Resource Logo

(BETA) Report from the 3rd Conference on Retroviruses and Opportunistic Infections


The tone of this year's 3rd National Conference on Retroviruses and Opportunistic Infections, held January 28 - February 1, 1996 in Washington, DC, reflected hope and enthusiasm. Over the past year, there have been great strides in understanding the mechanisms of viral activity and HIV disease progression as well as how to treat and prevent secondary infections. The news about antiretroviral therapy, particularly the protease inhibitors, sparked great excitement at the conference.


How the Virus Works

Anthony Fauci, MD, of the National Institutes of Health, proposed an expanded model of HIV infection and progression.

Primary infection leads to trapping of virus in the dendritic cells of the lymph nodes; CD4 cell interaction with the trapped virus results in infection within the node. Massive replication and widespread dissemination throughout the entire body follows. Intense immune response permits partial immunologic control of viral reproduction, but nonetheless allows viral escape. This escape from immunologic control sets HIV infection apart from most viral infections and necessitates further research.

A state of dynamic equilibrium is reached, formerly known as the "latency" period, during which the infected individual is generally asymptomatic. In this state, accelerated viral replication is opposed by rapid immune cell turnover.

Clonal Exhaustion

Initial HIV infection causes CD8 cell expansion. There are 24 basic families, or types, of CD8 cells that can respond by clonal expansion, or proliferation of cells of a certain family. HIV-specific cells belong to these families in varying numbers, and are the only cells activated to expand. In some people, cell expansion may thus be limited to only 1 family or a few families of CD8 cells. In others, cloning, or cell proliferation, in response to infection may occur in a broad range of families. Among HIV positive people whose response is monoclonal, or involves only 1 family, the expansion of that family is massive. Ultimately, that family apparently disappears entirely despite the fact that HIV is still present. No new family appears, so this disappearance does not reflect mutation. Rather, Fauci proposed, it disappears because it has been "used up," or completely destroyed.

Those who respond to infection with massive monoclonal CD8 cell expansion appear to progress quickly, reflected by a rapid and continuing drop in CD4 count. Those whose CD8 immune response involves many families do better. CD4 counts drop less, then return almost to baseline; further losses are delayed and gradual. An intermediate rate of progression is seen in people whose pattern of clonal response falls between the 2 extremes.

Fauci suggested that viral escape was in part a reflection of clonal exhaustion. Moreover, a person's initial response to infection may help forecast the speed of their progression.


Following sequestration of the virus in the lymph node, local secretion of cytokines helps to create a delicate balance which actually favors HIV replication. Once infection is established, cytokines are released to keep replication down. These include interleukin 2 (IL-2), IL-4 and the chemokines, which have been found to influence the rate of viral replication. Chemokines are produced locally in response to local irritants or other cytokines.

To summarize, once the virus is established in the lymph nodes, the immune system fails to ever completely contain or eradicate it. Rather, a dynamic equilibrium is reached, with intensive viral replication opposing intensive immune response. Fauci postulated 2 newly discovered phenomena underlying the mechanisms that lead to this state: clonal exhaustion, and the existence of newly identified soluble mediators of immune response.


Multiple studies have shown that viral load is predictive of clinical outcome in HIV infection. Its reliability is equal or superior to that of the CD4 cell count. Viral load testing makes it possible to assess disease course and prognosis earlier than has been possible using CD4 counts, offering a rationale for continuation or change of treatment tailored to individual response. Its use in research permits more timely and efficient study of antiretroviral therapies, as efficacy can be seen sooner.

Another important application for viral load testing relates to prenatal care for HIV-infected women and prenatal prevention of vertical transmission. Maternal viral load can help predict the likelihood of transmission of HIV to a child, as well as length of survival of infected infants.

Test results were reproducible for a given patient as long as the same technology was used. A poster presentation by W. A. O'Brien, MD, of the Department of Veteran's Affairs in Los Angeles, and others stated that a change in viral load of greater than or equal to 0.5 log is a useful indicator of response, particularly when accompanied by a 10% or greater change in CD4 counts.

Michael Saag, MD, of the University of Alabama, commented that the cost of the HIV branched DNA (bDNA) assay to participating laboratories was about $60. Laboratories charging patients $150 or more should be asked to justify their pricing.


