The Body: The Complete HIV/AIDS Resource
Follow Us Follow Us on Facebook Follow Us on Twitter 
Professionals >> Visit The Body PROThe Body en Espanol
  • Email Email
  • Printable Single-Page Print-Friendly
  • Glossary Glossary

GMHC Treatment Issues
January 1997


Medical Marijuana: Research Priority, Hoax or Civil Right?

by Elisa Nelan

Medical marijuana, long a subject of interest to HIV-affected persons, has become the subject of broader national interest since the elections last November and the passage of two key voter initiatives, Propositions 215 and 200 in California and Arizona, respectively. The propositions reflect popular belief that anecdotal reports of the benefits of medical marijuana warrant legal permission for that use. Heated debates continue between proponents and opponents of medical marijuana at the community, state and national political levels. Although local and state drug laws vary, much of the current debate will be waged in the federal arena. The Clinton Administration has already threatened to penalize doctors who recommend marijuana; medical marijuana advocates in response are discussing First Amendment legal action; and as a result, the ultimate consequences of the California and Arizona voter initiatives remain clouded.

The federal government's anti-marijuana stance was first enunciated in 1937 (on the heels of Prohibition), when the Marihuana Tax Act made marijuana illegal for non-medical use. That law made the procurement of marijuana even for medical purposes so difficult that it was effectively removed from the pharmacopoeia. Marijuana ultimately was designated a Schedule I drug, meaning among other things that it lacks an accepted medical use and is unsafe even under medical supervision. Although there have been continuing reports of marijuana's use by people suffering pain and complications from such maladies as cancer, glaucoma and, more recently, AIDS, there has been a consistent official denial ever since of any potential medical benefit. This political atmosphere has largely prevented scientific research to test the benefits of smoking marijuana, despite research proposals by well-respected scientists and advocates. One concession on the part of the government was the creation in 1976 of a compassionate use program that permitted a few people (reportedly fewer than ten) to receive marijuana for medical use. By 1990, with the burgeoning of the AIDS crisis, the number of applicants to the program swelled. But in 1992 the Public Health Service totally dismantled the program.


Enter Marinol

Marijuana's well known anti-nausea and appetite-stimulating effects did lead to some study of the drug as an anti-emetic and appetite stimulant in people with AIDS or cancer suffering from wasting.1 Since most of this effect was attributed to tetrahydrocannabinol (THC), the psychoactive component and best studied metabolite of marijuana, a synthetic version was designed. In clinical studies, synthesized THC was shown to increase appetite and help people gain weight.2,3,4,5,6 These findings led to the approval of the pharmaceutical grade THC, dronabinol (Marinol), in 1985 as a treatment for nausea and vomiting related to chemotherapy in people who failed on conventional anti-emetic treatments. Subsequent studies in people with AIDS led to the December 1993 FDA approval of Marinol for AIDS-related anorexia associated with weight loss.7

The key study that led to the AIDS indication was a double-blind study involving 139 persons with AIDS randomized to take either Marinol or placebo. Those taking Marinol were said to experience increased appetite, decreased nausea, and "trends toward improved body weight and mood." Long-term follow-up of 94 of these participants, who took open-label Marinol for approximately one year, indicated that appetite improvement was sustained over time. However, 17% dropped out of the study because of such side effects as dizziness, muddled thinking and drowsiness.8

Developers of Marinol tout the drug for its " purity" relative to marijuana, which contains many components other than THC. Yet it is precisely some of those other components or cannabinoids that certain researchers are interested in studying. Preliminary evidence suggests that the other cannabinoids help confer the benefits experienced by people who smoke (or eat) marijuana compared to those taking Marinol. There are no actual human trials, though, to confirm this hypothetical superiority of marijuana over Marinol. Ironically, the manufacturers of Marinol note that there have been over 25 human clinical trials of Marinol but no clinical trials of smoked marijuana's safety or efficacy, apparently implying that this lack is an indication of Marinol's superiority.9

Marijuana vs. Marinol

Marinol itself is not for everyone. According to Lester Grinspoon, M.D., of the Harvard Medical School and author of Marihuana: the Forbidden Medicine, (Yale University Press, 1993) " Marijuana has distinct advantages over Marinol, for a number of reasons. First, Marinol must be swallowed. Obviously, some people with nausea and vomiting, one of the indications for Marinol, may be unable to ingest it. Second, Marinol is subject to huge bioavailability variations, between individuals, but also shifting day to day for the same person. It's very difficult to define appropriate doses for the individual. Third, the therapeutic value of marijuana is dependent on many other cannabinoids than THC (which is all Marinol is). In my experience, people given a choice consistently choose marijuana over Marinol. Fourth, marijuana has an immediate effect, which is what people with migraines or epilepsy, for example, want. In contrast, the effects of Marinol may not be felt for one to two to three hours after taking it. Fifth, people smoking marijuana rarely report anxiety, whereas people taking Marinol more frequently do. This is because THC taken alone leads to anxiety. Cannabidiol, which one ingests along with THC and other cannabinoids when consuming marijuana, has an anti-anxiety effect. Finally, Marinol costs significantly more. Even at the street price inflated by what I call the prohibition tariff, marijuana is much less expensive. If it were government-approved, it would become far less expensive than it is presently; I estimate that an ounce might cost $25-30, so that a marijuana cigarette would be about 30 cents. And this is actually one of the problems: marijuana would never be a money-making proposition, and, since it would replace more expensive pharmaceuticals, the pharmaceutical industry stands to lose.

