AIDS TREATMENT NEWS No. 071 - December 16, 1988
John S. James

For other serious diseases such as cancer, a system called compassionate has filled such gaps. But for reasons discussed below, a Federal consensus has recently developed that compassionate use of non-approved drugs should not be allowed except as part of scientific studies to determine whether or not the drugs work. While plausible in theory, this policy may cause severe problems in practice, problems which threaten not only the lives and health of patients, but continued progress in scientific drug research as well.

We have heard that the manufacturer of foscarnet is getting five to six calls a day from physicians and patients seeking that drug because for medical reasons they cannot use the only other available treatment, ganciclovir. Because of the recent movement against compassionate use, these patients can seldom get foscarnet, either--but for bureaucratic reasons, not medical. As a result, many of these people are now going blind.

Behind this situation is a bizarre and complex history -- and it includes good news as well as bad. This history touches on some of the most central issues in U.S. drug development -- not only on what is wrong, but also on steps now being taken to improve the process.

Exclusion from treatment for CMV retinitis could become a major, bitter, and confusing dispute. The people affected need to know what their options are. And they may need political support; therefore the larger community must be informed.

CMV Retinitis

CMV retinitis is an infection of the retina of the eye by cytomegalovirus (CMV). The opportunistic infection causes blindness if left untreated. Because damage can occur very rapidly, persons with AIDS should see a physician immediately if they notice blind spots, blurred vision, or loss of peripheral vision. Sometimes there are no symptoms, and the disease is discovered during a routine eye examination.

While it has not yet been documented, experts strongly suspect that the risk is much greater for persons with fewer than 50 T-helper cells. Since 15 to 45 percent of persons with AIDS will eventually develop CMV retinitis (according to a press release from Syntex Corporation, which has the most experience in treating this condition), such persons would probably benefit from a screening for the disease, or from routine eye check-ups from a private physician. Apparently CMV retinitis is becoming a larger problem because persons with AIDS are living longer.

CMV can also affect other organs besides the retina of the eye, such as the intestines or the lungs. It may also be a cofactor causing HIV itself to progress further. One study of autopsy reports found that half of the persons with AIDS who died had active CMV infections unsuspected while they were alive (see AIDS Treatment News #65, September 23, 1988).

The same treatments used for retinitis may also be effective for treating CMV infections elsewhere. A program for distribution of the drug ganciclovir (see below) intends to allow anyone with life-threatening CMV to obtain the drug, even if they do not have retinitis. For more information, call the ganciclovir study center (phone number below).

Ganciclovir (DHPG)

During the past three years, over three thousand persons with CMV retinitis have been treated with ganciclovir, which stops the progression of the disease and usually saves the person's remaining eyesight. Unfortunately ganciclovir has side effects, especially bone-marrow toxicity; therefore it usually cannot be combined with AZT, which has similar toxicity. Patients have often had to give up AZT in order to use ganciclovir to save their sight. Also, ganciclovir must be given intravenously.

Ganciclovir has not been approved by the FDA. But its developer, Syntex Corporation of Palo Alto, CA., has made the drug available for several years through compassionate use at no charge to the patient, at a cost to the company of over 25 million dollars, mostly in research and development expenses.

Unfortunately the widespread availability to ganciclovir by compassionate use led to a bureaucratic nightmare involving the manufacturer and the FDA. Everybody knew that the drug worked, but the only way to "prove" under the regulations that it did work would involve a trial in which patients were deliberately allowed to go blind, which everyone agreed was unethical. FDA rules insist that a drug must be shown to work through formal scientific trials. But Syntex could not ethically run such a study

It would be bad enough if a drug could not be approved in spite of the fact that it was known to work. But this situation was even worse. Ganciclovir could not be tested and approved because it was known to work. This knowledge itself meant that complying with the regulations necessary for approval would be universally recognized as unethical.

In an attempt to end this impasse, Syntex presented published articles and reports of thousands of cases to the FDA, asking for an exception in this case because it was so obvious that the drug was effective. But over a year ago, on October 26, 1987, an FDA advisory committee shocked the medical community by recommending against approval (see "DHPG Blindness Drug Threatened", AIDS Treatment News #44, November 6, 1987).