Carl Grunfeld, MD, of the University of California at San Francisco, reviewed the role of cytokines in HIV-associated wasting. While Grunfeld stated that failure to recover lost lean body mass (muscle) is due to underlying HIV infection, he emphasized that secondary infections are the primary precipitants of wasting. Active secondary infection causes the body to expend more energy and reduces appetite. Skeletal muscle is the primary source depleted to provide necessary protein. "Rapid weight loss with anorexia is a sign of an event, particularly secondary infection in AIDS," Grunfeld stated.

Donald Kotler, MD, of St. Lukes-Roosevelt Medical Center in New York, discussed the use of medical foods with specialized nutrients (Advera, Sustecal, Lipisorb, Ensure, etc.) to combat malabsorption, which he ties to a CD4 count of 200 cells/mm3 or fewer. He recommended combining medical foods with appetite stimulants, and judging response by clinical outcome. He urged consideration of malabsorption in diagnosing HIV-associated wasting.

Morris Schambelan, MD, of San Francisco General Hospital, reviewed the successful use of recombinant human growth hormone (Serostim) to treat HIV-associated wasting. In a study performed by Schambelan and Kathleen Mulligan, PhD, of UCSF, two-thirds of 178 HIV positive participants with evidence of wasting received growth hormone for 3 months, while one-third received placebo. Growth hormone recipients gained weight and lean body mass, while placebo recipients did not. After 3 months, all subjects received growth hormone; those who had first received placebo also gained, while the larger group maintained their initial gains. Reported quality of life improved, as did exercise capacity measured by treadmill work output. Schambelan reported that new data suggest a reduced incidence of secondary infections, particularly Pneumocystis carinii pneumonia (PCP), in the group originally assigned to growth hormone.

Schambelan also discussed the role of androgenic steroids such as testosterone in the treatment of HIV-associated wasting. He reported on a study by Judith Rabkin, MD, of the New York State Psychiatric Institute, and Kotler in which HIV-affected men were given 400 mg of testosterone every 2 weeks. After 11 weeks, average weight had increased, although not significantly. However, lean body mass had increased by 2.6 pounds, which was statistically significant.

Antiretroviral Therapy

While information was presented on many drugs currently in wide use in clinical practice, the real excitement in Washington centered on protease inhibitors. Drugs in this class are the most potent antiretroviral drugs to date. Recently released data indicate that participants in a clinical study who received the protease inhibitor ritonavir (Norvir) experienced fewer opportunistic infections and deaths than matched controls who received placebo.

Emilio Emini, MD, of Merck Research Laboratories, gave the state-of-the-art lecture on this class of drugs. The goal of an antiretroviral therapy, he stated, should be to reduce viral load as much as possible, for as long as possible, while minimizing resistance, with few or no side effects. Issues to consider are bioavailability (how much drug gets inside the body for use), pharmacokinetics (how long a drug lasts in the body, what concentration it reaches, what compartments or parts of the body it enters, and how it breaks down), plasma protein binding (how much drug is bound inertly to other molecules in the blood, and how much is free), and selection for resistance (how the virus changes to avoid the effects of a drug, and what impact that change has on viral response to other drugs).

While Emini focused on indinavir (Crixivan), Merck's protease inhibitor, he reviewed studies available for all 3 protease inhibitors.


Saquinavir (Invirase) is well tolerated and is the only protease inhibitor currently available by prescription.

In study SAQV13330, saquinavir was tested against and in combination with AZT (Retrovir) in people who had never used antiretroviral therapy. After 16 weeks, viral load was reduced by 0.6 log in the AZT-only group, by 0.2 log in the saquinavir-only group, and by 1.0 log in the combination group.

Study ACTG 229 evaluated the effects of saquinavir in AZT-experienced subjects. The group receiving ddC (Hivid) plus saquinavir had lower viral load by 0.3 log. The AZT plus saquinavir group showed no change from baseline; and the group receiving AZT, ddC and saquinavir showed a reduction of 0.5 log. Less than half of the patients in the saquinavir plus AZT arm developed resistance at 1 year; even fewer subjects in the other 2 arms were saquinavir-resistant.


Short term data indicate that Abbott's ritonavir (Norvir) increased survival, decreased incidences of opportunistic infections, and decreased viral load in study participants.