" In the new edition of our book Marijuana: the Forbidden Medicine, we list about 30 medicinal uses for marijuana. For example, the new anti-emetic drug ondansetron costs $120 to $160 for oral administration but, because of nausea and vomiting, often requires IV administration which raises the cost to $600. If marijuana were legal and available for use with chemotherapy, you could prescribe someone to smoke a marijuana cigarette about 20 minutes before their cancer treatment -- at a cost of $0.30. That would prevent nausea just as well. Even if marijuana therapy was reimbursable under a patient assistance program, the extreme low cost would hardly endanger payment for other treatments." 10 Although Marinol is generally well tolerated, some people report intolerable levels of psychotropic side effects, i.e., an unacceptably " stoned" feeling. To ameliorate this effect, there have been suggestions that an anti-emetic drug called prochlorperazine could be used in combination with Marinol.11 However, many people report that the individualized dosing and administration schedule permitted by smoking marijuana easily prevents the ill effects associated with Marinol. Still other people with AIDS, who cannot take Marinol or other oral antinausea medications because of gastrointestinal difficulties (like the severe vomiting giving rise to the need for the medication in the first place), report that they can better achieve the desired antinausea benefits from smoking.

Marijuana, Wasting and TNF

Although people with HIV/AIDS have used marijuana for diverse reasons, including relief of pain and nausea, one of the most prominent reasons for use is appetite stimulation, especially by those with wasting. However, wasting or cachexia is not simply a loss of weight or appetite; various factors and mechanisms contribute to the syndrome. One of the most significant factors in AIDS-related wasting is a disproportionate loss of lean body mass (muscle). A syndrome with a complex etiology, wasting may occur even in spite of an apparently adequate dietary intake. Recombinant human growth hormone (Serostim, manufactured by Serono) is the only approved drug that has been shown to increase lean muscle mass (with the possible exception of anabolic steroids), but its exorbitantly high price currently renders it a non-option for many HIV-infected people.

Tumor necrosis factor alpha (TNFa), which the immune system produces as part of the inflammatory response to infection, plays a particular role in the development of wasting -- by contributing to lean body mass destruction -- and in a number of other chronic disease-related conditions.12,13,14 TNFa also stimulates HIV proliferation.15

There are no studies that establish that reducing TNFa levels, which are notoriously difficult to measure, will be beneficial in people with HIV. Researchers nevertheless have been pressing ahead to find ways to reduce those levels. Among other agents, THC has been shown to reduce TNFa in the test tube. In a study involving macrophages exposed to the excitory cytokine interferon gamma (IFNg) along with bacterial lipopolysaccharides (LPS), the addition of THC to the mix was shown to impair macrophage activation. The macrophages altered their responses to normal "triggers" like IFNg and LPS by failing to carry out their normal sequential expression of various proteins. One of the proteins the THC-exposed macrophages were unable to express was TNFa.16

Another study looked specifically at the effect of " physiologically relevant" concentrations of THC on TNFa synthesis by human large granular lymphocytes (LGL), a type of white blood cells Researchers found that THC decreased two primary LGL functions, TNFa production and anti-tumor activity.17

But other components of marijuana may also be involved in reducing undesirable inflammatory factors. Cell culture studies show that the essentially nonpsychoactive cannabidiol (CBD), in particular, potently reduces levels of TNFa as well as IL-1,18,19 another inflammatory white blood cell secretion involved in wasting and HIV replication.20 Watzl and others studied the in vitro effects of both psychoactive and nonpsychoactive components of marijuana on white blood cells. They found that concentrations of THC and CBD comparable to what would be found in the blood after smoking marijuana decreased cells' secretion of IL-1 and TNFa, increased IFNg and had no effect on IL-2. The effect of these other marijuana constituents on TNFa especially provides a rationale for the clinical evaluation of marijuana in people with AIDS-related wasting.

Objections to Smoking

An often raised concern is the potential hazards of smoking any substance. Donald Kotler, M.D., an expert on AIDS-related wasting at New York's St. Luke's Roosevelt Hospital, feels that medical marijuana use by people with AIDS may be helpful, but cautions that research into the possible health hazards of smoking in people with AIDS (e.g., the risk of pneumonia), is highly desirable. Dr. Grinspoon points out that it is not exposure to THC and other cannabinoids that is at issue, but rather the burning of the vegetative or plant matter. The combustion process creates problems with noxious substances like tar and carcinogens, whether tobacco or marijuana is smoked. He emphasizes that the average medical marijuana smoker would be exposed to far fewer toxic byproducts, relative to a tobacco smoker, since she or he would be smoking very few " cigarettes."

" Doctors don't like smoking, " said Dr. Grinspoon, " but it would be a simple matter to develop a vaporizing gadget for medical marijuana use that would obviate smoking-related hazards. And, speaking of financial incentives, unlike marijuana itself, which no one can patent, they definitely can patent medical devices developed to better administer it."

Conducting Further Research

The time is ripe for legitimate scientific research into these questions. Perhaps more than any other US researcher, Donald Abrams, M.D., of San Francisco is intimately acquainted with the difficulties of actually accomplishing this goal. His proposed study of smoked marijuana in people with AIDS has been stymied by the Drug Enforcement Agency and National Institute of Drug Abuse (NIDA),21 and the story of Abrams' attempts and the government's responses has been widely publicized over the past several months (for background, see Treatment Issues, May, 1995). His research protocol for a pilot study of smoked marijuana, approved by the Food and Drug Administration and Dr. Abrams' local Independent Review Board (IRB), was blocked by NIDA in April, 1995 and again this summer by a National Institutes of Health grant review committee. That study may get another hearing. An addiction research team at Brown University is reviewing the rejected research grant application, and collaborative meetings soon will be held to discuss strategies for meeting a May, 1997 reapplication deadline.

The federal government meanwhile continues with its anti-marijuana strategy, but that strategy now appears to be subtly softening. Clinton Administration " Drug Czar" Barry McCaffrey, a retired Army general, only this month backpedaled from his previous assertions that medical marijuana was a " cruel hoax" with no medical benefit to announce that the federal government will spend $1 million dollars on a meta-analysis of existing research on potential benefits. At a January 13 press conference, Gen. McCaffrey went further, stating, " There's an open door to approval for any substance that provides proven benefits."

With all the marijuana studies so far conducted in the test tube rather than in human beings, one wonders how Barry McCaffrey's meta-analysis could possibly prove marijuana's benefits. Dr. Grinspoon believes that the government cannot ignore established anecdotal evidence, especially the vast amounts that exist for marijuana. But others feel that many questions remain that can only be resolved by further research. Dr. Kotler, for one, thinks that marijuana may be good medicine for some people, but adds, " The word of the day is combination therapy. If things aren't going well nutritionally, then relying on marijuana is probably not a comprehensive enough plan of attack." Exactly how marijuana fits into an overall regimen against HIV infection and its manifestations is still undetermined.


