Moods Brighten as Statistics Lighten
In February 1997, Treatment Issues commented that the 4th Conference on Retroviruses and Opportunistic Infections, "reflected the gradual accretion of promising developments leading toward fulfilling the now years-long hope of making HIV infection a 'chronic manageable disease'." This year's 5th Conference (held in Chicago on February 1-5) was even more low-key, with unexpected developments exceedingly rare. Many of the reports had in fact been anticipated in previous months' Treatment Issues. The big news lay in epidemiology and basic science, where last year's triumphs and conundrums found sequels in this year's reports.
A conspicuous feature of recent AIDS/HIV conferences has been the increasing banter from the podium. The speakers' relaxed, frequently jocular demeanor reflected the latest statistics on the dropping incidence in opportunistic infections and deaths. Kevin De Cock of the federal Centers for Disease Control and Prevention established the upbeat atmosphere on the conference's first full day by announcing that AIDS-related deaths in the US were down 45% between the first halves of 1996 and 1997 (lecture L2). That's on top of a 23% decline between 1995 and 1996.
One factor in the falling death rate is the reduced HIV transmission in the late eighties, but reports from the conference showed a strong link between triple drug therapy and extended survival. In the Adult Spectrum of Diseases cohort, an observational study of 18,000 people with AIDS sponsored by the CDC, the risk of death was 2.5 times lower for those on triple combination therapy than for those on monotherapy (slide presentation 10). Prophylaxis for PCP or MAC further reduced the risk of death by about 20%. Similarly, reports from the New York City Department of Health described a 33% decrease in AIDS deaths between the last half of 1996 and the first half of 1997, following a 29% drop in 1996 over 1995 (slide presentation 9b). (Comparable declines occurred for men and women and for every ethnic group.) A follow-up report from the city argued that the reduction in death was closely linked to use of protease inhibitors (late breaker 7) based on a case-control comparison of 150 people who died of AIDS in 1996 and 150 people with similar characteristics not known to have died that year. Thirty-one percent of those who survived 1996 received protease inhibitors versus only 11% of those who died. Two-thirds of both groups were taking some sort of antiviral therapy.
The decline in deaths was an echo of the reduced incidence of all major opportunistic infections. The best data on this at the Conference came from INSERM, the French national health research institute, which has tracked the experiences of 60,000 people with HIV (poster 182). The incidence of all major OIs tumbled between the first halves of 1996 and 1997: The rates of CMV and cryptosporidiosis were down 80% and the rates of KS and PCP declined 75% and 70%, respectively. The brain infection PML had by far the least reduction, about 29%. A 1,400-person observational cohort at Johns Hopkins University in Baltimore found that here, too, protease inhibitor use correlated with a 60% reduced risk of opportunistic infections (poster 184).
(These observations were seconded in an article that appeared at Treatment Issues' press time, in the March 26, 1998 New England Journal of Medicine. The authors measured a greater than two-thirds fall in deaths and OIs from 1995 through the first half of 1997 among 1,255 persons with at least one CD4 count less than 100. Death and illness rates dropped by the same proportion in all demographic groups. The extent of the reductions was associated with the "intensity" of therapy -- monotherapy, dual nucleoside analogs and protease inhibitor-containing combinations each provided additional benefit.)
The Johns Hopkins group also noted that OIs were nonexistent in people whose CD cell count rose above 200 as a result of therapy. Other studies found a significant reduction in OI risk after CD4 cell count increases of as little as 50 (lecture S29). Such observations led Richard Chaisson, M.D., of Johns Hopkins University to exclaim, "The best prophylaxis for opportunistic infections is antiretroviral therapy!"
The reversal of disease progression that parallels therapy-induced improvements in CD4 count and viral load is widely taken as a sign that the immune system is recovering its capacities. The Retrovirus Conference presented evidence from a wide range of laboratory studies showing that such "immune reconstitution" is taking place to a surprising extent, though the ultimate limits of this reconstitution remains to be determined.