In one reported study, ritonavir was studied in patients with CD4 counts greater than 50 cells/mm3. At doses below 1,200 mg a day, a 1.2 log drop in viral load occurred initially, but the decrease was not sustained. At 1,200 mg a day, however, a 1.2 log drop was sustained for at least 32 weeks. CD4 count rose by an average of 200 cells/mm3, and was also sustained for at least 32 weeks.

When tested in combination with AZT (200 mg 3 times a day) and ddC (0.75 mg 3 times a day), a 2.5 log drop in viral load was sustained at 20 weeks. CD4 count was increased by 100 cells/mm3 from baseline.

One newly reported study, presented at a late breaker session (important studies submitted after the conference deadline), looked at 1,190 extensively pre-treated subjects with CD4 counts less than 100 cells/mm3 (median CD4 count 19 cells/mm3). Entry requirements included the receipt of at least 9 months of approved antiretroviral therapy. Combination therapy, other than ritonavir, was restricted to a maximum of 2 approved antiretrovirals. Ritonavir or placebo were added to patients' current regimens. Study participants were not required to take other antiretrovirals during the study, although most subjects did so.

Secondary markers were examined in a nested subgroup study (a smaller group taken from within a large study, and looked at for a particular purpose) over 16 weeks. CD4 counts rose by 40-50 cells/mm3 in the ritonavir-treated group, and were unchanged in the placebo arm. CD8 cell counts also increased in the ritonavir recipients, and remained high for the 16 weeks of the substudy.

This important trial showed improved clinical outcome in a group whose extensive pre-treatment might predict resistance to therapy, and whose underlying immune competence was poor. In such a group, because progression of disease was more likely, it was possible to show clinical benefit in a short period of time.

Side effects of ritonavir included occasional large increases in uric acid, which sometimes results from tissue breakdown, as well as elevations of AST and SGTP, enzymes which may reflect liver or biliary tract inflammation or obstruction. Creatinine phosphokinase (CPK), an enzyme found in many tissues but especially in muscle, was sometimes transiently elevated. Nausea, vomiting and diarrhea were reported by some participants. Levels of red blood cells and neutrophils in the blood were often markedly improved.


Indinavir (Crixivan), made by Merck, was approved by the FDA in March 1996 as monotherapy or in combination with nucleoside analog drugs for the treatment of late-stage HIV disease.

Indinavir was tested alone and in combination with other approved antiretroviral drugs in people who had never taken AZT. In the indinavir-only group, an initial 2.2 log drop in viral load was reduced to 1.5 logs at 24 weeks. Forty percent (40%) sustained a drop of 2 logs or greater, with a viral load below 200 copies/mL in 15%. One of 13 subjects showed AZT resistance.

The AZT-only group maintained a 0.5 log reduction in viral load at 24 weeks. None had a viral load under 200 copies/mL. Eleven of 16 were AZT-resistant at 24 weeks.

Combination AZT and indinavir caused a 2.5 log drop in viral load, which was sustained at 24 weeks. Sixty percent (60%) sustained a greater than 2 log drop; 60% kept viral load under 200 copies/mL. Only 1 of 13 participants in the combination group developed AZT resistance. Of those remaining on treatment at 48 weeks, 40% still show viral load under 200 copies/mL. (These later figures may be biased by the fact that patients who do well remain on a study, while those who do poorly tend to drop out.)

Six of the people in the combination group have had serial lymph node biopsies. Five showed a 2 log drop in nodal viral load; interestingly, treatment had been interrupted in the subject who did not.

In Merck protocol 19, indinavir alone was tested against a AZT/ddI combination and against indinavir/AZT/ddI. Viral load dropped by 2 logs in the indinavir monotherapy group, by 1.6 logs in the AZT/ddC group, and by 2.9 logs in the triple combination group. CD4 counts in this last group increased by a median 150 cells/mm3. White blood cell counts and absolute neutrophil counts increased. None of the 78 participants developed kidney stones, a known side effect of indinavir that can be prevented by drinking a lot of water.