1 Doblin RE and Kleinman MA. Journal of Clinical Oncology. July, 1991:9(7): 1314-9.
2 Beal JE, et al. Journal of Pain & Symptom Management. February, 1995:10(2): 89-97.
3 Lane M et al. Journal of Pain & Symptom Management. August 1991:6(6):352-9.
4 Cat LK and Colemann RL et al. Annals of Pharmacotherapy. May 1994:28(5):595-7.
5 Nelson K et al. Journal of Palliative Care. Spring 1994 19(1):14-8.
6 Gorter R et al. AIDS. January 1992; 6(1):127.
7 Roxane Laboratories, Inc. Marinol product information.
8 Beal JE et al. IX International Conference on AIDS. June 1993; PO-B36-2354.
9 Roxane Laboratories, Inc. op cit.
10 Grinspoon L. Personal communication. January 1997.
11 Plasse TF et al. Pharmacology, Biochemistry and Behavior. November 1991; 40(3):695-700.
12 Gearing AJ et al. Nature. August 18, 1994; 370(6490):555-7.
13 Llovera M et al. Journal of the National Cancer Institute. August 18, 1993; 85(16):1334-9.
14 Espat et al. Surgical Oncology. October 1994;3(5):255-62.
15 Goletti D et al. Journal of Virology. April 1995;69(4):2540-6.
16 Cabral GA and Fischer-Stenger K. Life Sciences. 1994; 54(23):1831-44.
17 Kusher DI et al. Cellular Immunology. March 1994;154(1):99-108.
18 Watzl B et al. International Journal of Immunopharmacology. 1991;13(8):1091-7.
19 Watzl B et al. Advances in Experimental Medicine and Biology. 1991;288:63-70.
20 Strassman G et al. Journal of Immunology. March 15; 150(6)2341-5.
21 Abrams D et al. The New England Journal of Medicine. September 7, 1995;333(10):671.

NIH Panel Examines Consequences of AZT in  Pregnancy

by Saundra Johnson

Conflicting data from two studies of AZT's ability to induce tumors in mice exposed to the drug before birth has raised considerable consternation and confusion at the National Institutes of Health (NIH). Since 1994, the Public Health Service (PHS) has advised administering AZT to pregnant HIV-positive women and their babies to reduce the chances of mother-to-child transmission of the virus. This advice was based on the outcome of the NIH-sponsored trial ACTG 076. On January 14, the NIH convened a panel to decide whether the new mouse trial data was reason to alter the previous guidelines. The panel's goal was to bring together basic and clinical scientists, bioethicists and mothers with HIV to scientifically review and interpret the data and develop recommendations for publicizing the latest information and for future research.

The two studies were very different in concept and result, and these differences made drawing inferences from the data very difficult.

One study, conducted by the National Cancer Institute (NCI), was designed to potentiate any tendency of AZT to cause tumors. Mice received doses of AZT near the maximum tolerated dose. There was a significant dose-related increase in tumors found in the lungs of both male and female mouse offspring and in the livers of the males. Also, seven of the study's 46 AZT-exposed female mice experienced reproductive tract tumors versus none of the 30 female control mice. (But note that incidence of hematopoietic cancers -- lymphomas, leukemia and the like -- was dramatically lower in the male and female mice exposed to AZT in utero.)

The other mouse study, which was carried out by AZT's manufacturer, Glaxo Wellcome, was designed to replicate the dosing rate and schedule used in ACTG 076 and look at carcinogenic effects in offspring over time. This study found no increase of tumors following the pregnant mice's exposure to AZT.

After presentations to the NIH panel by the two studies' investigators, a discussion ensued over whether cancers seen in mice are predictive of those in humans under the same conditions. There are only a few drugs that have been tested in this way, and of these, DES is the only one on which we have long term follow up. The offspring of mice given DES developed tumors similar to the tumors seen in the children of women given DES to prevent miscarriage. It is believed that DES and AZT cause cancer in different ways, however.

The panel was unanimous in concluding that the benefits of AZT in preventing perinatal transmission appear to far outweigh the concerns raised by the NCI mouse study. It considered that the data in and of itself does not suggest the need for changes in treatment recommendations or public health policy. The panel then drafted several research, clinical, public information and policy priorities that tempered its position in support of AZT:

  • Reassessment of the PHS guidelines in the broader context of the rapid change in state-of-the-art treatment of women with HIV since the guidelines were first developed and published (single agent versus combination therapy).

  • Improvement in public awareness of the industry-sponsored, and now under-used, pregnancy registry of those exposed to antiretroviral drugs before birth and aggressive monitoring of all such exposed children.

  • Greater emphasis on studying perinatal interventions that maximize the safety and minimize the likelihood of long-term side effects.

  • Completion of the full two-year follow-up of the mice in the NCI study.

  • Additional research on transplacental carcinogenesis of nucleoside analogs like AZT, including investigation into the tumor-inducing mechanism and its relationship to the amount of AZT administered (with results confirmed in at least one other species).

  • Determination of the relationship between AZT pharmacokinetics and AZT incorporation into blood cells and solid tissues.

  • Full disclosure to women of the theoretical risk of cancer in their offspring during counseling about interventions to reduce perinatal HIV transmission.
As Treatment Issues went to press, Eric Goosby, Director of the Office of HIV/AIDS Policy, announced that a date would be set in the coming week for reconvening the original panel to review the current PHS guidelines. A formal statement will be sent to all women involved in perinatal transmission trials and to the PHS for dissemination to physicians.

Contradictory Roles for GM-CSF

by Theo Smart

Granulocyte-macrophage colony-stimulating factor (GM-CSF) is used to raise low neutrophil cell counts (neutropenia), a frequent drug- or AIDS-related condition that impairs the body's ability to ward off infection. But there have been reports that GM-CSF increases HIV activity as well as boosting cell proliferation. Such findings have led many clinicians treating neutropenic patients with HIV to use a competing drug, G-CSF, which does not stimulate HIV.

