Despite the enthusiasm at the Eleventh International Conference on AIDS in Vancouver over recent advances in anti-HIV therapy, it will be a long time until there are manageable combinations of easy-to-tolerate drugs that work for everyone. New agents are needed particularly for the many individuals who already have failed on protease inhibitors. Researchers at Vancouver reported upon an array of new agents that have not seen substantial clinical testing. Most of these compounds seem promising in the test tube but have yet to demonstrate anti-HIV activity in people.
Given the activity of established drugs in the same class, some of the new reverse transcriptase and protease inhibitors seem destined to proceed smoothly through the clinical trials process. It is unclear to what extent such agents truly will increase people's therapeutic options because of problems such as cross-resistance and intolerance. The most significant advances may arise from the work of researchers who are looking for other aspects of HIV's lifecycle that could be as successfully exploited as reverse transcriptase and protease have been.
Eric de Clerq, M.D., of the Rega Institute for Medical Research in Belgium has long been working with compounds that block the way HIV binds to cells. Among these is dextran sulfate, one of the first anti-HIV drugs studied in people. De Clerq's team now has demonstrated that the virus can become resistant to the compound in cell culture studies by changing certain amino acids in the V3 loop of HIV's gp120 envelope protein (conference abstract Th.A.152). Although clinical study of dextran sulfate was discontinued because of its toxicity, the finding is significant because it helps explain where the virus is susceptible to attack by compounds such as dextran sulfate. Less toxic compounds might be developed to exploit this weakness in the virus envelope. One such drug, PRO 2000, was reported to be safe when administered as a ten-minute infusion to HIV-negative volunteers (abstract We.B.3129).
"T-20 is the most potent agent that we've ever looked at in [a mouse model for HIV infection]," according to Paul Black, M.D., Ph.D., of the Food and Drug Administration. In his study, the drug completely blocked HIV infection in a mouse model. Since the drug is a protein derived from part of HIV, there is a fear that its administration could provoke an antibody response that may inactivate it. Another potential problem is that the compound cannot be taken orally since it is broken down by acids and enzymes in the digestive tract. In its current form, the drug would have to be given as an injection two or three times a day. To cut down the number of injections, Trimeris is looking at second-generation drug candidates that persist longer in the blood. They also are considering administering the drug via "mini-osmotic pumps," small implanted devices that secrete a constant level of drug over time. These pumps could be implanted in the arm, supplying drug for weeks or perhaps longer. First, though, Trimeris Pharmaceuticals intends to determine whether the current form of T-20 has activity in people. Dose-ranging studies are planned to begin by the end of the year.
The surprising activity of Glaxo-Welcome's new nucleoside analog 1592U89 (see August's Treatment Issues) has shown that it may be possible to optimize the potency of this class of drugs. Dr. Jan Balzarini, also of the Rega Institute for Medical Research in Belgium, presented data on a modified form of d4T that is 50- to 100-fold more potent in cell culture studies, because it enters cells more readily (abstract Mo.A.1012). Meanwhile, a group at Emory University working with a lipid-associated form of ddI in mice, found that the new agent more effectively reached and persisted in lymphoid tissue than standard ddI (abstract Mo.A.1079).
The National Cancer Institute has moved a new nucleoside analog, F-DDA, into phase I/II dose-ranging studies (abstract Mo.A.1076). The drug's active form is similar to ddI, though the ddI-resistant virus remains susceptible to F-DDA. Like ddI, F-DDA's activity in cell culture is increased by the addition of hydroxyurea and ribavirin. Unlike ddI, the new drug does not need to be formulated with a buffer, and thus should not cause the gastrointestinal side effects that make ddI so hard to tolerate. If the drug is found to be active in the dose-ranging study, it will be developed by U.S. Bioscience, makers of the anti-PCP drug trimetrexate.
Two posters presented the antiviral activity, resistance and safety profile of Hoechst-Bayer's NNRTI (non-nucleoside reverse transcriptase inhibitor) HBY-097 (abstracts Mo.A.1102, Mo.B.1326). In four studies with doses ranging from 375 to 3000 mg per day, the drug reduced viral load by an average of 1.38 to 1.63 logs (95.8 to 97.7%). Four out of ten patients treated with higher doses of HBY-097 developed a rash similar to that seen with nevirapine and delavirdine. The drug is currently in dose-ranging and drug interaction studies (including a study with indinavir, and another with AZT). Pivotal clinical studies are slated to begin next year. For more information call 800/TRIALS-A (800/874-2572).
Preclinical data were reviewed for DMP-266, an NNRTI discovered by Merck and now licensed to DuPont-Merck (abstract Mo.A.1077). This drug has established antiviral activity and is currently in phase II clinical trials (see May's Treatment Issues, page 8). While most NNRTIs are defeated by the evolution of just one resistance-conferring mutation in the HIV reverse transcriptase, it takes two to protect HIV against this drug. Nevertheless, two mutations in the enzyme have been shown to readily occur when NNRTIs are used with AZT alone or in other suboptimal combinations. It remains to be seen how quickly resistance to DMP-266 will occur in the current studies that combine the drug with indinavir or AZT/3TC.
Using NNRTIs in combination with more potent regimens may be the best hope for defeating NNRTI-resistant virus. A second option would be the development of compounds that can attack HIV resistant to nevirapine or delavirdine. Preclinical work has suggested that the virus would need to develop a unique mutation to evade the effects of HBY-097. In human studies, this novel mutation has yet to be seen. A mutation that causes resistance to delavirdine, though, was seen in two out of ten patients treated with HBY-097 at study end (fourteen days), but it is not yet clear how much or whether these mutations will allow the virus to resist the effects of this new drug.