Roy Gulick, MD, of New York University Medical Center, reported on an ongoing placebo-controlled indinavir study. This study is looking at AZT-experienced patients, many of whom have used other antiretroviral drugs as well. Participants must have CD4 counts between 50 and 400 cells/mm3, a viral load greater than 20,000 HIV RNA copies/mL, at least 6 months of prior AZT experience, and no history of protease inhibitor or 3TC use. Nonwhite subjects comprise 28% of the study population, and women 15%. Ninety-seven (97) subjects are enrolled, grouped into an indinavir arm (800 mg twice a day), a 3TC/AZT arm (standard doses) and a 3TC/AZT/indinavir arm.

While this study is still in progress, the triple combination arm is showing a profound reduction in viral load. Taking indinavir alone has reduced viral load by 40% so far in this group; 3TC/AZT reduces viral load by 20%; the 3 drugs together reduce viral load by 85%, or almost 1 log. At 24 weeks, effects have been sustained, and the regimen is well tolerated. No clinically significant drug interactions have been seen. Only 1 subject has withdrawn from the study. Bilirubin was elevated in 17 of the patients on the indinavir arms. Kidney stones were found in 2 indinavir patients, but did not require study withdrawal.

Based on the promising results of this study so far, all participants will be offered triple drug therapy at the end of 24 weeks.


Protease inhibitors are more powerful than the reverse transcriptase inhibitor drugs which preceded them. They reduce viral load to a greater degree. Their chance for sustained activity depends on their power to decimate the population of virus by limiting viral replication and reducing opportunities for viral mutation. Subtherapeutic drug levels, whether caused by reduced dosing, drug interactions, malabsorption or reduced bioavailability due to other factors, or self-imposed drug holidays, all permit increased viral replication and increased opportunity for mutation and resistance.

Resistance to saquinavir requires 2 viral mutations. Resistance to indinavir or ritonavir requires multiple sequential viral mutations at various sites. Indinavir- and ritonavir-resistant viral strains are cross-resistant to both drugs. Two-thirds or more of such strains are saquinavir-resistant as well. Saquinavir-resistant isolates, however, are susceptible to indinavir and ritonavir.

Based on early evidence, resistance is best avoided by maintaining therapeutic drug levels and by using protease inhibitors in combination with antiretrovirals from other classes. While the concept of treatment with more than 1 protease inhibitor is interesting, powerful drug interactions may result. Barring a clinical trial in a research setting where blood levels can be monitored, the risks of experimentation are high. As Joseph Eron, MD, of the University of North Carolina at Chapel Hill, said during the BETA Live! conference call, "Do not try this at home."

Drug Interactions

Protease inhibitors interact with many other drugs. They increase levels of rifabutin, the use of which is not recommended with this class of drugs. Drug interactions have also been reported with rifampin and ketoconazole. Phenytoin (Dilantin), phenobarbital and other drugs which induce the P450 system (a metabolic pathway for drug breakdown in the liver) speed clearance of protease inhibitors, and may drop levels into the subtherapeutic range. Using protease inhibitors in combination also changes pharmacokinetics and affects drug levels. Ritonavir, for example, when administered with saquinavir, can raise saquinavir levels 8 times. Indinavir can increase saquinavir concentrations even more, to 36 times the expected level.

The bottom line, then, on protease inhibitors, can be reduced to:

* Treat early. * Treat consistently. * Use in combination with other antiretrovirals. * Watch for drug interactions with other HIV-associated medications.

Other Antiretrovirals

A French study reported that d4T in people with early-stage HIV disease with a measurable viral load showed sustained activity. While combination therapy emerged from this conference as the clear treatment of choice, this study showed that benefit can be obtained from early antiretroviral treatment.

B. Gazzard, MD, of Chelsea and Westminster Hospital in London, gave a brief update on the European and Australian Delta Treatment Group studies. Delta I, which compared AZT monotherapy to AZT plus ddI in antiretroviral-naive subjects, showed fewer clinical events and a significant survival benefit in the combination arm. Delta II, which examined AZT-experienced patients, also showed a survival advantage with the addition of ddI to AZT.

Combination ddI and d4T resulted in sustained reductions in viral load and increases in CD4 count. The regimen was well tolerated, with no significant increase in toxicity, even where toxicities overlap.