But if GM-CSF is too dangerous to use in people with HIV, why has Dr. David Ho proposed (last November at the Third International Conference on Drug Therapy in HIV Infection in Birmingham, England) conducting studies that would throw it in the mix with triple drug antiretroviral therapy? And why is Immunex, the Seattle company that markets GM-CSF in the U.S., now enrolling a very large study using this drug in people with late-stage AIDS who are not neutropenic? The reasons are somewhat unconventional and have little relation to each other. The results of these studies may determine whether GM-CSF is to have a supporting role in the treatment of HIV infection when combined with the new highly active antiretroviral therapies or a larger role as broad opportunistic infection preventive for advanced disease.

Fixing the Phagocytes

GM-CSF belongs to a family of cytokines called colony-stimulating factors that promote the production of various blood cells in the bone marrow. GM-CSF regulates the development of neutrophils, macrophages, monocytes and eosinophils. The other CSFs include granulocyte-CSF (G-CSF), which specifically induces neutrophil production, and macrophage-CSF (M-CSF), which stimulates macrophages and monocytes. The pharmaceutical versions of these natural immune cell hormones are produced by genetically altered yeast, E. coli or mammalian cells.

The use of CSFs to treat neutropenia in cancer patients has been long established. Since neutrophils play a major role in fighting microbes such as bacterial, fungi and parasites, patients with neutropenia are particularly susceptible to bacterial and disseminated fungal infections. In a study of GM-CSF in neutropenic cancer patients, the agent was shown to reduce mortality caused by fungal infections. 1

Neutropenia is fairly common in people with AIDS. It may be caused by a number of factors including HIV infection itself, opportunistic infections, nutritional deficiencies (vitamin B12, folate) or treatments such as AZT, Bactrim/Septra, ganciclovir and cancer chemotherapy. Both G-CSF and GM-CSF have been shown to increase neutrophil count in people with HIV, but a number of laboratory studies suggest that both GM-CSF and G-CSF do more than encourage the proliferation of neutrophils. There is a substantial body of data showing that these cells are functionally impaired in people with HIV, 2,3,4,5 although just how these deficits occur is not fully understood. The CSFs have shown the ability to restore the function of these damaged cells, enhancing their phagocytic activity, microbe killing ability and antigen presentation. Specifically, GM-CSF-treated neutrophils, monocytes and macrophages have exhibited increased activity against bacteria, HIV, fungi, MAC, TB, cryptosporidia and Pneumocystis carinii.

Thirty-five HIV-infected individuals treated with GM-CSF for a mean of 9.7 weeks had no major opportunistic infections while on treatment in one report. 6 Another GM-CSF study involved 32 KS patients with chemotherapy-induced neutropenia. No bacterial infections occurred, and only 19% of the patients developed an opportunistic infection over the median seven-month duration of the study.7

A few small GM-CSF studies have been carried out in people with HIV who were not neutropenic. Two of four AIDS patients with resistant esophageal candidiasis experienced a complete response in a study of GM-CSF at 150 mg/kg of body weight per day in combination with itraconazole or amphotericin. One participant had a partial response, and the other, who had several other AIDS-related complications, died before receiving two weeks of treatment with GM-CSF.8 Improved anti-mycobacterial activity was seen with GM-CSF in one MAC study,9 and better clinical and radiologic responses were seen in pediatric AIDS patients randomized to receive GM-CSF or nothing in combination with standard TB therapy.10

Immunex's Gamble

All the laboratory data and the small pilot studies have encouraged Immunex to run a large opportunistic infection prevention study in over 500 people who are not neutropenic or only mildly so. The six-to-twelve month multi-center trial will enroll very advanced patients with a CD4 count less than 50, or less than 100 if they have had a prior AIDS-defining illness (other than KS). Study participants will receive 250 mg of GM-CSF or placebo three times a week. Given the possibility that GM-CSF may increase viral load, study subjects are expected to take antiretrovirals (the study will be discontinued if a ten-fold increase in viral load is observed after three months). Other immunomodulators are forbidden.

The study's primary endpoint will be the total number of opportunistic infections over time, including recurrent and bacterial infections. It will also evaluate the time to progression of disease, survival, incidence of specific opportunistic infections and quality of life. At the higher doses used to treat neutropenia, GM-CSF can cause adverse reactions such as fever, bone pain, joint and muscle pain, headache and other flu-like symptoms. Analgesic or anti-fever medication usually can control such side effects. (For more information on this study, which is open to enrollment, call 800/TRIALS-A.)

Immunex is taking a big chance. For one thing, a 500 patient study is an expensive proposition. Furthermore, although GM-CSF may improve microbe killing over the short term in a cell culture dish, what is going on in the body of a person with HIV is extremely complex. GM-CSF administration may very well increase the level of other cytokines known to have harmful effects, such as tumor necrosis factor alpha.

The company additionally is conducting two studies in HIV-positive volunteers with oral candidiasis unresponsive to fluconazole. The first study will compare continuing on fluconazole alone to the effect of adding daily GM-CSF to the fluconazole. In the second study volunteers will be switched to intravenous amphotericin B with or without GM-CSF. (The GM-CSF will be administered as a 250 mg/day subcutaneous injection in both studies.) These two trials are still on-going, but already, unofficial sources report that the fluconazole trial so far is not detecting any positive benefits to adding GM-CSF.

Purging the Reservoirs

Immunex is not convinced that GM-CSF substantially contributes to viral load. Even if it does, though, the company believes that this danger could be offset by the improvement in immune function elicited by GM-CSF. In contrast, Dr. Ho foresees using GM-CSF in conjunction with the potent new anti-HIV combination therapies precisely because of its HIV-stimulatory effect.

Current anti-HIV therapies only affect new virus particles produced in cells with active HIV. Administering agents known to excite cells latently infected by HIV might provoke them into viral production. Cytokines like IL-2 might shorten the time to clearance of HIV from the pool of quiescent CD4 cells, and factors like GM-CSF would increase the turnover in the reservoir of latently infected monocytes and macrophages.

