Published in the Bulletin of Experimental Treatments for AIDS April 1999 issue, by the San Francisco AIDS Foundation.

Selected Highlights from the 6th Conference on Retroviruses and Opportunistic Infections
-- Main Page
-- Clinical Research
-- Protease Inhibitor Side Effects
-- Anti-HIV Drugs and Combinations
-- HIV Transmission and Origins
-- HIV-Related Conditions

April 1999 Table of Contents

Main Page

beta@sfaf.org

Selected Highlights from the 6th Conference on Retroviruses and Opportunistic Infections

Clinical Research

By Harvey S. Bartnof, MD

• Multiply Exposed, HIV Negative Persons May Carry HIV Genetic Material
• HIV Infection Controlled in Six Persons
• People on HAART with HIV Rebound Still Have CD4 Cell Count Benefits
• Genotypic Resistance Testing Improves Response to Therapy
• Drug-Resistant HIV Continues to Increase in Newly Infected Persons
• Thymus Gland Function in Adults

Multiply Exposed, HIV Negative Persons May Carry HIV Genetic Material

• HIV genetic material was detected in resting CD4 cells.
• Results indicate that latent HIV infection can occur without positive antibody status.

One of the more interesting presentations at CROI was authored by researchers from the Fred Hutchinson Cancer Research Center at the University of Washington in Seattle. Research in the past several years has documented the existence of a small number of HIV negative persons who had a large number of high-risk, unprotected sexual exposures to HIV (see BETA, April 1998).

Examples include a few men who were sexual partners of gay men with AIDS, and female prostitutes in certain African countries with high rates of HIV infection. These people were shown to be repeatedly negative for HIV antibodies in their blood, and they had negative HIV viral culture tests. Many of them had HIV-specific cellular immune responses against HIV antigens in the laboratory; however, none were ever found to have HIV specific DNA (genetic material) incorporated into their human chromosomes. Such HIV "resistance" had been partially linked with specific genetic markers.

In an oral presentation at CROI, the Seattle researchers reported on 37 individuals identified in 1996 who had multiple high-risk, unprotected sexual contacts with HIV-infected partners during the next two years. All 37 remained HIV antibody negative. Viral load tests for HIV RNA in the blood plasma of these people also were persistently negative (below the limits of quantitation of the tests used). Somewhat surprisingly, all 37 persons repeatedly had HIV genetic material detected in their resting CD4 cells. The genetic material included all three of the main HIV genes: envelope (env), core (gag), and polymerase (pol). Detection of the HIV genes was performed in several laboratories and at different times to help exclude potential cross-contamination from other laboratory specimens. The specific HIV DNA sequences detected changed very little during the two-year observation period.

The authors concluded that their results represent "the first demonstration that latent HIV-1 infection can occur without subsequent seroconversion." They also concluded that latent HIV infection can occur "extremely early, before the onset or despite the absence of overt systemic infection." In addition, the researchers noted that HIV replication did not occur, or did so only at extremely low levels. Lastly, the authors noted that this latent, antibody-negative state seemed to protect these individuals from later infection with other HIV strains.

These findings add to our understanding of the spectrum of HIV infection and have significant implications for vaccine development and new treatment approaches.

HIV Infection Controlled in Six Persons

• HIV infection was controlled in six persons after two to three cycles of HAART.
• All six started anti-HIV therapy during very early HIV infection.
• After each successive therapy cycle, the time until HIV viral load rebound increased.
• Similar findings occur in a monkey model of AIDS.
• This experimental approach to anti-HIV therapy is dangerous at the present time for almost all persons.

Immediately prior to CROI, AIDS researchers and physicians were discussing a front-page report that appeared in the January 25, 1999 issue of the Wall Street Journal. This article described seven persons with early-stage HIV infection who appeared to have "controlled" their respective HIV infections after starting, stopping, restarting, and then restopping HAART. Several of these persons were described in detail in three separate presentations at the conference. Similar results also were reported concerning a monkey model of AIDS. It appears that the initial increase in HIV RNA viral load after first stopping HAART may have acted as an "internal vaccine" that may have primed the immune system to subsequently control HIV replication after HAART was subsequently restarted and restopped.

The first report was an update of a late-breaker presentation from the 5th CROI in February 1998 (see BETA, April 1998). Julianna Lisziewicz, MD, Franco Lori, MD, and colleagues from the RIGHT Institute in Washington, DC, described a person who is now referred to as "the Berlin patient." Fifty-seven days after HIV infection, a 29-year-old gay man was treated with indinavir (Crixivan), ddI, and hydroxyurea. His HIV RNA viral load decreased from 89,000 copies/mL to undetectable (limit of quantitation 500 copies/mL). Due to epididymitis (testicle infection), the man's anti-HIV combination regimen was stopped for eight days, and his HIV viral load rebounded to 5,356 copies/mL.