Ashley Haase, M.D., of the University of Minnesota started the conference off with a description of immune repair among people taking part in an AZT/3TC/ritonavir trial (lecture S1 and late breaker 13). Researchers took frequent lymph tissue samples (lymph node and tonsil biopsies) from these participants to examine infected and uninfected cell populations and tissue structure using special staining techniques to produce graphic photographs. According to Dr. Haase, persons with HIV contain about 100 billion CD4 cells in their lymph tissue, which is approximately half of normal and is unexpectedly high considering the CD4 cell depletion in the blood. About 0.12% of lymph CD4 cells are actively proliferating in those with HIV whereas the normal percentage is about 0.4%. In the trial volunteers, CD4 cell repopulation took place at an initial rate of 80 million cells per day. This reproduction roughly offsets the number of cells previously lost per day to HIV and is about equal to the same low rate seen in adult cancer patients after extensive chemotherapy. It is not enough to provide a major rebound in CD4 count.
Most stunning is the reassembling of normal lymph tissue architecture within six months of therapy. In place of the now well known disordered mass of lymph node tissue, a healthy structure emerged. But the functionality of the restored tissue remains to be determined. Dr. Haase summarized his observations by saying, "In some individuals you see almost a full restoration of immune function; in others less or none. It depends on several factors, including age and the extent of damage -- although we don't know the point of no return. Some individuals with a CD4 count of zero respond quite well."
One laboratory technique for gauging the return of immune responses is to measure how well patients' immune cells proliferate when exposed to particular immunogenic proteins such as tetanus toxoid or candida (yeast) antigen. A University of Pittsburgh group led by Charles Rinaldo followed five persons receiving AZT/3TC/indinavir as part of Merck-sponsored protocol 039 (poster 169). Little change in proliferation responses was seen for the first 36 weeks, but by the second year significant CD4 and CD8 cell responses became evident. Brigitte Autran and colleagues in Paris followed up their initial observations on immune reconstitution presented at last year's conference with a cell proliferation study similar to the one from Pittsburgh (poster 172). The Paris analysis also found a return in proliferation responses, this time after the third month of therapy.
Neither group of researchers could find any renewal of proliferation in response to HIV antigens, however. Bruce Walker, M.D., of Massachusetts General Hospital elaborated on this point in a major conference address (lecture L4). Dr. Walker noted that in cell cultures, human CD8 cells are able to virtually wipe out HIV infection, yet in the body these cells (of both the cell-killing, or CTL, and virus-suppressing varieties) tend to peter out. In his studies of long-term nonprogressors, Dr. Walker found that such people uniquely retain high CD4 cell proliferative responses to HIV. In the body, these cells produce the cytokines that stimulate and maintain the CD8 response. He has also found that combination antiviral therapy in people who have contracted HIV within the last six months seems to protect the anti-HIV CD4 cells. HIV presumably kills off these cells during primary infection as they become activated and therefore highly vulnerable to the virus. Starting highly active antiretroviral therapy ("HAART") after the six-month window of opportunity does not seem to allow their reappearance, at least over the two-year periods observed so far.
For Once, Naïveté is Welcome
The absence of this highly desirable immune response may indicate that the immune system has suffered irreparable damage. Assuming that infection has obliterated the anti-HIV cells, they could only return as part of an overall recreation of a new, "naïve" CD4 cell population. The naïve cells are a diverse group, each responding to specific disease antigens that the body has not encountered before. They are mostly created during childhood and mature in the thymus gland, an organ that becomes increasingly nonfunctional with age. Several characteristic surface receptors distinguish naïve cells from the memory cell clones warehoused after past battles with infectious agents. Dr. Haase commented that when he and his colleagues examined lymph tissue, naïve CD4 cells constituted only 26% of the total, compared to 45% in HIV-negative individuals, but their number gradually increases by the end of the first year of therapy-engendered viral suppression.