Another promising group of agents has yet to enter clinical trials: Compounds owned by Uniroyal, related to UC10 (see Treatment Issues article, Overcoming Resistance to Reserve Transriptase Inhibitors; September 1995, page 9) have been shown to be active against virus containing all five of the major mutations that confer resistance to the other NNRTIs (abstract Mo.A.1051). Furthermore, the addition of the Uniroyal compounds to 3TC produces, in Dr. de Clerq's words, "a knock-out combination." In his laboratory experiments the virus could not become resistant to both drugs at the same time and still retain the ability to replicate. Though the Uniroyal compounds are not yet in the hands of an entity that develops drugs, Dr. de Clerq claims there is a good chance that they may be licensed to a pharmaceutical company in the near future and moved into clinical testing. He notes that these compounds are stable in human blood, and some have been shown safe and bioavailable in animals.
A number of researchers spoke of the HIV integrase enzyme as the next important target of antiretroviral therapy. Integrase is an enzyme that integrates HIV genetic material into the host cell's DNA.
In one presentation, Edward Robinson, Ph.D., of University of California Irvine reported preclinical data on a class of compounds derived from Bolivian plants that inhibit integrase activity. Merck Pharmaceutical's Daria Hazuda, M.D., Ph.D., presented a poster on integrase inhibitors found by screening other natural products (abstract Mo.A.1020). Though the fungus-derived compounds she reported upon are very potent blockers of integrase catalytic activity in drug screening tests, they do not show activity against HIV when cultured with cells. Lack of activity in cell culture is a common problem with many of the integrase inhibitors that have been identified. Either the compounds do not effectively get into the cell, or doses that inhibit HIV are also lethal to the cells the drug is meant to protect.
Dr. Hazuda believes that further structural studies using these compounds, and others Merck has yet to announce, may show them how to modify the structures to create a potential drug candidate. This will take time: Dr. Hazuda commented, "We are several years away from developing clinically useful integrase inhibitors."
The one integrase inhibitor to reach human studies is Aronex's ARR-177, which now sports the name zintevir (abstract Th.B.943). This compound does not inhibit the "active" part of the integrase enzyme. Instead, it keeps integrase from binding to the HIV genetic material in the first place. Dose-ranging studies have recently begun in San Francisco.
HIV's zinc fingers are amino acid structures on the surface of one of HIV's core proteins. The zinc fingers capture and help package HIV genetic material into newly budding virions, and also appear to play a role during the earlier stages of cell infection. Two zinc finger inhibitors are currently in clinical studies. One owned by Parke-Davis has completed initial pharmacokinetic studies and will enter phase I/II dose-ranging studies within a month (see Treatment Issues article, Zinc Fingers-Next antiviral Target; October, 1995, pages 7-8). ADA, the other compound, is owned by the Dutch company Van de Velde and is now the subject of a European phase II study. According to William Rice, Ph.D., of the National Cancer Institute, the zinc finger mechanism of action for this drug was identified only recently. Both of these compounds, though, may become inactivated by common substances in the body. For example, Parke-Davis's zinc-finger inhibitor breaks down when exposed to glutathione, an antioxidant present in most cells.
Dr. Rice presented data on a new class of zinc finger inhibitors that are as active as the earlier compounds but not sensitive to glutathione or other antioxidants (abstract Th.A.150). Members of this class of chemicals are currently being studied for safety and activity in animal models.
Dr. Masatoshi Taneka of the National Cancer Institute reported that his agency, in collaboration with the Kyoto Pharmaceutical University and Japan Energy Co., has discovered a protease inhibitor, KNI-241, that is not stopped by HIV that has developed resistance to other available protease inhibitors (abstract Tu.A.260). The same group had taken one protease inhibitor, KNI-272, into clinical trials, but this drug is defeated by the same resistance-conferring mutations that inactivate ritonavir and indinavir, and did not seem to be as potent as the approved drugs. Development of KNI-272 is being discontinued in favor of KNI-241.
Finally, Suvit Thaisrivongs, Ph.D., of Pharmacia & Upjohn presented preclinical data on its third and most powerful protease inhibitor (abstracts Tu.A.261 and Mo.A.1084). Because of its potency, the company believes it will be active despite its tendency to become bound to albumin in the blood, a problem that led to the discontinuation of Pharmacia & Upjohn's two previous lead compounds. The drug is relatively easy to make and is active against HIV resistant to any of the approved protease inhibitors. Phase I dose-escalation studies will begin in Michigan before the end of the year.
While many left the XI International Conference on AIDS in Vancouver pondering the prospects of HIV "eradication" and marveling over the suggestive combination therapy data, others left impressed by how the diagnosis, treatment and prophylaxis of opportunistic infections are improving. Viral infections (cytomegalovirus -- CMV), bacterial infections (Mycobacterium avium complex -- MAC), malignancies (Kaposi's sarcoma -- KS) and fungal infections (oral and esophageal candidiasis) were all featured in separate symposiums. Most of the treatment and prophylaxis trial data concerned FDA-approved drugs like ganciclovir and clarithromycin, which already have widespread use in the U.S. As it turns out, the data generated from these studies will make a major contribution to managing patients with AIDS.