The data on whether GM-CSF stimulates HIV is somewhat equivocal. Most studies show that it does promote HIV,11,12,13 but a number of cell culture studies suggest that it inhibits HIV replication in macrophages. An Australian study reported that GM-CSF blocked viral replication in macrophages sometime after reverse transcription in a dose-dependent fashion.14 A Japanese study published about a year ago found that certain steps of HIV-replication were inhibited in macrophages stimulated by GM-CSF, and that HIV did not replicate well in these cells. The authors concluded that "the suppression of HIV replication in GM-CSF-induced macrophages may provide a model of the latency of HIV infection in vivo."15

Some of GM-CSF's effect may be time-dependent, or, as Immunex has argued, moderated by the immune activation that GM-CSF incites. A study presented in Washington last year at the Third Conference on Retroviruses and Opportunistic Infections suggested that HIV-infected monocytes and macrophages produced 15-fold more virus particles when activated by GM-CSF, but there was no increase in the rate of new cell infection when compared to controls.16 A study using SIV also found that GM-CSF treated cultures produced more viral material, but after controlling for the increased number of cells, there was no increase in SIV production on a per cell basis.17

he few studies that report that GM-CSF upregulates HIV production used levels of HIV p24 antigen to evaluate HIV activity, a somewhat suspect measure.18,19 Other studies based on p24 report no consistent change or reductions in viral activity in persons on GM-CSF.20,21,22The context in which GM-CSF is used, such as in patients on chemotherapy for secondary infections or cancer, is difficult to interpret. It should be remembered that chemotherapy suppresses lymphocyte proliferation and therefore

may temporarily reduce viral activity by reducing the pool of infectable cells. In one of the studies that reported increases in p24 antigen while on GM-CSF, viral activity fell when AZT was reinstituted.23 In vitro studies suggest that GM-CSF increases the activity of AZT (and d4T) in macrophages, possibly by increasing the necessary cellular processing of the drug.24 The same study reported reduced activity for ddI and ddC, it should be added. Two recent investigations employing the new PCR assay to monitor viral load in patients on stable AZT report that use of GM-CSF has no consistent effect on viral load.25,26

The lack of effect on overall viral load should not come as a surprise in light of the small proportion of the viral reservoir constituted by infected macrophages. According to Dr. Ho's estimates, infected monocytes and macrophages contribute less than 1% of the total viral load. But that small trickle is important to clear out. Whether GM-CSF is the best agent for effecting this clearance remains to be seen. Cell culture studies suggest that M-CSF may promote HIV replication more efficiently. Dr. Ho also is considering that compound, which is licensed to Chiron but not yet marketed in this country.


1 Rowe JM et al. Blood. July 15, 1995; 2(86):457-62.
2 Smith PD et al. Journal of Infectious Diseases. May, 1990; 161(5):999-1005.
3 Collins HL and Bancroft GI. European Journal of Immunology. June, 1992; 22(6):1447-54.
4 Robin G et al. Lymphokine Cytokine Research. August 1991; 10(4):257-63.
5 Flø RW et al. AIDS. June 1994; 8(6):771-7.
6 Manfredi R et al. Journal of Chemotherapy. August, 1996; 8(4):214-20.
7 Nassar F. Antimicrobial Agents and Chemotherapy. September, 1994; 38(9):2162-4.
8 Baldwin GC et al. Blood. October, 1989.74(5):1673-7.
9 Kemper C et al. The Thirty-Fifth Interscience Conference on Antimicrobial Agents and Chemotherapy. September 17-20, 1995; Abstract G109.
10 Gorbea MC et al. The Ninth International Conference of AIDS. June 6-11, 1993; Abstract PO-B07-1196.
11 Perno CF, et al. Blood. August 15, 1992; 80(4):995-1003.
12 Koyanagi Y et al. Science. September, 1988; 241:1673-5.
13 Kitano K et al Blood. April 15, 1991; 77(8):1699-1705.
14 Maerz A et al. Annual Conference of the Australia Society of HIV Medicine. November 3-6, 1994; 6:248.
15 Matsuda S et al. AIDS Research and Human Retroviruses. September, 1995; 11(9):1031-8.
16 Perno CF et al. Third Conference on Retroviruses and Opportunistic Infections. January 28-February 1, 1996; abstract 139.
17 Walsh DG et al. American Journal of Pathology. October 1991; 139(4):877-87. 18 Pluda JM et al. Blood. August 1, 1990; 76(3):463-72.
19 Kaplan LD et al. Journal of Clinical Oncology. June, 1991; 9(6):929-40.
20 Levine JD et al. Blood. December 15, 1991, 78(12):3148-54.
21 Scadden DT et al. Am J Clin Oncol. 1991; 14(suppl 1): S40-4.
22 Davey RT Jr et al: Journal of Infectious Diseases. July, 1991;164(1):43-52.
23 Pluda JM et al. Blood. August 1, 1990; 76(3):463-72.
24 Perno CF et al. AIDS Research and Human Retroviruses. August, 1990; 6(8):1051-5.
25 Scadden DT, et al. AIDS Research and Human Retroviruses. August 10, 1996; 12(12):1151-1159.
26 Davison FD et al. Journal of Clinical Pathology. September 1994, 47(9):855-7.

HIV/AIDS-Related Fatigue

by Lillian Delmonte, D.Sc.
"Fatigue impacts significantly on the psychological well-being and quality of life of patients with HIV/AIDS," says William Breitbart, M.D., attending psychiatrist and director of Clinical Care Training for the Psychiatry Service Memorial Sloan-Kettering Cancer Center in New York City. Fatigue, generally measured by patients' responses to questions concerning reductions in daily activity, is one of the most prevalent and most undertreated problems experienced by patients with HIV infection and AIDS. Yet many physicians tend to ignore HIV/AIDS-related fatigue. They do not consider it to be a complaint requiring clinical intervention even though fatigue can be as disabling as pain.