Despite fears that the company might then abandon the drug, it continued to make it available for compassionate use, while working with the FDA and the National Institutes of Health (NIH) on what to do next. But meanwhile, all other compassionate access to AIDS drugs became much more difficult, if not impossible. Companies were determined that what happened to Syntex would not happen to them. It is clear that many deaths have resulted from this side effect of the ganciclovir rejection; but it is difficult to address this fact politically because it would be difficult to prove which ones. Physicians seldom ask for access to a drug if they know they not get it. So in most cases no record of the rejection, and no audit trail of the public policy behind it, is ever made.

It is clear that the FDA feared that approving the drug without the scientific trials normally required would set a precedent of allowing companies to do an end run around the regular process of drug approval, by means of widespread compassionate use and the resulting political pressure. The FDA was determined that Syntex would not get away with this, would not receive approval without doing a trial. (The thousands of cases recorded under compassionate use did not constitute a formal trial, because these cases occurred as physicians treated patients, not according to a randomized protocol designed in advance by scientists to exclude potential errors such as biases in patient selection, or poor record keeping.) So the impasse remained.

In jurisprudence there is a saying that hard cases make bad law. In the case of ganciclovir, a public-policy stampede against compassionate use was triggered by an unusual situation. Scientifically designed trials are not only the right way to prove drug efficacy, they are usually the cheapest way too, because uncontrolled use requires far more patients to get convincing information. There should be no need to abolish compassionate use in order to force companies to do proper trials, because they have an economic incentive to proceed that way anyway.

But when ganciclovir was first discovered, Syntex and Burroughs-Wellcome were engaged in litigation over rights to the drug; Syntex eventually won. It appears that the companies were unwilling to invest in trials because they did not know whether or not they would have the rights to the drug.

But Syntex made ganciclovir available by compassionate use, to a few patients at first. Once it was clear that the drug did work, the company was stuck. The FDA refused to accept the drug without scientific clinical trials, and the company was ethically barred from conducting the trials required.

Oral Ganciclovir

Ganciclovir is always given by intravenous infusion, usually requiring daily visits to a medical center. An oral form has been created, however, and a successful but very preliminary human test was published over a year ago (Jacobson MA and others, Human pharmacokinetics and tolerance of oral ganciclovir, Antimicrobial Agents and Chemotherapy, volume 31 number 8, pages 1251-1254, August 1987.) As far as we know, no further testing has been done.

Syntex has applied to the FDA for an IND for permission to test oral ganciclovir. That IND is pending. Syntex will not make public when it first applied, but many months ago we had heard that a proposed trial of oral ganciclovir could not get an IND and therefore would not take place.

An aerosol form of ganciclovir has also been tested, and found effective against CMV lung infection in an experiment in animals. (Debs RJ, Montgomery AB, and others. Aerosol administration of antiviral agents to treat lung infection due to murine cytomegalovirus, The Journal of Infectious Diseases, volume 157 number 2, pages 327-331, February 1988).

The New Treatment IND and Clinical Trial

Recently a new attempt to resolve the problem was announced by the National Institute of Allergy and Infectious Diseases (NIAID, a branch of the NIH), the FDA, and Syntex Corporation.

Under the plan, which will replace the compassionate use system in effect until now, everybody who has already been using ganciclovir will be allowed to continue receiving it. New patients will be examined by an ophthalmologist to determine whether the CMV lesions are near the center of the field of vision and therefore immediately sight threatening. Those with immediately sight-threatening disease will be allowed to receive ganciclovir under a system called the "treatment IND", new rules for earlier access to experimental drugs approved in mid-1987 but seldom put into effect since then ("IND" stands for "investigational new drug").

Those with CMV retinitis which is not sight threatening will instead be directed to a controlled clinical trial. No placebo will be used, but patients will be randomly assigned either to receive immediate treatment, or to receive later treatment if there are signs of deterioration (allowing them to remain on AZT longer if appropriate).

Among persons with AIDS and physicians we have heard from, this plan has been greeted with mixed feelings. On the negative side, it reduces treatment options for those with less serious disease. We have also heard medical questioning of specifics of the study design.

The overall plan (the treatment IND combined with the clinical trial) may be a brilliant and partly successful effort to find a medical reason for a study really being done for regulatory-political purposes. The ostensible question is whether it is better to start ganciclovir immediately or wait in cases of non-sight-threatening disease; but what the study is really trying to prove is that Syntex and other manufacturers cannot get away with substituting compassionate use for clinical trials.

Yet despite these negatives, we see the new ganciclovir plan as possibly a watershed advance, moving toward a new compromise and consensus on how to reconcile the scientific requirements of good studies with the needs of patients and their physicians.