The CMV symposium was mostly comprised of clinical trial data (follow-up and final analyses) which had been presented at previous AIDS meetings. The two most interesting (and competing) studies presented were Syntex/Roche Study 1654 by Steve Spector, M.D., (abstract Th.B.302) and CPCRA 023 by Carol Brosgart, M.D. (abstract Th.B.301). In a nut shell, Syntex/ Roche 1654 demonstrated that oral ganciclovir was approximately 50% better than placebo in preventing CMV retinitis, and CPCRA 023 found that there was no statistically significant benefit of oral ganciclovir over placebo.
Dr. Brosgart's presentation was much anticipated because she was going to reveal the "reason" why oral ganciclovir "didn't work." After a run-through of the basic study data, Dr. Brosgart revealed the reason: ddI. Patients on the placebo arm who were taking ddI (alone or in combination with other antivirals) had a very low rate of developing CMV: 64 of the patients in the placebo arm were taking ddI and only one developed CMV disease. Conversely, patients in the oral ganciclovir arm who were on ddI had a high probability of developing CMV disease.
Of note, Dr. Spector reported that the Syntex/Roche 1654 team did a post hoc analysis of all patients taking ddI but did not find the same results seen in CPCRA 027. Dr. Brosgart stood by her team's analysis but cautioned that the results should be interpreted carefully and that the supposed negative ddI and oral ganciclovir interaction should be investigated further. Other groups have previously noted that :
Dr. Spector's presentation focused on the fascinating (yet depressing) virology data of 619 of the 725 patients from Roche/Syntex 1654. This was one of the first CMV prophylaxis trials with a CMV DNA PCR (viral load) sub-study. (Note: This is cytomegalovirus viral load not HIV viral load.) Out of all patients who were CMV PCR positive at study entry (approximate 45% in both arms), CMV disease developed in 43% of the placebo patients and 26% of the oral ganciclovir patients. Among the CMV PCR-negative patients at study entry, CMV disease developed in 1% of the oral ganciclovir patients and 14% of the placebo recipients, a highly significant difference.
Quantitatively, CMV PCR-positive patients were stratified by their CMV viral load count into three groups, those with: less than 2,500 copies; 2,500-50,000 copies; and 50,000 to 150,000 copies. The patients with CMV counts below 50,000 benefited from oral ganciclovir: 20% of the oral ganciclovir patients developed CMV disease versus 40% of those on placebo. For those with CMV loads more than 50,000, oral ganciclovir's protective effect was no better than placebo. Of note, there were ten patients with over 150,000 copies (all randomized to the oral ganciclovir arm), all of whom went on to develop CMV disease.
Jay Lalezari, M.D., of San Francisco (filling in for Dorothy Friedberg, M.D.) presented data from Syntex/Roche study 2226, a trial comparing three doses of oral ganciclovir (3, 4.5, or 6 grams) with IV ganciclovir (5 mg/kg daily) for CMV maintenance therapy (abstract Th.B.305). The median time to first progression was 41 days for 3 grams; 50 days for 4.5 grams; 57 days for 6 grams; and 70 days for IV treatment. The difference in time to progression between the low dose arm versus the IV arm was statistically significant.
These are important -- yet disquieting -- results. The 3 grams dose is what is currently recommended to thousands of patients, but it is not as effective as higher oral doses or using IV ganciclovir. Moreover, 3 grams is twelve pills a day and costs at least $15,000 a year. If one is to start taking 6 grams a day, that would be an impractical 24 pills a day costing over $30,000 a year.Hoffmann-La Roche at present is developing a 500 mg capsule to reduce the number of required pills. The price needs reducing, too.
The MAC symposium also was comprised of clinical trial data which had mostly been presented at previous AIDS meetings. The two most important studies presented, ACTG 196 (clarithromycin versus rifabutin versus a combination of the two for MAC prophylaxis) and the Canadian MAC treatment study CTN 010 (a combination of three drugs versus four) have been detailed in Treatment Issues articles (Summary of Data from ACTG 175 and Delta Trials, and ACTG 175 and Delta; October 1995, pages 2-3).
Of interest, and eliciting considerable protest by ACT UP, were the results (and their subsequent mishandling) of CPCRA 027 (abstract: LB.B.6025). CPCRA 027 was a MAC treatment study comparing two doses of clarithromycin (500 mg twice a day versus 1000 mg twice a day) in combination with other agents. This study was stopped early because a high mortality rate was observed in the group of patients receiving 1000 mg twice a day. As of February 6, ten of 45 (22%) patients in the 500 mg arm had died versus 17 out of 40 (43%) patients in the 1000 mg arm, a significant difference. No plausible explanation was given for this increase in mortality; to this date these findings still puzzle the principal investigators and statisticians. These findings of increased mortality with high-dose clarithromycin are similar to those observed in two previous MAC treatment trials.3 CPCRA 027 -- a study some believe to have been unethical from its inception -- was, in part, conducted in order to confirm or disprove those earlier findings of higher mortality.
The results of a trial of Doxil versus the combination of bleomycin plus vincristine (BV) for the treatment of KS were presented in a late-breaking poster (abstract LB.B.6026). Doxil -- a liposomal encapsulated form of adriamycin, a common, yet powerful, cytotoxic chemotherapeutic agent -- was recently granted accelerated approval by the FDA for the treatment of relapse or refractory KS.
Both regimens were administered every three weeks by intravenous infusion, consisting of 15 mg/m2 of bleomycin and 2 mg of vincristine versus 20 mg/m2 of Doxil. Mostly conducted in Europe, this study enrolled 241 patients with 102 evaluable patients in the BV arm and 116 evaluable in the Doxil arm. Seventy percent of the patients had under 50 CD4 cells and no G-CSF was allowed. Of the Doxil patients, seven (5.8%) achieved a complete response and 64 (52.9%) achieved a partial response compared to one (0.8%) complete response and 27 (22.5%) partial responses achieved by the BV patients. The difference was statistically significant. At the end of treatment (when patients stopped drug) the response rates were lower in both arms: the rate of complete response was 3.3% and 35.5% for partial responses on Doxil versus no complete responses and 14.2% partial responses on BV.