The acute epididymitis resolved with standard antibiotic therapy during those eight days, and he subsequently restarted the same anti-HIV therapy. He continued to take his medications for four months. During that time, his viral load again became undetectable. Then, due to acute hepatitis A, the man stopped his antiretroviral therapy a second time. During the 17 days of hepatitis A illness, his HIV viral load was undetectable by one measurement. He then restarted his triple anti-HIV regimen and took it irregularly for five weeks. After a total of just under six months from first starting anti-HIV therapy, the man decided to stop his treatment permanently.

Somewhat surprisingly, his viral load did not rebound, and it remained undetectable for the next year and a half. It has been two years since the man first started anti-HIV therapy (i.e., data is now available for two years for this individual). The three treatment periods totaled 176 days of antiretroviral therapy.

Detailed laboratory testing revealed that HIV DNA was detectable in the man's resting CD4 cells and in lymph node cells, but at extremely low levels. His CD4 cell count became normal, as did the number of naive T-cells (those that are able to respond to new infections). His T-cells also demonstrated very potent anti-HIV responses when tested in vitro.

Two other persons who appear to have "controlled" their HIV infection were treated at the Aaron Diamond AIDS Research Center at Rockefeller University in New York City. This report was authored by G.M. Ortiz, MD, and colleagues including David Ho, MD. These two persons were part of a group of 12 HIV positive persons first described by Martin Markowitz, MD (see BETA, March 1997). Both of them had been infected with HIV for less than three months and were treated with ritonavir (Norvir), AZT, and 3TC. Due to poor adherence, both stopped therapy, subsequently restarted, and then later stopped therapy again. The first course of antiretroviral therapy led to undetectable HIV RNA viral loads in both persons. After the first treatment discontinuation, both experienced viral load rebound. Fortunately, when their anti-HIV regimen was restarted, their viral loads again became undetectable.

When the two persons later stopped anti-HIV therapy again, their viral loads did not rebound, and their viral loads have remained undetectable for 14 and 21 months, respectively. Interestingly, the researchers found that the anti-HIV cytotoxic T-lymphocyte (CTL) responses of both persons increased after their drugs were first stopped.

The authors also described a third person from the original group of 12 who, after a second discontinuation of therapy, maintained an undetectable viral load for four months and then rebounded. A fourth person who employed a similar start-stop-restart-restop treatment pattern immediately experienced HIV viral load rebound. These last two participants demonstrate that this treatment pattern is not effective for all people. The Aaron Diamond AIDS Research Center researchers concluded that "intermittent drug discontinuation can be associated with boosting of HIV-1-specific CTL responses, which may contribute to the prolonged suppression of viral replication when drug therapy is stopped [for those with early HIV infection]."

Three other HIV positive persons with a similar story were described in a late-breaker presentation by Lori. (He also reported on three macaque monkeys with very early simian immunodeficiency virus (SIV) infection that demonstrated a similar clinical course with successive cycles of anti-SIV therapy.)

The three HIV positive people were treated with HAART (ddI, hydroxyurea, and either a protease inhibitor or d4T) for three weeks. Then, after one week of no therapy, two successive three-month treatment cycles were given. The second and third cycles were started when the persons' HIV viral load levels rebounded to greater than 5,000 copies/mL. Their viral loads again became undetectable (limit of quantitation 400 copies/mL) when HAART was restarted each time. After the first treatment cycle (three weeks), viral load rebound occurred after seven days. After the second treatment cycle (three months), viral rebound occurred after 14 days. After the third treatment cycle (again three months), viral rebound occurred after 37 days. The pattern of increasing time until viral rebound after each successive course of anti-HIV therapy is striking. Lori concluded that "serial treatment interruptions may prolong the time to viral rebound in humans."

Lori then described three SIV-infected macaques with early infection that were treated with successive cycles of hydroxyurea, ddI, and PMPA (related to adefovir [Preveon]). As in the human study, the time until viral load rebound increased with each successive treatment cycle. After each successive cycle, viral load rebounded to progressively lower maximum levels (lower viral load set-points).

The monkey report plus the three reports of six persons who appear to have "controlled" their HIV infection with successive cycles of anti-HIV therapy add to our understanding of HIV infection and disease progression. Understanding the mechanisms involved may expand the options for HIV treatment and the possibilities for an effective anti-HIV vaccine. Several research institutions are already expanding studies that will examine similar strategies in HIV positive persons with early infection.

It must be emphasized that all persons described in these reports had been HIV-infected for less than three months. Stopping anti-HIV therapy in people infected for longer periods almost always leads to immediate viral load rebound. People should not try this approach until more is known about the mechanisms involved. Stopping anti-HIV therapy almost always leads to viral load rebound and the risk of developing drug-resistant virus.