Michael Lederman, M.D., of Case Western University argued in a lecture (S30) that most of the CD4 count rebound immediately after the commencement of potent therapy is due to a redistribution of cells from the lymph nodes. As HIV levels drop, there is a reduction in inflammation and markers for cell activation (poster 159). Dr. Haase and others also have noted declines in the tendency for cells to die upon activation (through programmed cell death or apoptosis, a process that usually clears the body of unneeded or improper immune activation). CD4 counts in the blood immediately shoot up, with both the memory and naïve components increasing, but there is little evidence of any new cells appearing. After the first few weeks, the CD4 count is largely stable through the first year of HAART, with the proportion of naïve cells slowly increasing, a trend documented by a Dutch report (slide presentation 20).
Increases in naïve cells might indicate that new cells are coming from the bone marrow and maturing in the thymus. That does happen in adult cancer patients after chemotherapy, albeit at a very slow rate (lecture L7). But the thymus, the major site of CD4 cell maturation, is heavily damaged during HIV infection. As disease progresses, it becomes more and more invaded with infiltrate containing mature white blood cells and its structure collapses, according to lecture by Barton Haynes, M.D., of Duke University (lecture S44). Immune cell maturation in the thymus, which normally is present at residual levels after childhood, declines with progressive HIV infection, and the organ ends up resembling other lymph tissue.
Besides examining the thymus glands of people with HIV, Dr. Haynes has also examined people with HIV whose thymuses have been previously removed. One person he described was placed on potent HIV therapy in 1996 and responded like a normal patient, with increases in naïve and memory CD4 cells. Another, who had been thymectomized when he was 16, long before he was infected with HIV, was a long-term survivor and had normal naïve cell levels even before HIV treatment. To Dr. Barton, these patients' experience indicated that some thymus-independent CD4 maturation does exist and is available to persons with HIV.
Residual Virus, or the Receding Eradication Horizon
The ability to regenerate the naïve CD4 population and particularly an anti-HIV immune defense may be critical to controlling the virus. A large amount of residual HIV replication occurs even with apparently successful HAART therapies. Dr. Pat Bucy of the University of Alabama calculated that at the minimal viral load of 20 copies/ml, 350,000 cells in the body are actively producing HIV (lecture S17 and posters 178, 519 and 574). But in various HAART trials, 25% to 60% of participants still have detectable viral loads by the new ultrasensitive assay that is accurate down to 50 copies/ml (see table).
In addition, several different groups using a variety of techniques all reported a trickle of HIV coming out of long-lived latently infected cells (see Treatment Issues, November 1997 and conference lectures S15a, S16, S17 and S19, among other presentations). David Ho, M.D., of New York's Aaron Diamond Research Center, is now estimating that some 200 latently infected cells turn on each day at the start of therapy (lecture S16). He now says that completely suppressive treatment must continue for 5 to 20 years to eliminate all these cells, considering their long lifespans. At last year's conference, his estimate was a minimum of only 2.5 years. Dr. Anthony Fauci's lab at the National Institute of Allergy and Infectious Disease is in close agreement with Dr. Ho this year: Tae-Wook Chun, a colleague of Dr. Fauci's, estimated at the conference that the required length of treatment was six to eight years, according to his lab's observations of latently infected cells (slide presentation 515).
Various proposals have been put forward to accelerate the activation and degeneration of latently infected cells. Anthony Fauci pointed out at the conference that trials involving IL-2 might already furnish some data along this line (lecture S43). (In test-tube experiments, IL-2 can stimulate latent cells to produce virus, but infection of new cells and further viral production could be terminated by HAART.)
Past research has observed that tetanus vaccinations raise viral load, but has not characterized the source of the increased virus, whether from newly infected cells (activated cells are more likely to be infected by HIV) or by cells that were formerly latently infected and now activated. A small study at Dr. Fauci's lab found evidence that the extra HIV produced under the influence of the vaccine is of an older variety than the one currently predominating in the patient (slide presentation 286). This indicates that the HIV is "archival" virus from cells that were latently infected.