In a recently concluded study (presented at the Forty-third Annual Meeting of the Academy of Psychosomatic Medicine last November), Dr. Breitbart evaluated interrelations between fatigue, psychological well-being and quality of life in 438 ambulatory adult AIDS-diagnosed volunteers recruited primarily from three private New York hospitals from the end of 1992 through early 1995. Volunteers had been HIV seropositive for an average of 4.4 years before participating in the study. The probable sources of their HIV infection were intravenous drug abuse (52.8 per cent), homosexual contact (28.3 per cent) or heterosexual contact (sixteen per cent), and blood transfusions (0.7 per cent). Patients were assessed both by self-reported description of their own condition and by clinician-rated evaluation of psychological distress, quality of life and functional ability. Seven different scales were applied to grade fatigue, physical distress, psychological distress and performance status. The presence of fatigue was determined using the PSC (AIDS-Specific Physical Symptom Checklist) item regarding persistent fatigue lasting for at least two weeks and the MSAS (Memorial Symptom Assessment Scale) item covering "lack of energy." To be classified as having fatigue, volunteers had to respond positively to the relevant questions in both scales.

Fifty-four per cent of the study participants were considered to have fatigue on the basis of this evaluation. An analysis of multiple variables found that fatigue was more prevalent in women and in respondents with a greater number of AIDS-related symptoms, particularly persistent or frequent pain. As expected, there was a strong correlation between high scores for depression in the survey and fatigue, but 58% of the 231 individuals with fatigue did not have elevated scores for depression. In those who did have depression associated with a high level of fatigue, not all the fatigue symptoms could be accounted for by psychological factors.

The investigators concluded that more structured interviews were needed to better differentiate fatigue that is a reflection of depression from fatigue arising as a direct symptom of HIV disease. It also needs to be better determined and how frequently depression results from fatigue rather than the other way around. Notably, the incidence of fatigue did not vary according to CD4 cell count, indicating that more elaborate techniques, such as viral load assays, are required to document the relationship between HIV and fatigue.

The current Sloan-Kettering data confirms a survey of HIV-related fatigue conducted by Stanford University researchers and reported at the Ninth International Conference on AIDS in 1993 (abstract no. PO-B01-0885). That study compared fatigue in 394 HIV-negative persons and 917 people with varying stages of HIV infection. In this study, 55.8%, 28.4%, 17.3%, and 5.1% of the AIDS, ARC, asymptomatic and HIV-negative participants, respectively, reported experiencing fatigue in the seven days prior to the interview. Fatigue appeared to be a treatment-related side effect in 5.4% of respondents with AIDS, 6.2% of those with ARC and 4.3% of those with asymptomatic HIV.

Causes of Fatigue

HIV/AIDS-related fatigue is multidimensional. It has no single cause, but rather a constellation of interactive causes, of which psychological distress and depression are just one aspect. Physical symptoms, drug side effects, sleep disturbances, malnutrition and wasting, AIDS dementia, hormonal insufficiency (due to low adrenal gland output, for example) and muscular weakness (HIV- or AZT-related myopathy) all contribute to fatigue. In particular, the Memorial Sloan-Kettering study noted a small, but statistically significant association between fatigue and anemia (low oxygen transport by the blood).

Anemia may affect two-thirds of people with AIDS . Its causes include:

  • Disease-related anemia: Long-term disease of any type can lead to "anemia of chronic disease." Iron is poorly incorporated into red blood cell (RBC) precursors, and hemoglobin levels are low. In addition HIV or opportunistic infections (such as Mycobacterium avium and B19 parovirus) can infect and destroy the bone marrow cells that develop into RBCs. The inflammatory agent tumor necrosis factor, whose production is stimulated by HIV, MAC and other pathogens, also decreases RBC production.

  • Drug-induced depression of stem cells in the bone marrow, leading to inadequate red blood cell (RBC) production. AZT, most notoriously, has this effect in persons with AIDS. (15 to 20% of PWAs treated with AZT experience anemia -- but AZT causes anemia in only a few per cent of patients with less advanced HIV infection.)

  • Disease- or treatment-related kidney damage. Cells in the kidney regulate the bone marrow's production of red blood cells by secreting erythropoietin (EPO) in response to lowered oxygen levels in the blood. Damage to these cells leads to a blunting of the EPO response and anemia.

  • Nutritional deficits: Cryptosporidiosis and other gastrointestinal infections can lead to malabsorption and deficiencies in folic acid and, especially, vitamin B12, which are necessary for RBC formation. Iron deficiency can arise from chronic blood loss, for example as a result of intestinal Kaposi's sarcoma or cytomegalovirus infection.

Pneumocystis carinii and cytomegalovirus and other active opportunistic conditions directly trigger fatigue. Part of this fatigue results from the increase in inflammatory cytokines as the immune system responds to the OIs, but PCP lowers oxygen levels in the blood by interfering with lung absorption, and CMV can reduce hormonal levels by infecting the adrenal glands.

Treating the Medical Causes

Dr. Breitbart stressed the importance of appropriate medical treatment and psychosocial support for reducing a patient's fatigue symptoms and improving quality of life. Addressing the medical causes of fatigue, which may be as simple as switching patients to alternative drugs, can have a rapid, positive effect A better diet and special nutritional supplementation to combat malabsorption or the special needs of those with chronic infection can be useful in combating fatigue.

Raising hemoglobin levels by administering blood transfusions, recombinant erythropoietin (brand names: Epogen or Procrit) or nutritional therapy also gives people improved energy levels and greater functional capacity. Red blood cell transfusions are a "quick fix" for treatment-associated anemia, but they do carry a small risk of immune reactions and transmission of blood-borne infections. Recombinant erythropoietin has the advantage of being free of significant side effects. It requires three weeks to elevate hemoglobin levels, though, and occasional blood transfusions may still be required.

The team at Sloan-Kettering at present is conducting a placebo-controlled, six-week study on the ability of two commercially available psychostimulants (Ritalin and Cylert) to reduce fatigue in ambulatory HIV-positive individuals. This trial is still enrolling (call Monique Kaim, Ph.D., at 212/583-3002 for more information). According to Dr. Breitbart, "Stimulants are an umbrella kind of therapy that help you deal with fatigue of any cause." They also can interfere with sleep and cause loss of appetite, though, both of which might further fatigue in the long run.