Lisziewicz, J. and others. Immune control of HIV after suspension of therapy. 6th CROI. Abstract 351.

People on HAART with HIV Rebound Still Have CD4 Cell Count Benefits

At the 37th Interscience Conference on Antiretroviral Agents and Chemotherapy (ICAAC) in September 1997, Steven Deeks, MD, from San Francisco General Hospital presented disappointing evidence that over half the people followed at the hospital experienced HIV viral load rebound ("failure") on HAART regimens that include a protease inhibitor (see BETA, January 1998).

Now, Deeks has presented data indicating that despite such rebound, CD4 cell counts often remain increased and that many people continue to derive clinical benefit. Deeks and colleagues have also measured the half-life (the amount of time required for an original amount to be decreased by half) of CD4 cells in various groups. They found that in their HAART-treated participants, the half-life of CD4 cells was almost twice as long as that of HIV positive persons not taking HAART who have similar HIV viral loads.

The hallmark of HIV infection is a progressive decline in the CD4 cell count resulting primarily from a sustained high level of HIV replication. CD4 cell count decreases occur due to decreased production of CD4 cells, increased destruction of these cells, or both. Through glucose labeling experiments, Deeks and colleagues measured the mean half-life of CD4 cells in three different groups of people: seven untreated HIV positive controls, nine persons successfully treated with HAART, and four persons taking HAART who experienced HIV viral load rebound (a viral load greater than 5,000 copies/mL) for one year yet maintained a CD4 cell count greater than 100 cells/mm³ above their pretreatment level. The mean viral load measurements in the three groups were, respectively, 94,000 copies/mL, undetectable (limit of quantitation 400 copies/mL), and 112,000 copies/mL. The mean CD4 cell counts in the three groups were, respectively, 342 cells/mm³, 509 cells/mm³, and 266 cells/mm³. The CD4 cell half-lives for the three groups were, respectively, 24 days, 77 days, and 43 days.

These results indicate that CD4 cells in HAART-treated persons with viral load rebound survive almost twice as long as those in untreated HIV positive persons with similar viral loads. However, the increased CD4 cell half-lives in persons with viral load rebound are still not as long as the half-lives of CD4 cells in HAART-treated persons with sustained undetectable viral load. Limitations of this study include the "snapshot" nature of the report. Doing sequential analyses of the participants over time will reveal long-term trends. It is quite possible that the increased CD4 cell half-lives in HAART-treated persons with viral load rebound will not persist over several years (see also BETA, July 1998).

Genotypic Resistance Testing Improves Response to Therapy

Whether the results of drug resistance testing will have any clinical benefit for HIV positive persons remains to be proven (see BETA, January 1999). Interim results from the Terry Beirn Community Programs for Clinical Research on AIDS (CPCRA) study 046 suggest that there are benefits to having access to genotypic resistance test results when deciding on a new anti-HIV treatment regimen after a person experiences viral load rebound. The study authors, John Baxter, MD, and colleagues from the University of Medicine and Dentistry of New Jersey, indicated that short-term results were encouraging. A total of 153 persons were enrolled in the trial after experiencing HIV viral load rebound while taking a protease inhibitor-containing regimen that also included two nucleoside analogs. The participants were randomized into two groups. In group 1, the results of a genotypic resistance test (ABI brand) along with expert recommendations for a new regimen were made available to the treating physician. In group 2, genotypic testing was performed, but the results were not given to the treating physician. Viral load changes in the two groups after four to eight weeks of the new regimen were averaged and compared.

The participants in group 1 had a mean viral load reduction of 1.2 log copies/mL. This was a significantly greater reduction than experienced by those in group 2, who had a reduction of 0.6 log copies/mL. The percentage with an undetectable viral load (limit of quantitation 500 copies/mL) in group 1 (50%) was over twice the percentage in group 2 (23%). These results were statistically significant. The benefits of genotypic resistance testing occurred at all CD4 cell count levels. As expected, the viral load response in both groups correlated with the number of active drugs (as determined by resistance testing) prescribed in the new regimen. The authors concluded that the results of genotypic resistance testing with expert interpretation led to significantly improved short-term HIV viral load responses. More studies will be needed before the Food and Drug Administration (FDA) approves these tests.

Drug-Resistant HIV Continues to Increase in Newly Infected Persons

• Decreased drug sensitivity is present in 21-28% of newly infected persons.
• First report of protease inhibitor-resistant HIV in semen is documented.
• First case of a newborn with multidrug-resistant HIV is reported.
• First case of heterosexually transmitted HIV resistant tonevirapine and delavirdine is documented.

Two late-breaker reports presented data showing that the rate of drug-resistant HIV in recently infected persons continues to increase. The results suggest that drug resistance testing will likely have a significant role when starting therapy in persons known to have been infected within the last one to three years.