W.W. Freimuth, MD, of the Howard Hughes Medical Institute, reported on an ongoing study of delavirdine, a non-nucleoside reverse transcriptase inhibitor made by Upjohn Pharmaceuticals, tested in different doses with or without AZT. Subjects had CD4 counts of 200-400 cells/mm3, were largely asymptomatic and were divided between AZT-naive (40%) and AZT-experienced (60%). The mean increase in CD4 count for the combination regimens at 60 weeks was 20-30 cells/mm3 (40-50 cells/mm3 increase in the naive group), with a 0.6 log reduction in viral load at 52 weeks at the higher doses studied (300-400 mg 3 times a day).

Another study, protocol 17, looked at the benefit of delavirdine/ddI treatment compared to ddI alone in a group of 800 largely symptomatic, AZT-experienced subjects. Twenty-five percent (25%) were ddI-experienced as well. CD4 counts at entry ranged from 0 to 300 cells/mm3. CD4 counts in the combination arm were increased by 5-20 cells/mm3 above those in the ddI-only arm at 40 weeks.

In 366 AZT-naive patients in study NUCA 3001, 3TC/AZT combination therapy resulted in a peak drop in viral load of 1.5 logs, with a CD4 count rise of 60-80 cells/mm3. At 52 weeks and beyond, viral load has remained suppressed at 1 log below baseline and CD4 count elevations were unchanged. Fewer adverse clinical endpoints were noted than in the AZT-only arm of the study.

In a follow-up study with AZT-experienced patients, study NUCA 3002, subjects were randomized to AZT/high-dose 3TC (300 mg twice a day), AZT/low-dose 3TC (150 mg twice a day), and AZT/ddC. Both 3TC arms showed decreases in viral load at 52 weeks (about 0.5 log) and increases in CD4 counts (23 cells/mm3 in the high-dose group and 42 cells/mm3 in the low-dose group). During the 52 weeks of the study, no significant difference in progression to clinical events was observed. In sum-mary, in AZT-experienced patients, treatment with AZT/3TC combination therapy is better than with AZT alone, and effects appear to last out to 52 weeks.

When NUCA 3002, a prospective double-blind study, was unblinded, it was found that those who had dropped off the study were those whose viral load had returned to baseline. In other words, the patients clinical judgment mirrored a return to viral load baseline. Thus another marker for progression, in addition to viral load and CD4 count, may be patient intuition.

Resistance to non-protease inhibitor drugs occurs in stepwise increments. It appears to develop faster when drug levels are lower. Strains of multidrug-resistant virus may emerge, but the virus which results will be less fit. For example, development of resistance to 3TC results in an attenuated, less virulent virus with reduced resistance to AZT. Delavirdine selects for a virus that replicates less efficiently.

An interesting study reported by T.G. Tachedjian, MD, of the Macfarlane Burnet Center for Medical Research, showed that HIV resistant to AZT will be sensitive to foscarnet; conversely, the development of foscarnet resistance confers AZT sensitivity.

Again, the appropriate approach is to treat early, while less virus is available to mutate; treat aggressively, at the maximum safe and effective dose; and treat in combination. Convergent therapies, which attack replication at different points, can box in the virus, forcing mutations that purchase its survival at the expense of fitness.

Women and HIV

Although many presentations addressed issues of importance to women, the range of subject matter remained narrow. Most studies focused on vertical transmission, or infection of the newborn by its mother. Those studies that addressed diagnosis and treatment of HIV-affected women for their own sake were chiefly concerned with gynecological matters.

Perinatal Transmission

ACTG 073 examined the role of AZT treatment in perinatal HIV transmission among 419 mothers with CD4 counts of over 200 cells/mm3. As in ACTG 076 (see BETA, June 1995, pp. 41-43), a 3-fold decrease in transmission was seen among the mother/infant pairs that received AZT before, during and after delivery. High maternal viral load, as well as low CD4 cell count, CD4 percentage, and CD4/CD8 ratio were associated with increased risk of transmission. However, there were no viral load levels below which transmission did not occur.

Other factors believed to influence transmission rate mode of delivery (vaginal or Cesarean), duration of rupture of membranes before birth and gestational age (premature or full-term) were not found to be predictive of HIV infection in this study. Thus researchers concluded that there was no way to assure that a fetus could be protected from transmission. In addition, the benefit of AZT treatment appeared unrelated to its effect on viral load in the mother. The value of AZT in blocking transmission may relate more to its effects during delivery and after birth than its effects on the mother during gestation.