While Doxil was found to be efficacious in treating patients' lesions, it was more bone marrow-suppressive with 76.1% of the Doxil patients registering an absolute neutrophil count (ANC) below 1500 compared to 52.5% of the BV patients. Paresthesia (numbness and tingling in the mouth) was more common for the BV patients (14.2% versus 3.3%) and oral candidiasis was found to be more common in the Doxil patients (28.9% versus 17.5%). No survival data were given.
These results are similar to those seen in a study comparing Doxil against ABV.4 Since both studies were conducted in chemotherapy-naive subjects and demonstrated Doxil's superiority, the FDA is likely to grant, Sequus, the makers of Doxil, full approval to market Doxil as first-line therapy.
New data on itraconazole oral solution (cyclodextrin) for the treatment of oral and esophageal candidiasis were presented in a number of sessions. Cyclodextrin is 30% to 60% more bioavailable than the standard itraconazole capsule when it is taken on an empty stomach. It is not available in the United States at present.
A study presented by Rabih O. Darouiche, M.D., compared the safety and efficacy of two doses of cyclodextrin with fluconazole (200 mg per day) for the treatment of thrush or oral candidiasis (abstract: Mo.B.117). A total of 190 patients (179 evaluable with a median CD4 cell count of 94) were enrolled to receive either itraconazole solution 200 mg daily for seven days or 14 days, or fluconazole 200 mg on day one and then 100 mg daily for 14 days.
There was no statistically significant difference between the three patient arms in terms of clinical response (freedom from lesions), fungal eradication (negative Candida culture) or relapse rates after four weeks' follow-up. Even so, the investigators concluded that itraconazole solution for 14 days was at least as effective (if not a bit better) as fluconazole in suppressing symptoms of oral candidiasis and in achieving a clinical and anti-fungal response.
Dr. Darouiche conducted an identically-designed study comparing the itraconazole solution to clotrimazole troches for the treatment of oral candidiasis. 149 patients (83% HIV-positive) were randomized to receive either the fourteen day dose of itraconazole or clotrimazole troches at 10 mg five times daily for 14 days.
There was no statistically significant difference between the treatment arms in their ability to achieve a clinical response: 77% for itraconazole solution versus 66% for clotrimazole. Itraconazole solution, however, was determined to be more effective in decreasing fungal colony counts and culture positivity: the itraconazole solution had a 66% response rate compared to 47% for clotrimazole.
Bruce Moskovitz, M.D., presented data on 110 patients enrolled in a multicenter, double-blinded, placebo-controlled study comparing itraconazole solution to fluconazole for the treatment of esophageal candidiasis (abstract: Mo.B.116). Patients who had both endoscopy-proven and culture-positive esophageal candidiasis received either itraconazole solution at 100-200 mg daily or fluconazole at 100-200 mg daily for two weeks beyond symptom resolution (minimum of three and maximum of eight weeks of therapy). Patients were assessed clinically every week and received an endoscopy at the end of treatment.
In the itraconazole arm, 48 of the 53 (91%) patients were symptom-free at the end of treatment as compared to 49 of the 57 (86%) patients in the fluconazole arm. The median time to clinical response was 27 days for the itraconazole arm and 28 days for the fluconazole arm. At the end of treatment, an endoscopy determined cure occurred more often in the itraconazole patients (90%) than in the fluconazole patients (80%). Likewise, 92% of the itraconazole patients as compared to 78% of the fluconazole patients achieved fungal eradication. Relapse -- during the four-week follow-up period -- occurred in 18% of the itraconazole patients and 27% of the fluconazole patients.
All three studies support the efficacy of cyclodextrin/itra
conazole oral solution and will probably be used in Jannsen's application to the FDA. If it is priced fairly, cyclodextrin might be a more suitable option over fluconazole given the latter drug's high price, drug interactions and potential to cause resistance.
- Medina DJ et al. Antimicrobial Agents and Chemotherapy. May 1992; 36(5):1127-30.
- Gaines K et al. Tenth International Conference on AIDS. August 7-12, 1994; 10(1):7 (abstract 0048).
- Chaisson RE et al. Annals Internal Medicine. December 15, 1994; 121:905- 911.
- Northfelt DW et al. Third Conference on Retroviruses and Opportunistic Infections. Washington DC, January 28-February 1, 1996; page 123 (abstract 379).
Outdated yet Out-of-Reach: AIDS Society Treatment Guidelines at the International Conference
by Gabriel Torres, M.D.
A thirteen-member international panel of HIV experts convened by the International AIDS Society-USA (IAS) released new antiretroviral treatment guidelines amid a turbulent crowd of demonstrators, researchers, HIV clinicians, people with HIV and AIDS and other participants of the XI International AIDS Conference in Vancouver, Canada, on July 6, 1996. These guidelines are essentially an updated version of those drafted at a meeting in January and published in the July 10 issue of the Journal of the American Medical Association. The panel endeavored to address the most common HIV and AIDS treatment scenarios in clinical practice, using the latest information available at the time the guidelines were drafted. But discussions at the Conference indicated that the guidelines have significant shortcomings because they do not take into account either the rapid evolution in anti-HIV therapy or the real-world limitations and complications practitioners will encounter in following even these now outdated recommendations.