Dr. Breitbart suggested that patients should be encouraged to develop their own methods of coping with fatigue. The first steps patients can take are pacing their daily lives, altering activity/rest patterns, including frequent rest breaks, and delegating activities to others. Working out at the gym or jogging is a form of natural psychological and physical stimulation, but more moderate exercise, like walking, can be helpful, too. People have also found such meditative exercise forms as yoga, tai chi or chi gong to be very restorative even when their physical capacity is limited by disease. Massage, therapeutic touch, acupuncture and other alternative therapies that claim to restore the body's "energy balance" also may have a role to play, if only for their meditative aspects, which relieves mental tension and depression.

Psychosocial counseling and support groups can be important for helping the patient to cope with emotional stress or anxiety. Occupational therapy can be a valuable strategy for distracting patients from focusing on their disease, symptoms and emotions. Finally, antidepressant medication can be used in cases of recalcitrant depression.

Commentary: Reinventing Expanded Access

by Theo Smart
Less than ten months after the approval of indinavir and ritonavir, a sizable number of people have already failed on these purported wonder drugs. Oh, there may be one or two antiretrovirals left that they have not taken, but given HIV's facility for mutating to achieve drug resistance, single and most dual agent therapy offers little more than a temporary delay in disease progression. It is easy to evolve resistance to one or two drugs. Only a few mutations are required, and one and usually two drugs do not suppress HIV replication well enough to prevent them from emerging.

New classes of antiretrovirals such as integrase, fusion blockers and zinc finger inhibitors all seem a long way off, and people's needs are more immediate. There are, of course, the new drugs now well along in the development pipeline, but getting the pharmaceutical companies to release those drugs to all who need them has not been an easy task. The last six expanded access programs in the U.S. have been extremely small, opening only a few months before FDA approval of the drugs in question. Participants were generally only offered access to the new treatment if they had very low CD4 cell counts and had failed or were intolerant to approved therapies. And the programs for the three protease inhibitors now on the market included lotteries for eligible applicants to further restrict the amount of drug distributed.

No one would contest that those most ill should receive experimental drugs first, but it has become clear that combination therapy can turn around the course of disease in people with AIDS. In contrast, the effect of monotherapy or virtual monotherapy (the addition of the expanded access agent to an ongoing, failing regimen) is transient at best and potentially harmful due to the drug-resistant HIV to which it gives rise. This is not what activists had in mind when they first fought for novel expanded access programs like parallel track.

There are those in the pharmaceutical industry who agree: "Someone needs to say, once and for all, that expanded access can't work the way it used to," said Maureen Myers, Ph.D., Director of Clinical Research--Virology for Boehringer-Ingelheim, the company that makes the now-approved nevirapine. "When designing these programs, you have to position the drug in such a way as to offer some benefit to the patients taking it."

Boehringer's nevirapine is a perfect example of a drug that offers little benefit when misused. Nevirapine needs to be taken in combination with several other agents that a person's has not yet had a chance to develop resistance to -- otherwise, resistance develops within weeks. Extending nevirapine access only to those who have failed all other medications would be an exercise in futility. That futility probably was the experience of most of the participants in the American expanded access program, who were treatment-experienced and had CD4 counts under 50. (We will never know for sure , though, what enrollees' experience was because of the minimal amount of data collected under the nevirapine program.)

Boehringer wants to take a different approach in Europe, according to Dr. M yers. Although not all the details have been worked out, the company is considering a larger program, open to those with higher CD4 cell counts. The protocol would expressly recommend nevirapine use in combination with other drugs that a person has never tried before. Companies have allowed participants in some previous expanded access schemes to use additional antiretrovirals -- this would be the first time that it is mandated.

Boehringer is taking a similar approach in the United States, where it has adjusted its pediatric expanded access program to grant nevirapine to children with less advanced disease and those who are treatment-naive, as well.

Agouron: Nelfinavir

Agouron has the largest expanded access program currently in place and is now enlarging it. The company is now offering its protease inhibitor nelfinavir (Viracept) to children between the ages of two and 13. According to Joy Schmitt, Agouron's Manager of Public Relations, the company should be able to extend access to children below two years of age within the next month or so, once Agouron considers that it has enough safety data. The importance of this initiative cannot be overemphasized, since pediatric formulations are not available for the three protease inhibitors now on the market.

The adult (age 13 and up) program also has changed substantially. It is now open to anyone unable to take indinavir and/or ritonavir due to intolerance and/or failure (as defined by a patient's personal physician) and who has had a CD4 count below 100 at some point over the history of one's illness. To enroll in either program call 800/621-7111 Monday through Friday, between 8 a.m. to 6 p.m., Eastern Standard Time.

The new adult protocol is considerably more generous than the old one. The former program had required applicants to fail or be intolerant to all approved protease inhibitors, (including saquinavir). Now people have the option of using nelfinavir in combination with saquinavir. According to Agouron, nelfinavir increases blood levels of saquinavir five-fold, not as potently as ritonavir (which effects a 20-fold increase), but significantly nonetheless.

The old protocol's worst feature was the requirement that an individual's CD4 cell count had to be below 50 at the time of entry into the expanded access program. If that count was still elevated from earlier, transiently effective antiretroviral therapy, the applicant would have to wait until it slipped below 50, regardless of whether he or she had high HIV levels or was afflicted with potentially fatal opportunistic infections.

Of course, program participants still must fail or be intolerant to ritonavir "and/or" indinavir. This poses a problem, because of broad spectrum cross-resistance among protease inhibitors. Nelfinavir may give rise to a unique HIV drug resistance mutation when it is the first protease inhibitor used, but scientists from Merck and Abbott believe that most who fail on their drugs will benefit little from switching to nelfinavir. Agouron's drug resistance specialist, Amy Patick, Ph.D. says that she has no data to confirm whether or not indinavir and ritonavir cause cross-resistance to nelfinavir but commented, "It wouldn't surprise me [if cross-resistance arises] because the 82 mutation is in such a critical site on the enzyme." (The mutation at amino acid 82 is the key HIV protease mutation conferring resistance to indinavir and ritonavir.)