The Ariel Project, which also looked at vertical transmission in 151 mothers, found no evidence of any viral load threshold below which infants were not infected, or above which infants were consistently infected.

Other presentations on perinatal transmission addressed breast-feeding (which predisposed to HIV transmission), length of time in birth canal (the first twin born is more likely to be infected), duration of rupture of membranes (proportional to the rate of transmission) and type of delivery. Vaginal birth appears to involve a higher risk of transmission than Cesarean delivery. Data are blurred, however, as many babies delivered by Cesarean section still spend a long period in the birth canal. Therefore, the potential benefit of Cesarean section in reducing transmission may be understated.

Recommendations are to minimize invasive monitoring of the infant, e.g., not using fetal scalp devices, in order to reduce the risk of infection during birth. An attempt to reduce the rate of transmission in Africa by cleansing the birth canal prior to delivery was unsuccessful, despite careful study design.

Cervical Dysplasia

The incidence of cervical dysplasia (cell changes which may indicate cervical cancer) is increased in HIV-infected women; recurrence after treatment is increased 4-fold when compared to HIV negative women. While it remains unclear whether cervical cancer occurs more frequently in HIV positive women, it is clear that, once present, its course is more aggressive. Cancers of the vagina and the vulva (the area outside the vagina) are also more frequent in women with HIV.


Candida, or yeast infections, are often the first sign of HIV infection in women. Herpes simplex outbreaks occur more often, and asymptomatic shedding of herpes simplex virus is common. Pelvic inflammatory disease (PID) occurs more often, can be more severe and may result from different organisms than those found in PID in HIV negative women. Genital ulcers are more frequent.

Menstrual Irregularities

One study reported no HIV-related increase in menstrual irregularities or symptoms compared to matched HIV negative controls.


One interesting study compared length of life after diagnosis in HIV-infected women and men in Puerto Rico. Although matched for CD4 count with men in the study, and although they had fewer AIDS-associated events, women in the study lived only about half as long as men. A thoughtful analysis of social factors showed that women were more likely to have contracted HIV by heterosexual transmission, more likely to be widowed and more likely to live in rural areas and to have dependents. Researchers postulated that widows infected by husbands and left to raise the family had less money, less access to health care and greater responsibilities, and that these factors directly influenced their survival.

Opportunistic Infections


Mark Jacobson, MD, of San Francisco General Hospital, discussed treatment and prophylaxis of cytomegalovirus (CMV). "Treatment is always required," says Jacobson. "Watchful waiting in the presence of disease is an invitation to progression."

Intravitreal implants (drug-containing devices inserted into the eye) are effective, doubling the time to progression of retinitis. However, there are considerable risks; vision may be disturbed or lost as a result of the procedure itself. Jacobson reports a 10% risk of visual impairment as a result of implantation, compared to standard intravenous (IV) treatment with ganciclovir or foscarnet, which offers a 5% risk of blindness. He recommends systemic treatment for peripheral retinal damage unless Zone 1 (central vision) is threatened, in which case implants may be worth the risk. In addition, implants provide no systemic benefit against CMV manifestations in other parts of the body.

Oral ganciclovir as maintenance therapy is not as successful as continued IV treatment. Oral ganciclovir allows more disease, more bilateral disease (in both eyes) and more involvement of Zone 1. However, there may be a role for oral ganciclovir in primary prophylaxis of CMV once CD4 counts fall below 50 cells/mm3. Jacobson suggests CMV PCR studies every 3 months after a CD4 count of 50 cells/mm3 has been reached to screen for CMV in those most likely to benefit from prophylaxis.

In cases of progression, ganciclovir and/or foscarnet may be increased or combined. Co-administration of the standard dose of each has been shown to double the time to progression.


Parvovirus, a viral organism which causes a common childhood illness, was found in the bone marrow of 16% of AIDS patients at autopsy in one series. It should be considered when neutropenia (a shortage of neutrophils, a type of white blood cell necessary for immune defense) is present and no other cause has been found.


C. Van der Horst, MD,of the University of North Carolina at Chapel Hill, reported on new approaches to the treatment of cryptococcal meningitis. He stressed the importance of relieving intracranial pressure with repeated lumbar punctures (spinal taps) or the placement of a shunt. Headaches are relieved and confusion reduced when pressure is lowered.