When to Initiate Therapy
The panel agreed that treatment needs to be initiated before irreversible immunologic damage occurs based on an assessment of disease progression. For asymptomatic patients, treatment was recommended for those with any of the following conditions: CD4 count less than 500, a CD4 cell percentage below 25, CD4 count greater than 500 if viral loads are higher than 30 to 50 thousand or those with a rapidly declining CD4 count (loss of greater than 300 cells/mm3 over twelve to eighteen months). Some of the panel's experts would defer therapy in patients with stable CD4 counts between 350 to 500 and viral loads less than 5,000 or 10,000.
Those with CD4 counts greater than 500 who are considering initiating therapy should be aware that there are no available data to support treatment at this early stage of HIV disease. Difficulties concerning a patient's tolerance to and acceptance of a given treatment, the expense of the treatment, the potential for development of viral resistance to the treatment and other problems associated with long-term toxicity are all major considerations in starting therapy at this point.
The panel also felt that therapy for those with HIV levels greater than five to ten thousand should be considered, but some members suggested that all patients with viral loads greater than 5,000 definitely should be treated. In addition to these guidelines, it was strongly recommended that all symptomatic patients (including those with oral candidiasis, hairy leukoplakia and other symptoms of mild immunodeficiency) should be treated with antiretroviral drugs, regardless of CD4 count or HIV viral load.
The panel did not address the critical issue that many clinicians and patients still do not have access to HIV viral load testing, and thus the recommendations based on viral load are not applicable for a substantial number of people.
What Types of Drugs to Use for Initial Therapy
The panel felt that the strongest data supported the use of AZT plus ddC, AZT plus ddI or AZT plus 3TC for initial therapy. ddI monotherapy was less favored. For patients with a high risk of progression (rapidly dropping CD4 count, presence of symptoms, or rising viral loads), the use of protease inhibitors in an initial treatment regimen may be warranted. If a protease inhibitor is to be added to any of these regimens, the panel felt that the choice should be based primarily on antiretroviral potency and then on safety, tolerability, durability of antiviral effects, drug resistance patterns, the potential for limiting future treatment options and cost. The panel came to this conclusion after discussing whether the most potent regimen, which is usually a triple regimen that includes a protease inhibitor, is appropriate during the initial stages of treatment or whether such therapy should be reserved for patients at higher risk of progression.
The panel omitted the role of nevirapine in initial or follow-up regimens or the use of two protease inhibitors together, data concerning which were released in Vancouver. The recommendations also ignored d4T monotherapy altogether even though it is one of the stronger monotherapies available and widely used.
None of the recommended regimens included more than three drugs, despite the fact that many in the community are starting to use four or five antiretroviral drugs together, often in conjunction with supportive agents such as interferon alpha, interleukin-2, hydroxyurea and many other alternative and complementary therapies. Furthermore, the panel did not rank its recommendations based on preference. Various options appeared to be given equal weight in the report though some were supported by large-scale trials measuring disease progression and others were not.
When to Change Therapy
The first reason for changing therapy discussed by the panel was treatment failure -- a rebound to 0.3 to 0.5 log (one-half to two-thirds) of pretreatment viral load, a CD4 count decline after an initial rise or a return to the pretreatment value or new opportunistic infections. Other reasons mentioned for switching included a patient's inability to comply with a prescribed regimen due to toxicity or lifestyle factors or use of a suboptimal regimen such as AZT or ddC monotherapy.
Just how a patient's therapy should change would, of course, depend on the reasons for altering it. For example, if a patient's intolerance to a given drug is the primary reason for change, then finding drugs that the patient can tolerate is paramount. If the problem is treatment failure, then drugs with greater potency and different mechanisms of action are recommended, particularly those who do not exhibit cross-resistance. A change to the most potent regimen available, based on virologic, immunologic and clinical characteristics of the individual patient is recommended.
Several common examples of clinical scenarios were presented for patients who have been on the following regimens and require a change in therapy:
AZT monotherapy: If the patient has stable disease, adding ddI, ddC or 3TC or switching to ddI monotherapy (less preferred) are all acceptable options. If the patient has advanced disease or extensive prior zidovudine history, adding 3TC or switching to another combination with or without a protease inhibitor were recommended.
- Combination therapy with two nucleosides or ddI monotherapy: Patients on these regimens should change to a two- or three-drug regimen containing at least two new drugs, one of which could be a protease inhibitor.
- Protease inhibitor regimens: For patients already using a protease inhibitor-containing regimen, the panel recommended a change to at least two new drugs, preferably including a protease inhibitor with a different pattern of resistance mutations in HIV.
This last switch alternatively could include addition of a non-nucleoside reverse transcriptase inhibitor (NNRTI) like nevirapine. Again, the panel did not consider NNRTIs when the guidelines were originally drafted, and the NNRTIs' precise role remained poorly defined in the eyes of many attending the IAS seminar.
When to Stop Antiretroviral Therapy
The panel also addressed the question of when to stop antiretroviral therapy in selected patients. It concluded that stopping is only appropriate for patients with advanced HIV disease who have significant toxicity and declining quality of life associated with therapy. Consequences of increased viral replication after withdrawal of therapy, including the potential neurologic effects, should be considered prior to discontinuation of an antiretroviral regimen as well.
Three Special Cases
- Acute or primary (new) infection
The panel recommended considering entering such patients in a clinical trial or treating with the most optimal combination of drugs (two or three) for at least six months. The panel did not address the treatment of persons with recent HIV infection (less than six months), even though there is evidence that such patients may receive similar benefit from therapy (see Treatment Issues article, Nipping HIV in the Bud; August 1996).