(1) In the past, clinical trials usually required eligible patients to stop other treatments and then risk getting a placebo. And persons who did not meet the scientific entry criteria for the study were simply abandoned. By contrast, this plan for ganciclovir serves everyone who needs the drug -- and takes patients' needs into account.

(2) In the past, trials of AIDS treatments typically used death, pneumocystis, or other severe deterioration as "endpoints" of the study. Patients have often been subjected to bad medical management--such as being required to stop aerosol pentamidine--in order to make these endpoints more likely for the placebo arm, increasing the power of the trial to tell whether the drug worked better than no drug. But here, the patients not given ganciclovir will receive "very close monitoring for clinical deterioration", according to an FDA press release announcing the new ganciclovir treatment IND and clinical trial--presumably to take them out of the study and give them the drug before serious damage is done.

(3) The ganciclovir plan sheds light on what we believe is a hidden but awesomely important impediment to the overall success of clinical trials, as well as to compassionate treatment access by patients. We call this problem the "mandate for ignorance" which follows from certain ethical requirements on placebo-controlled or other randomized trials, as they are now designed. (See "Drug Trials: A Mandate for Ignorance", below.)

For More Information on Ganciclovir

For information about entering the ganciclovir program, physicians and patients can call the Ganciclovir Study Center, 301/497-9888. There is no charge to patients, as costs are paid by Syntex Corporation.

Foscarnet

Foscarnet, often used in Europe as a first choice for treating CMV retinitis because it does not require that the patient stop using AZT, may also have some effectiveness against HIV. Its manufacturer, the Swedish company Astra, had intended to test the drug as an AIDS treatment in the U.S. But according to an editorial in The Wall Street Journal (April 21, 1987), Astra left those trials due to frustration with bureaucratic obstacles to AIDS research in the U.S.

Foscarnet, like ganciclovir, must be given intravenously. It does not cause bone-marrow toxicity like ganciclovir. But it can cause kidney toxicity, which may be reduced by giving the patients enough water, sometimes by intravenous infusion.

Published reports of European experience in using foscarnet for CMV retinitis are mixed. It seems clear that the drug does stop the disease during a high-dose "induction" phase of the treatment. But patients are then put on a lower "maintenance" dose, and relapses during this maintenance phase have often been reported. According to information from the National Eye Institute, a branch of NIH, a 70-85 percent rate of favorable response has been reported.

NIH is now running two different studies of foscarnet. One, by the National Eye Institute, was reported in AIDS Treatment News November 4, 1988. This study requires trips to Bethesda, MD, near Washington, DC, but all expenses are paid.

The other NIH study is a multi-centered trial taking place in San Francisco, New York, and Los Angeles. In San Francisco (and probably the other locations also), these studies are recruiting patients now who have not been treated with ganciclovir or foscarnet before. The main advantage of foscarnet over ganciclovir is that patients can continue using AZT; foscarnet is also believed to have some anti-HIV activity. Patients in the San Francisco area interested in enrolling in this study can call the protocol coordinator at 821-5139.

Both these trials are excluding patients who have previously used ganciclovir. NIH does plan to start a multi-center "salvage" protocol for such patients, hopefully early next year. Meanwhile, for patients who have failed ganciclovir the only access to foscarnet at this time is in Houston, TX; patients must pay their own travel, lodging, and hospitalization expenses, unless their insurance will cover part of the cost.

One other group of patients may also be left out--those who are using ganciclovir successfully, but should be on foscarnet instead so that they could use AZT. We brought this potential problem to the attention of the group designing the protocol.

For those who fall between the cracks, compassionate use of foscarnet should be available. But the FDA recently led the charge against such access. NIAID is trying to devise plans which can include everyone. When all goes well, these plans may do their job. But when things go wrong and studies are delayed, the current policy against compassionate use makes the patients pay.

It seems clear that foscarnet is not a miracle drug, although it will probably have an important role in treatment. European physicians are using it to treat CMV retinitis, often as first choice over ganciclovir. When nothing else works, U.S. patients and physicians should have the option of using it.

Other Treatment Possibilities

Besides ganciclovir and foscarnet, other possible treatments have been suggested. We have not been able to research these fully by press time.

Acyclovir. This drug may have some effect. NIH plans to study a combination of acyclovir and AZT to treat CMV retinitis.

