DHPG Blindness Drug Threatened?

On October 26 the FDA's Anti-infective Drug Advisory Committee voted 11-2 to recommend against approval of DHPG (also called Cytovene, or ganciclovir), on the grounds that the Committee did not have enough data to prove the drug effective for CMV retinitis. The FDA itself is not bound by the recommendations of such panels, but usually follows them. A final decision may be made in as little as two to three weeks.
Tests in thousands of persons, many published in medical journals, have shown beyond any doubt that DHPG is helpful in treating CMV retinitis, as well as CMV colitis and probably other systemic cytomegalovirus infections. No other drug has been approved for CMV retinitis. According to a Reuters news report on the controversy, CMV affects up to 98 percent of persons with AIDS, and if untreated will cause blindness or death in up to 25 percent of them. According to Syntex Corporation of Palo Alto, CA, the manufacturer of DHPG, untreated CMV retinitis inevitably leads to blindness.
Syntex has provided DHPG free to over 2,500 patients in a compassionate-use program, at a cost to the company of about 25 million dollars, and submitted the resulting data to the FDA. Because of serious side effects, mainly hematologic toxicity, the drug has been limited to those who urgently need it. It is usually given intravenously, but an experimental oral form has now been developed.
The Advisory Committee wants dose-response data on DHPG from prospectively controlled trials. It will probably take years to fund, set up, conduct and analyze this research, and gain approval for the result. Meanwhile Syntex will presumably be allowed to continue compassionate use of the DHPG. But there is fear in the PWA community that the drug could become more restricted or even unavailable, resulting in thousands of cases of unnecessary blindness or death.

Comment

DHPG has long been in a regulatory limbo. It is so certainly effective that it would be grossly unethical to do a
placebo study to prove it. Yet it is very hard to gain FDA approval without a double-blind study, which would usually
require a placebo if no substitute drug is available. DHPG has become close to a de facto standard therapy on an experimental, compassionate basis, a situation enormously expensive to Syntex, the manufacturer, which may not be able to bear the burden indefinitely. The result is a catch 22 in which the drug cannot be approved just because it so clearly works and has no substitute.
The FDA's Anti-infective Drug Advisory Committee, a group of outside scientists not employed by the FDA but selected and instructed by the agency, did not ask for placebo trials. It wants prospectively controlled dose ranging trials, presumably meaning that patients must be randomly assigned to different doses in a standardized procedure not always responsive to the physician's judgment of what is best for the patient. Such trials may mean further restricting access to the drug in order to force persons to volunteer.
Analyzing and reporting the results of compassionate use of the drug in over 2,500 patients--results also published in a number of medical journals--was not good enough, despite the fact that this experience showed beyond any doubt whatever that the drug is in fact effective.
The rejection of DHPG (like the similar rejection in May 1987 of TPA, a major advance in treating heart attacks after they have already begun) tells the pharmaceutical industry that the FDA disfavors compassionate use and will largely ignore any information gained from such practical experience. This policy will make it even harder to get manufacturers to make their drugs available to desperately ill patients. Compassionate use is already very expensive for drug companies, because of the Federal paperwork required. If they cannot use the resulting data toward gaining FDA approval, then they have no incentive to be involved.
Discouraging compassionate use and refusing to consider knowledge gained from it will also slow the development of new knowledge by tightening the most critical bottleneck in AIDS research--clinical testing to learn exactly when and how to use the many drugs already well known to science and medicine which appear very likely to be useful in the treatment of AIDS or ARC but are not being tested or brought into use.
One physician who has used DHPG pointed out that over 80 percent of patients treated with it improved or stabilized,
compared to under five percent of those left untreated. CMV in the urine became negative in over 90 percent. As for side effects, he has treated dozens of patients and never had to discontinue the drug permanently; sometimes he had to stop temporarily when certain blood counts became low. He also noted that today many more AIDS patients are outside of major medical centers, making it much harder for them to obtain treatment by compassionate use. The costs of taking care of these people will greatly increase, he pointed out, if we let them go blind. He also noted that the Advisory Committee's request for dose- response data is in fact a political euphemism for a placebo trial, as some of the doses would presumably be small enough to be ineffective; otherwise there would be no point in the study.
If the recommendation against approval was really so bad, why did the committee of experts vote for it? There seem to be several reasons. First, the clinical information available on DHPG is admittedly not clean or elegant science, because it comes from case reports not from a controlled study designed in advance. Second, the FDA's regulations require two prospectively designed controlled studies, unless there are very compelling reasons not to have them; with DHPG, for ethical reasons, there were none. Third, most of the panelists were not experienced with DHPG or eye diseases. For example, one suggestion was to use visual acuity tests to make sure DHPG worked. But in fact such tests would not be appropriate, because with CMV retinitis visual acuity depends entirely on whether the part of the retina
destroyed happens to include the macula, the small spot most important for vision; a visual field test should be used instead. The two votes for approval were from ophthalmologists familiar with DHPG. We don't know why the others put the letter of the law ahead of the views of those panelists with experience in the matter being decided.
There is no appeal from an FDA decision. And the public does not realize what is happening because those close enough to know are afraid to speak publicly. They fear with good reason that if they alienate the officials on whose decisions they depend, they will no longer be effective in accomplishing the goals they believe are important--whether gaining approval for a drug they are championing, or furthering any other project.
This situation will persist until Congressional committees, media, and private organizations realize that they will learn
what is going on only through long-term commitment to this investigation, with private, off the record conversations in
which scientists, physicians, and officials explain what is happening and what the problems are--without the well-founded
fear of damaging or destroying their work.