had nothing to do with the triple therapy. The 241 study, like ACTG 155, was conducted in AZT-experienced persons; thus the benefit of combination therapy was lessened because of pre-existing AZT resistance.
Because studies like 155 and 241 suggested that combination therapy did not work in AZT-experienced people, it took longer than it should have to carry out similar studies in AZT-naive persons. Eventually a AZT/ddI/nevirapine study (called INCAS) would show that triple combination therapy, when used in AZT naive persons, could reduce viral load beneath the limit of quantitation in 50% of participants -- with associated clinical benefit. Unfortunately, these results were three years off and wouldn't be presented until the Vancouver AIDS Conference in 1996. It is tempting but ultimately futile to look back ten years on and wonder how many people could have been saved if highly active antiretroviral therapy -- with two nucleosides plus nevirapine, rather than plus a protease inhibitor, used in people before they had taken AZT as monotherapy -- had been discovered a few years earlier.