- Prophylaxis of high-risk exposures
For serious needle stick exposures, two nucleoside analogs (usually AZT plus 3TC) were recommended. A three-drug regimen (two nucleoside analogs and a protease inhibitor) was preferred if the exposure is likely to involve a drug-resistant HIV variant. No recommendation was issued for high-risk sexual exposures since no human studies have evaluated a "morning after" prophylaxis. Animal studies, however, do indicate that there may be a role in using antiretroviral agents to abort HIV infections.
- Vertical transmission
AZT monotherapy was recommended for HIV-positive pregnant women and for their newborns whether or not the mother receives treatment. The paucity of data on other regimens kept possible alternatives off the recommended list.
The absence of anything but AZT drew criticism from many obstetricians and gynecologists, who felt that such treatment is suboptimal. Limiting the recommendations to AZT failed to address those pregnant women with extensive prior experience with zidovudine or women who had been taking other antiretroviral drugs -- must they switch to AZT monotherapy two trimesters before giving birth? Some complained that no expert with experience treating HIV-positive pregnant women was included on the consensus panel.
Other important patient populations were not addressed by the panel. What treatment, if any, should be offered to long-term non-progressors or to patients with moderate viral loads of ten to twenty thousand? And finally, what should be done for those with markedly depleted immune systems (CD4 counts below 50), particularly those who have failed most if not all available treatments?
Modifying the Guidelines
The panel acknowledges that its recommendations leave out much new treatment information coming out of the very conference at which they were presented and anticipates revision in the next few months. As it is, the guidelines represent the perceived wisdom of first-rate HIV specialists earlier this year. But with the field in a state of rapid flux, guidelines can only be helpful if they advise on the state of the art on such topics as viral eradication and the optimal use of combination therapies as new agents become available.
Did They Consult Darwin?
One of the primary criticisms of the IAS guidelines was that they lacked an urgent call to hit HIV "early and hard" by using at least two nucleoside analogs and a protease inhibitor in persons with asymptomatic disease and relatively low viral loads. When changes in this initial regimen are necessitated by a rebounding viral load, the guidelines' critics argue that physicians should introduce three or more new drugs to which the patient's virus does not appear resistant. In a rare public display of outrage at one's scientific colleagues, Joep Lange, M.D., of the University of Amsterdam complained during an International Conference presentation on viral "eradication" (a commentary concluding session no. 436) that the IAS experts were giving "obsolete" advice. Dr. Lange commented that the IAS was not taking into account present knowledge concerning how readily drug-resistant HIV can evolve under conditions of partial viral suppression: "I am stupefied by the ignorance of biology exhibited by the IAS," he concluded.
But many physicians were not swayed by Dr. Lange's outburst. They felt that asymptomatic patients would not tolerate years of triple drug therapy and that three drugs may be unnecessary in any case. "I can bring the viral loads of many of my patients below undetectable levels with just two drugs, so why do they need three?" asked Steven Miles, M.D., of the University of Los Angeles after Dr. Lange's session was over. - DG
Delavirdine/Protease Inhibitor Interactions
by Dave Gilden
On September 27, the FDA's Antiviral Drugs Advisory Committee will meet to consider licensing delavirdine (brand name: Rescriptor). This Pharmacia & Upjohn product is the second nonnucleoside reverse transcriptase inhibitor (NNRTI) to come before the committee. (The first was Boehringer Ingelheim's nevirapine -- see the June/July Treatment Issues). Delavirdine is currently available to the public through an expanded access program -- call 800/779-0070 for more information.
A major issue that remains unresolved with both delavirdine and nevirapine is how these drugs affect the body's metabolism of protease inhibitors. Like the protease inhibitor ritonavir, both NNRTIs affect the CYP3A enzyme system in the liver that breaks down many drugs. Nevirapine seems to increase destruction of protease inhibitors whereas delavirdine (and ritonavir) inhibits this loss. The result of these hepatic changes on protease inhibitor blood levels or efficacy has been little explored up to now.
Drug combinations that include both an NNRTI and a protease inhibitor appeal to many people with HIV. First are those who want to suppress HIV as much as possible early in disease through a four drug combination that also includes two nucleoside analogs. Then there are persons with advanced infection who have had long histories with nucleoside analogs. Their HIV may be largely resistant to these older agents.
As part of the run-up to the FDA meeting and to partially satisfy intense public interest, Pharmacia & Upjohn has released the preliminary results of its tests on combining delavirdine with the three marketed protease inhibitors. These initial data all come from tests with 13 or 14 HIV-negative volunteers and are hardly enough to make final recommendations.
- For saquinavir (600 mg three times daily) plus delavirdine (400 mg three times daily), there was a slight reduction in delavirdine's mean trough plasma concentration (minimum level between doses) while the mean tough levels of saquinavir increased almost six-fold. Raising saquinavir levels to this extent will make it a more potent drug, but the effect of delavirdine varied greatly from person to person. Furthermore, 13% of the volunteers experienced a rise in blood levels of ALT, a liver enzyme whose blood levels go up when the liver is inflamed. Doctors should monitor such levels in patients taking delavirdine and saquinavir.
- Ritonavir was tested at 300 mg twice daily (half the standard dose) along with delavirdine at 400 mg three times a day. Ritonavir was unaffected, but a slight lowering of delavirdine levels was seen. Full-dose ritonavir has yet to be studied. In the meantime, Pharmacia & Upjohn is advising "caution" when combining these two liver enzyme inhibitors.