Anti-CMV antibodies. Passive immunotherapy using donors with high CMV antibody levels was described as producing conflicting results for CMV disease (Reed, H.C. and Meyers, J.D., Treatment of Cytomegalovirus Infection, Clin. Lab. Med. (USA), volume 7 number 4, pages 831-852, 1987). But several European articles describe the use of 'Cytotect' from Biotest Pharma, a product containing CMV-specific hyperimmune globulin, for prevention or treatment of CMV. The option of using this treatment should be available to U.S. physicians.

(S)-HPMPA or (S)-HPMPC. These new drugs are highly active against CMV in the test tube. We do not know of any human use. For background, see De Clercq E and others, A novel selective broad-spectrum anti-DNA virus agent, Nature, volume 323 number 2, pages 464-469, October 1986. Or see De Clercq E and others, Antiviral activity of phosphonylmethoxyalkyl derivatives of purine and pyrimidines, Antiviral Research number 8, pages 261- 272, 1987.

BHT. Over two years ago, AIDS Treatment News reported interest in BHT as a possible anti-AIDS treatment (issue number 10, August 15, 1986). BHT, used in small amounts as a food preservative, has been an underground herpes treatment used by hundreds of people during the last several years; one clinical trial found some effectiveness in treating herpes in humans. One laboratory report found that BHT was highly effective against CMV in the laboratory (Kim KS and others, Inactivation of cytomegalovirus and semliki forest virus by butylated hydroxytoluene, The Journal of Infectious Diseases, volume 138 number 1, page 91-94, July 1978).

BHT must be used carefully, because large doses are toxic. Even lower doses involve some risk. Letters in medical journals have reported several cases of toxicity, especially gastrointestinal problems, from doses commonly used to treat herpes.

We have not systematically researched BHT since writing the August 1986 article. But recently we talked to a person with AIDS in San Francisco, who has used BHT as his only treatment for CMV retinitis, and is convinced that it did slow although not stop the progression of the disease.

This patient started using BHT in 1986 or early 1987. In January 1988 he stopped temporarily, for about six weeks, following a doctor's advice; during this period without BHT he developed CMV retinitis, which "started galloping"; he saw new blind spots daily. He went back on BHT, using a very large dose for about six weeks, four grams per day dissolved in oil and taken rectally; he reported an instant "high" from the mental effect of this dose. After six weeks, he went back to smaller doses, about two grams per day used orally (still a large dose). In the eight months since, there has been slow deterioration of vision; he can still read, but with difficulty. While this person is convinced that BHT has worked partially for him, and would not dare discontinue it again, he also emphasizes that there is no proof that it has helped.

AIDS Treatment News is preparing an information packet on BHT. For a copy, send a self-addressed stamped envelope to: AIDS Treatment News, P.O. Box 411256, San Francisco, CA 94141; be sure to mention BHT.

Toward a Way Out

Recently a top FDA official suggested an idea for reconciling the need for scientific, rigorous studies with the need of patients for access to treatments.

The approach is to have a small, tightly controlled scientific trials with rigorous entry criteria. But in addition, as a condition for approval to conduct a study, the company would also make the drug available on a compassionate basis or "treatment IND" to other patients who needed it--with careful monitoring and reporting, both to enhance patients' safety and also to collect more information about the drug.

What we like about this proposal, which is similar to what is already being started with ganciclovir, is that everyone receives treatment, no one is abandoned. (Placebo studies are now becoming less common, with controlled trials usually comparing one treatment option against another, not against no treatment. It is true that controlled studies reduce the patient's power to choose, since for scientific reasons patients must be assigned at random to the different treatment groups. If there are not enough volunteers, a lottery might be necessary to decide who goes into the controlled trial, and who receives treatment through the compassionate-access, monitoring study.)

This approach will not solve all the problems. But it may allow researchers to conduct tightly controlled studies without sacrificing vital interests of any patients, whether in the study or not. It would certainly improve on the present system, in which treatment is denied in order to force patients into ill-designed trials, with those who don't qualify being abandoned.

881216
ATN07101

AEGiS presents published material, reprinted with permission and neither endorses nor opposes any material. All information contained on this website, including information relating to health conditions, products, and treatments, is for informational purposes only. It is often presented in summary or aggregate form. It is not meant to be a substitute for the advice provided by your own physician or other medical professionals. Always discuss treatment options with a doctor who specializes in treating HIV.

Copyright ©1980, 1988. AEGiS. All materials appearing on AEGiS are protected by copyright as a collective work or compilation under U.S. copyright and other laws and are the property of AEGiS, or the party credited as the provider of the content. .