The question must be answered as quickly as possible, because there is more at stake than simply not responding to nelfinavir. Many entering the expanded access program have failed most of the approved therapies. If they are pre-resistant to nelfinavir and start it in combination therapy with the one or two commercially available antiretrovirals that they still have not taken, they risk exhausting the benefit from those remaining background therapies too. The French government considers the danger so great that it is refusing to implement a proposed nelfinavir expanded access program because the company has shown its regulatory agencies no data to suggest that nelfinavir would be of any use in patients resistant to the other protease inhibitors.

But a broader program that covers people who have not failed ritonavir or indinavir would conflict with Agouron's clinical endpoint study. This trial compares disease progression in those taking nelfinavir plus nucleoside analogs to the experience of those taking ritonavir plus nucleoside analogs. It will enroll 1,300 participants with CD4 counts below 100. (The study is being conducted by the Community Programs for Clinical Research on AIDS (CPCRA); call 800/TRIALS-A for site information.)

In any case, the expanded access issue is rapidly becoming moot in the U.S.: Agouron has completed nelfinavir's New Drug Application with the FDA, and the protease inhibitor is expected to be on the market in a few months, only about a half year after Agouron commenced the expanded access program.

Glaxo: 141W94 and 1592U89

An activist consensus statement has called on Glaxo Wellcome to release 1592U89, the most potent anti-HIV nucleoside analog ever tested in people with HIV, immediately through a compassionate use/salvage program and subsequently through a more traditional expanded access program. The proposed salvage protocol would be open to: 1) anyone with a CD4 cell count below 50 or viral load over 40,000 copies per ml who has failed approved reverse transcriptase inhibitors and at least one protease inhibitor; or 2) anyone with a diagnosis of AIDS dementia complex unresponsive to existing therapies. The second phase of the proposed expanded access would be a little more open. It would cover those who cannot participate in ongoing 1592 trials, who have HIV viral loads over 20,000 despite ongoing therapy, or evidence (such as genetic analyses) that their HIV is resistant to current therapies.

Other activists, including this author, have recommended that the second phase of this program be a dual expanded access program for both 1592 and Glaxo Wellcome's protease inhibitor, 141W94, open to patients with less than 200 CD4 cells who have failed or are intolerant to indinavir and/or ritonavir. (This would not conflict with 141W94's clinical endpoint study which will be a head-to-head comparison of 141 and indinavir plus participant's choice of additional anti-HIV agents.)

A dual program is possible since both drugs are in similar stages of clinical development. The protocol would be in effect a large simple trial that would randomly assign participants to one of three arms, 1592, 141 or the combination. All patients would be free to use their choice of additional therapy in addition to the experimental treatment(s). This allowance could extend to concurrent use of other protease inhibitors as drug interaction data becomes available.

An expanded access protocol of this sort could yield valuable information about the use of 1592 and 141 together. Unfortunately, data collection requirements generally impair entry into expanded access programs since most public health clinics and doctor's offices lack the staff and resources to do the paperwork.

Novel mechanisms for improving data collection must be developed and implemented in order to gather more complete information on the safety and efficacy of drugs supplied through expanded access. One solution might be to collect data in a sizable subset of the total expanded access population. Another is to employ a contracted research organization that would collect pertinent data directly from the patients' files, provided permission to do so is included in the informed consent form that patients sign to accept the expanded access program's conditions.

Glaxo Wellcome is meeting with a number of activists this month, to discuss expanded access proposals, and its own revised proposal should be available shortly.

Gilead: Adefovir

There are other experimental antiretrovirals that should be candidates for expanded access programs, too. Some, such as Gilead's adefovir, have been in clinical trials as long as, if not longer than, the two from Glaxo.

Adefovir has fallen through the cracks, overlooked by activists and clinicians. The half-log (70%) reduction in viral load that this drug achieved in studies of antiretroviral therapy-experienced people last year was overshadowed by the potency of 1592, the protease inhibitors and non-nucleoside reverse transcriptase inhibitors (NNRTIs). But, aside from 1592 and perhaps 3TC, the drug is as potent as any of the nucleoside analogs in the treatment-experienced population -- and few would dispute that agents such as AZT, d4T and ddI still have a role in combination regimens. Furthermore, adefovir's effect as a monotherapy is much more durable than the NNRTIs. Gilead claims that the drug's antiviral activity is sustained out to at least nine months, the longest anyone has been followed so far.

Of great importance to people who have exhausted most of the other antiretroviral options, HIV isolates resistant to all the other nucleoside analogs and NNRTIs remain susceptible to adefovir, according to data presented in November at the Third International Conference on Drug Therapy in HIV Infection in Birmingham, England. Add to this the drug's activity against CMV and hepatitis B, its lack of affect on the blood and cellular levels of other antiretrovirals and its once-a-day dosing, and you have a compound that could play a valuable role in combination regimens.

Gilead says that it is considering an expanded access program, but is unsure how to design it. Part of the problem is that such a plan could interfere with recruitment for adefovir's pivotal clinical endpoint (disease progression) trial, which opened this month at CPCRA sites across the country.

The trial has a targeted enrollment of 2,000 people with CD4 counts under 100. Participants will receive adefovir or placebo in addition to their choice of background therapy. This trial design is essentially the same Abbott used in the study that won full (as opposed to accelerated) FDA approval for ritonavir in advanced HIV infection.

Anyone included in the pivotal adefovir trial could be randomized to placebo even if their HIV still is sensitive to only one or two other antiretrovirals. Such people would then receive what is certainly suboptimal therapy. The most ethical solution is to exclude individuals with a long history of treatment failure and offer them adefovir through expanded access.

  • Email Email
  • Printable Single-Page Print-Friendly
  • Glossary Glossary

This article was provided by Gay Men's Health Crisis. It is a part of the publication GMHC Treatment Issues. Visit GMHC's website to find out more about their activities, publications and services.