The first presentation was from researchers at Virco, Inc. in Belgium. Virco receives samples from HIV-infected people throughout the developed world, but mostly from Europe and North America. The VircoGEN test measures genotypic resistance to HIV, while the Antivirogram measures phenotypic resistance. Genotypic resistance refers to genetic changes in the virus known to be associated with drug resistance, while phenotypic resistance refers to resistance detected when HIV is grown in the laboratory in the presence of an anti-HIV drug. Both tests are moderately expensive and are not currently FDA-approved.

Blood samples were analyzed from 114 persons known to be infected with HIV between 1996-1998. All were naive to anti-HIV therapy. The overall rate of any genotypic or significant phenotypic resistance was 21%. The rates of resistance were highest for non-nucleoside reverse transcriptase inhibitor (NNRTI) drugs, with lower rates for nucleoside analogs and protease inhibitors.

The second presentation was authored by researchers from several medical universities in the U.S. Phenotypic resistance testing was performed at ViroLogic, Inc. in South San Francisco. The ViroLogic test is not FDA-approved at this time. Newly infected persons from five cities (San Diego, Los Angeles, Dallas, Denver, and Boston) were evaluated. All 69 persons had been infected late in 1997 or in 1998. The results showed that decreased sensitivity to any anti-HIV drug occurred in 28%. Reduced sensitivity to the protease inhibitor drugs, nucleoside analogs, and NNRTIs occurred in 13%, 3%, and 17%, respectively.

The results of these two studies indicate that approximately one recently infected person out of every four or five will have some degree of anti-HIV drug resistance. Given the rates of poor adherence among many HIV positive persons taking HAART and the recently reported increases in unsafe sexual behavior, the results of these studies are not unexpected. They add to the weight of evidence in favor of drug resistance testing of newly infected HIV positive persons before starting HAART, and point to the importance of continued development of new anti-HIV drugs that will be effective against virus strains that are resistant to current drugs.

In a related presentation, researchers from Harvard Medical School documented the first report of protease inhibitor-resistant HIV in semen. Previously, protease inhibitor-resistant HIV had been detected in blood plasma and cerebrospinal fluid (the fluid surrounding the brain and spinal cord), but not in semen. Participants in this analysis were from AIDS Clinical Trials Group (ACTG) amprenavir (Agenerase) studies 347 and 850. The presence of amprenavir-resistant HIV in semen was associated with an increase in semen viral load. The results are significant in terms of potential sexual transmission of protease inhibitor-resistant HIV strains.

The first case of multidrug-resistant HIV in a newborn was presented at CROI by Victoria Johnson, MD, from the University of Alabama. The infant's mother was from North Carolina and had adhered poorly to her HIV treatment, which included various combinations of reverse transcriptase inhibitors (RTIs) with or without a protease inhibitor. The mother's HIV viral load was inadequately suppressed. After the baby was born on January 31, 1998, blood testing on day 24 revealed mutations indicating HIV that was resistant to both protease inhibitors and RTIs. The infant had taken AZT as per the ACTG 076 protocol. At age seven weeks the infant's regimen was changed to d4T/3TC/nelfinavir (Viracept).

The first case of heterosexually transmitted HIV that is resistant to NNRTIs was also reported at the conference. Researchers from the University of Cincinnati documented that an HIV positive man transmitted HIV to his 19-year-old female partner in 1996 during unprotected vaginal intercourse. Both had HIV strains that were resistant to delavirdine (Rescriptor) and nevirapine (Viramune), with the usual K103N resistance mutation. The man had been poorly adherent to his anti-HIV regimen, a factor clearly linked with the development of resistance. Fortunately, the HIV strains of these two persons remained sensitive to protease inhibitors. The woman subsequently responded to treatment with indinavir/d4T/3TC.

All of these reports indicate that drug-resistant HIV is increasing, and that resistance testing of recently infected persons likely has merit.

Thymus Gland Function in Adults

The thymus gland is known to be an integral part of immune system maturation and function (see BETA, January 1999). Classical immunology has long held that the thymus gland involutes (shrinks) and becomes inactive by adulthood. Recent research has challenged the traditional dogma. At CROI, researchers from the Aaron Diamond AIDS Research Center provided more evidence that the thymus remains active in adults. They also showed that HIV decreases thymus gland function and that HAART improves the output of immune system cells by the thymus.

The authors, including David Ho, MD, used polymerase chain reaction (PCR) technology in a new way to detect a small rearrangement in the genes for the T-cell receptor that occurs in newly produced thymus cells called thymocytes. The researchers concluded that their technique will further contribute to the understanding of HIV disease and the aging process of HIV negative persons.

Harvey S. Bartnof, MD, has been a member of the Scientific Advisory Committee of the San Francisco AIDS Foundation since 1987.

Page last updated 1 June 1999