Amphotericin B is the treatment of choice for cryptococcus; efforts to treat orally with fluconazole or itraconazole have failed to show a greater than 50% response. Unlike the azole drugs, amphotericin B actually kills fungal organisms. Studies conducted by David Stevens, MD, at Stanford University show that there may be synergy (a combined effect of 2 drugs that is greater than the separate effects of both added together) when a growth factor, granulocyte-macrophage colony-stimulating factor (GM-CSF), is used with fluconazole. Following treatment with amphotericin B, fluconazole may be used as maintenance therapy, or secondary prophylaxis; prophylaxis after infection should be at 200 mg per day.

A study from Burundi reported the successful administration of amphotericin B with intralipids (fats used for feeding through the vein). Side effects of amphotericin (fevers, chills and nausea), which may be severe, were resolved, although renal function was not affected.

Fluconazole and amphotericin B may be used in combination to treat cryptococcosis, but not to treat aspergillosis, where the drugs work against each other. Histoplasmosis, when treated with fluconazole, results in a fluconazole-resistant organism which is still sensitive to itraconazole.


Prophylaxis or long-term treatment of thrush, or oral candida infection, with fluconazole results in resistance. It can be briefly overcome with high doses; the effect is transient. Studies of daily vs weekly regimens show more unusual and resistant yeast species, and more resistant candida in the weekly regimens.


In the kickoff lecture at the conference, Barry Bloom, MD, of the Howard Hughes Medical Institute/Albert Einstein College of Medicine, reviewed the extent of tuberculosis (TB) in HIV/AIDS. Rates are increasing, largely as a result of decreased funding for eradication efforts. HIV positive people are 3 times more likely to be infected; HIV positive Latinos are 5 times more likely.

Infections previously had represented reactivation; now many infections are primary, or newly infected, cases. Primary infection with the same strain can occur in large numbers of HIV positive people. Homeless shelters have been shown to be instrumental in such mass infections.

Mycobacterium avium Complex

Incidence, severity of illness and deaths associated with Mycobacterium avium complex (MAC) are increasing. Rifabutin and clarithromycin are each now approved as monotherapy for primary prophylaxis. One study has shown that adding rifabutin to clarithromycin does not improve activity. Clarithromycin or azithromycin for prophylaxis also reduces community-acquired pneumonia, and may be synergistic against PCP. Azithromycin at 1,200 mg per week or 500 mg per day, and clarithromycin at 500 mg twice a day, have been shown to be superior to rifabutin for MAC.

Drug interactions must be considered. Clarithromycin can increase rifabutin levels; rifabutin can lower clarithromycin levels. There are also important drug interactions between these drugs and protease inhibitors; rifabutin should not be given with protease inhibitors, and clarithromycin may require dose adjustment.

While prophylaxis may be more costly than waiting to treat, it is important to remember that MAC infection triggers an intense immune response, and hence triggers wasting and viral replication.

Pneumocystis carinii Pneumonia

PCP prophylaxis delays progression to AIDS and prolongs survival. About 50% of breakthrough infections are found in people with CD4 counts of less than 75 cells/mm3; 75% have counts less than 50 cells/mm3. TMP-SMX (Bactrim, Septra) is better than dapsone, which is better than aerosolized pentamidine. No studies yet show the effectiveness of atovaquone (Mepron) compared to the others for PCP prophylaxis.


Microsporidium can cause intractable diarrhea in people with advanced HIV infection. There are 2 main species; 1 is sensitive to some experimental regimens. New studies are available to help determine the species, and hence guide treatment.


Copyright © 1996 -BETA, Publisher. All rights reserved to the San Francisco AIDS Foundation. Reproduced by permission. Reproduction of this article (other than one copy for personal reference) must be cleared through BETA: PO Box 426182, San Francisco, CA 94142-6182. Tel: 415 487 8060 Fax: 415 487 8069 San Francisco AIDS Foundation, Mail SFAF..

Information in this article was accurate in March 1, 1996. The state of the art may have changed since the publication date. This material is designed to support, not replace, the relationship that exists between you and your doctor. Always discuss treatment options with a doctor who specializes in treating HIV.