- Indinavir so far has been tested as a single 400 mg or 600 mg dose given to volunteers receiving 400 mg of delavirdine three times a day. Delavirdine appeared to double the effective dose of indinavir -- so that 400 mg of indinavir resulted in the plasma levels usually found after 800 mg is given. Based on this sketchy information, Pharmacia & Upjohn is suggesting reducing the standard indinavir regimen of 800 mg three times daily to 400 or 600 mg three times daily when delavirdine is coadministered. Beware that raising indinavir blood levels heightens the risk of kidney stones.
New York ADAP to Cover New AIDS Drugs plus Viral Load Testing
by Derek Link
After the longest and bitterest budget battle in state history, New York state legislators have passed a 1997 spending plan that for the first time commits state tax dollars to the local AIDS Drug Assistance Program (ADAP). Although most other states contribute their own resources to their ADAP programs, New York until this year gave almost nothing. The state will now contribute $8 million this year from the general fund and $12 million annually for the succeeding three years. These funds will come from a new health care program for uninsured New Yorkers.
The move will allow New York's beleaguered ADAP to stand on firmer financial footing. On January 1 of this year, citing growing demands and skyrocketing expenses, the state health department cut 129 drugs from the ADAP formulary and capped or eliminated primary care services for the more than 17,000 low-income, uninsured New Yorkers with AIDS who rely on the program.
A New York State health department advisory committee met on August 21 to formally decide how the new state money should be used. As of September 1, New York ADAP began covering HIV protease inhibitors, viral load evaluations and other crucial anti-HIV and opportunistic infection agents including nevirapine for HIV, cidofovir for CMV and DaunoXome for KS. Pursuant to the advisory committee's recommendations, New York ADAP also has restored some of the most important benefits and coverage of medications that were eliminated in January such as many antibiotic and psychotropic drugs. Two of the most expensive medications, Epogen and Neupogen, which are used to manage the bone marrow suppression caused by some of the covered therapeutic agents, are not expected to return to the ADAP list at this time, however.
New York's decision came after a massive grass-roots lobbying effort by AIDS organizations across the state. Thousands of New Yorkers registered their support with state officials, and editorial pages in central and western New York spoke in support of ADAP. Behind this campaign was a sense of indignation that as treatment breakthroughs were being announced in the news, New York was denying access to new therapies. The magnitude of New York's AIDS epidemic ensured a large public outcry.
No state has been hit harder by AIDS than New York. New York City alone has more AIDS cases than any state in the union, straining the heath care system to near breaking point. For the first time in 100 years, the life expectancy for males in New York dropped, due largely to the AIDS crisis, according to the City health department. Few New Yorkers remain untouched by this public health tragedy, and few political leaders can ignore this crisis.
Police Raid Buyers' Club
At 8 in the morning on August 4, California drug agents burst open the doors of the Cannabis Buyers' Club, which distributes marijuana for medicinal use to people in the San Francisco area with AIDS, cancer and other conditions. Besides making off with the club's supply of pure marijuana and drug-laced baked goods and $60,000 in cash, the agents seized confidential medical records for the organization's 12,000 members.
The buyers' club has been openly selling marijuana primarily to people with AIDS- or cancer-related weight loss since 1991. The organization's raison d'être enjoys popular support in the region, and elected city officials have protected the club from police harassment. State Attorney General Dan Lungren ordered the assault on the compassionate-use organization without consulting local officials.
The California Attorney General's office says it was gathering evidence that healthy individuals were obtaining marijuana for recreational use through the club. But many see the raid as an obvious attempt to derail Proposition 215. This ballot measure, which goes before voters in November, would grant people in California the right to grow or purchase marijuana for personal medicinal use. The Attorney General opposes Proposition 215. Many of the organizers of the campaign, who were involved in the buyers' club, now must make time to contend with criminal investigations rather than work on the ballot effort, which the raid helps to discredit. Anyone who wishes to help or make financial contributions to the organization should contact Californians for Compassionate Use at 415/621-3986.
CMV Retinitis Study Aborted
A study of the experimental anti-CMV drug MSL 109, made by Protein Design Labs, has been halted after an interim analysis found that the treatment offered no benefit over placebo. MSL 109, also called Protovir, is a monoclonal antibody targeted at CMV (see Treatment Issues article, The Changing Treatment Options for CMV Retinits; July/Aug. 1995, pages 4, 7-8). The failure of the compound comes as a surprise, since in an earlier uncontrolled study, it seemed to prolong time to progression of CMV retinitis when added to standard of care medications. That trial (reported in Antiviral Research, 1992; 16 (suppl 1):111), found progression was delayed by more than 200 days on average in seventeen volunteers, about twice the delay treatment with ganciclovir or foscarnet alone usually achieves.
On the basis of these early promising results, large randomized studies were initiated. The discontinued trial sponsored by the National Eye Institute's Studies of the Ocular Complications of AIDS (SOCA) and the AIDS Clinical Trial Group (ACTG), had enrolled over 200 of the planned 320 patients with newly-diagnosed or recurrent CMV retinitis. All patients were treated with currently approved anti-CMV therapy, but half were randomized to receive MSL 109 (60 mg every two weeks) in addition to the other medications.
The investigators chose to perform the interim analysis after a Data and Safety Monitoring Board noted that more deaths had occurred among the patients who were receiving MSL 109. The difference in mortality only occurred in those patients with recurrent retinitis. According to Dr. Curtis Meinhart of SOCA, this may have been a fluke, since the incidence of death was lower than usual in those randomized to receive placebo. Be that as it may, the interim analysis found that the addition of MSL 109 did not delay time to progression (65 days in the MSL 109 arm versus 66 days in the placebo).