In one small study, G-CSF (filgrastim) injected to stimulate immune progenitor (stem) cell production for a gene therapy trial raised viral load briefly by up to ten-fold (1 log) in two people with viral loads below 50 copies (slide presentation 514). A planned ACTG trial, protocol 378, is intended to test this "flushing" mechanism as a means of eliminating latently infected cells. Twenty-four participants with viral loads over 20,000 and with no current anti-HIV therapy will receive AZT/3TC/nevirapine/indinavir for 48 weeks. During weeks one to 14 they will also receive either GM-CSF, IL-12 or no immune modulator. IL-12 and GM-CSF stimulate cell activation while promoting HIV production in latently infected lymphocytes and macrophages, respectively. At the same time, they have protective effects. IL-12 enhances CTL activity while GM-CSF renders monocytes and macrophages more resistant to entry by HIV.
A completely different approach was advocated by Franco Lori, M.D., of the RIGHT Institute, who has been a pioneer in the use of hydroxyurea to potentiate the nucleoside analog ddI (see Treatment Issues, November 1997). Hydroxyurea works by restricting the cellular supply of adenosine, for which ddI is a defective substitute. According to Dr. Lori, hydroxyurea plus ddI should be especially useful in preventing the build-up of latently infected CD4 cells because the combination is the only one that penetrates and affects quiescent cells that can shelter latent virus.
At the Retrovirus Conference, Dr. Lori presented further reports on the hydroxyurea/ddI/indinavir study carried out by Dr. Heiko Jessen of Berlin (again, see Treatment Issues, November 1997). Lymph node samples of three patients who were treated during primary infection and whose HIV became undetectable were sent to Robert Siliciano of Johns Hopkins University to search for latent infection. Two of three were still undetectable in Dr. Siliciano's standard assay involving five million cells. But when 60 million cells were checked, a few infected cells were found (late-breaker 11, lecture S15a and poster 655). The HIV in these cells may well be nonviable, however. If the HIV is inactive, then the Holy Grail of viral eradication would have been achieved in these patients. (One had already been off treatment for a year.) Whether similar results could occur in persons beyond the primary infection stage is completely unclear.
It's Alright to Fail
Steven Deeks, M.D., reported further observations on 143 San Francisco General Hospital patients who were on protease inhibitor triple combinations but failed to achieve or maintain viral loads below 500 copies/ml (the limit of quantification for the bDNA assay in use and equivalent to about 1,500 copies/ml by PCR). Even at 18 months from the commencement of therapy and as long as eight or nine months after viral rebound, these people had significant CD4 count rises over baseline and many still had substantially reduced viral load as well (poster 419). The improvements varied according to the size of the initial viral load response. The numbers look as if a new, healthier equilibrium has been established between HIV and a partially restored immune system. (Possibly too, drug-resistance mutations had impaired these people's virus.)
As long as there is appreciable HIV replication, the virus will continue to evolve to more fit, drug-resistant forms. The immune system will continue to suffer a certain amount of incremental damage, too, even though the persons involved are outwardly healthy. Most experts, including Dr. Deeks, expect these patients to eventually deteriorate. But it may be that one can stabilize HIV infection for considerable periods of time even if therapy is not maximally suppressive. In the meantime, we are accumulating further experience with salvage therapies (to be touched on next month). Also new anti-HIV agents will continue to increase treatment options (see Treatment Issues, January 1998, plus the following article in this month's issue).
Poster Session 12290: Sixteen-Week Followup of Indinavir Sulfate (Crixivan) Administered Every 8 Hours versus Every 12 Hours in Combination with Efavirenz
This article was provided by Gay Men's Health Crisis. It is a part of the publication GMHC Treatment Issues. Visit GMHC's website to find out more about their activities, publications and services.