Two other large MSL 109 studies are still ongoing. One is a CMV prophylaxis study in bone marrow transplant patients. The Data and Safety Monitoring Board recommended introducing no changes to this study, after finding no increase in mortality or side effects in participants taking MSL 109.
A second study conducted by the ACTG, study #266, has been temporarily placed on hold. A number of this study's investigators were involved in the earlier small study that reported positive results from MSL 109. Since the doses in ACTG 266 are the same or lower than the ineffective doses used in the SOCA trial, the ACTG study team has recommended enrolling new patients at higher doses of MSL 109. Those already enrolled on the study would have the option of quitting the study or switching to the higher doses. This plan, though, is dependent upon approval by the specific medical centers supervising the trial, the ACTG, the National Institute of Allergies and Infectious Diseases and the FDA.
Anabolic Steroid Boosts Weight
A study of nandrolone decanoate (Deca-Durabolin) has found that the anabolic steroid can increase weight in people with HIV, according to a report at the Eleventh International Conference of AIDS in Vancouver (Mo.B.423). The study, conducted in Chicago and New Orleans, randomized 73 people (one woman and 72 men) with low testosterone, fatigue, depression or wasting to intramuscular nandrolone (100 mg per week) or placebo. After twelve weeks, those on placebo received nandrolone for another twelve weeks. All trial volunteers were encouraged to exercise.
After the first twelve-week period, those on treatment had gained a statistically significant total of 1 kilogram compared to an increase of 0.1 kg on placebo. The group receiving nandrolone experienced a greater increase both in fat-free mass (2.3 kg versus 0.5 kg in those on placebo) and body cell mass, another measure of lean body mass (2.3 kg versus 1.0 kg in those on placebo), although the latter measure did not reach statistical significance. A large amount of the gained weight may have been water. Still, according to one of the investigators, Daniel Berger, M.D., "patients felt much better on drug."
Dr. Berger noted that "patients in other parts of the country are doing higher doses, but mostly we wanted to know whether it was safe for our patients." The steroid had no observed negative effect on viral load or immune status and increased levels of hematocrit (a measure of red blood cell level) and hemoglobin, which are low in many people with HIV, especially those on AZT. Levels of free and total testosterone decreased in the treated patients. A few patients withdrew from the study due to mood swings, restlessness or increased libido. Insomnia, impotence and acne were reported by some patients.
There have been fears, based on anecdotal observations, that anabolic steroids promote Kaposi's sarcoma. One volunteer treated with nandrolone developed KS. No changes in KS during the study were seen in three persons on treatment who had KS at baseline -- but one developed pulmonary KS one month after the study concluded. Although, the occurrence or worsening of KS may have little to do with use of the steroid, the study presenter at the International Conference, Dr. Gary Bucher of the Centers for Special Immunology site in Chicago, concluded that "Deca-Durabolin should be used with caution in patients with KS until larger studies are done."
The experience of the one woman in the study illustrates a potential danger of anabolic steroid use in women with HIV. She gained a few pounds, but stopped menstruating. After going off treatment, her weight remained stable, but her periods have not yet restarted.
The same investigators from the Centers for Special Immunology are planning a study comparing nandrolone to growth hormone in patients with wasting. The trial is slated to begin in the next three or four months. Treatment with human growth hormone, which was just approved by the FDA for use in AIDS wasting syndrome, costs about 100 times treatment with 100 mg per week of nandrolone.
Studies Strengthen Link Between HHV-8 Virus and KS
Two recently published studies strengthen the evidence that infection with HHV-8, also known as KSHV or the KS Herpes Virus, leads to the development of Kaposi's sarcoma lesions. Both studies used new blood tests that detect antibodies to the virus.
In one of the studies, blood from 913 people was tested for HHV-8 antibodies by researchers from the University of California San Francisco (Nature Medicine, August 1996;2:862-863, 918-928). The team found strong evidence suggesting that the virus is sexually transmitted. The highest incidence of HHV-8 infection (35%) was seen among gay or bisexual men who became HIV-infected through sex. A high rate of HHV-8 infection was seen among HIV-negative people with syphilis, 8% of whom had antibodies to the virus. Only 3% of HIV-positive people with hemophilia (who acquired HIV through blood products) and only 1% of blood donors without sexually acquired diseases had evidence of HHV-8 infection. Finally, underscoring the link between the herpes virus and the malignancy, the researchers found that 83% of tested patients with KS had antibodies to HHV-8.
Very similar results were published in the New England Journal of Medicine (July 25, 1996; 335:233-41) by Shou-Jiang Gao, Ph.D. et al (in collaboration with Yuan Chang and Patrick Moore -- the Columbia University team that first identified HHV-8). These scientists found antibodies to HHV-8 in 80% of 40 HIV-positive gay men with KS. Only 18% of 40 HIV-positive gay men without KS tested positive for antibodies to the virus, and none of 122 HIV-negative blood donors, 22 patients infected with Epstein Barr Virus, or 20 men with HIV and hemophilia had HHV-8 antibodies.
Perhaps the most important finding of the study was that 21 of the 40 HIV-positive men with KS developed antibodies to HHV-8 six to 75 months prior to the appearance of KS lesions. The researchers believe that the development of antibodies to HHV-8 is not the result of reactivation of a latent virus in immune-compromised patients. They base this view on the very low rate of antibodies in comparable gay men with HIV but no KS. The data suggests that HHV-8 infection is newly acquired, probably through sexual relations.