The discovery of HIV in the mid-eighties led to an intensive search for therapies that might inhibit the virus' life cycle, a search that eventually produced the sixteen antiretroviral drugs that are on the market today. Finding treatments that might work by improving the function of the immune system -- immune-based therapies (or IBTs) -- has proven a more daunting task, mainly because the mechanisms by which HIV impairs immunity are still not fully understood. Without that understanding, IBTs have largely been shots in the dark, with some aiming to improve overall immune function (and thus prevent or delay opportunistic infections), and others attempting to specifically improve the immune response to HIV. Over the years, many approaches have been proposed and studied, sometimes to great fanfare, but all have so far failed to demonstrate any measurable health benefit. As yet, there are no IBTs approved for the treatment of HIV infection or AIDS.


Among the first IBTs to be studied were cytokines, the chemical messengers of the immune system. Interleukin-2 (IL-2) is the only such treatment to have progressed to the final stages of clinical trials (see article on IL-2). Many others have been investigated as possible therapies for HIV. Gamma interferon is produced by immune system cells in response to infection, and activates elements of both innate and acquired immunity. A commercial form (Actimmune) is approved for the treatment of chronic granulomatous disease, an illness characterized by increased susceptibility to bacterial and fungal infections. A pilot study conducted at Boston's Dana Farber Cancer Institute in the late eighties reported very limited improvements in people with AIDS and Kaposi's sarcoma (KS), but subsequent trials conducted by the AIDS Clinical Trials Group (ACTG) failed to show significant benefit. The toxicities reported in these studies have become the signature of many cytokine therapies: fever, chills, headaches and muscle aches.

When test tube studies suggested that the combination of gamma interferon and another cytokine, TNF-alpha, might inhibit HIV replication, a trial was quickly put together by researchers at San Francisco General Hospital. TNF-alpha is a cytokine released by macrophages and T-cells that is involved in promoting fever and inflammation. The small pilot study of eleven people, published in late 1989, reported inconsistent changes in CD4 cell counts, and the approach was not pursued further. It later became appreciated that TNF-alpha levels are in fact abnormally high in HIV infection, eliminating the rationale for trying this cytokine as an IBT.

Two other members of the interferon cytokine family have been studied in HIV. Beta interferon is naturally made by cells called fibroblasts in response to viral infection, and a synthetic version made it as far as phase II studies for the treatment of KS in the late 1980s. The results, presented in the Annals of Internal Medicine in 1990 by Steven Miles and colleagues from the UCLA (University of California at Los Angeles) AIDS Center, showed evidence of disease regression or stabilization in about half of the 39 study participants. Follow-up studies of beta interferon in combination with AZT were less successful, and this IBT seems now to have quietly dropped off the map. Two commercial versions (Betaseron and Avonex) are approved for the treatment of multiple sclerosis.

Alpha interferon is a cytokine that can be defined both as an IBT and as an antiviral. Of all the interferons, alpha is the most extensively studied. Normally produced by virus-infected white blood cells, this cytokine can both directly block viral replication and shift the balance of the immune response toward a more Th1 (virus-infected cell killing) profile (see article on immune system). Early trials conducted at the National Institutes of Health (NIH) showed that alpha interferon could reduce KS lesions in people with relatively healthy CD4 cell counts. As a result, alpha interferon was approved in 1988 for the treatment of KS in people with over 200 CD4 cells. The side effects, however, quickly became notorious. Similar to many other cytokines, alpha interferon causes flu-like symptoms that can be accompanied by bone marrow suppression and even mental disturbances such as severe depression. Despite years of study both as a single therapy and combined with antiretroviral drugs, alpha interferon has yet to prove successful as an HIV treatment. Instead, this cytokine has found use as a therapy for some cancers and chronic hepatitis B and C infections.

Interleukins are another family of cytokines, the above-mentioned IL-2 being the most famous. But other family members with possible therapeutic potential have been identified and evaluated in clinical trials, including IL-3, IL-4, IL-10, and IL-12. IL-3 can promote the production of multiple cell types (including red cells, granulocytes, macrophages and lymphocytes) by the bone marrow. A Harvard study in people with HIV (published in 1995) showed that IL-3 enhanced white blood cell, neutrophil and eosinophil counts without increasing viral activity. However, other bone marrow stimulants have produced more consistent results (see G-CSF and GM-CSF, below) and IL-3 has not been commercially developed.

IL-4 appears to stimulate B-cells and antibody production but may also influence CD8 T-cell responses and inhibit production of TNF-alpha. A UCLA study of IL-4 as a possible KS treatment, reported in the Annals of Oncology in 1997, proved unsuccessful, although a few participants experienced temporary declines in HIV viral load. The major activity of the cytokine IL-10 is to suppress antiviral T-cell responses, gamma interferon production and antigen presentation. Conversely, IL-10 can also stimulate B-cell responses. A small pilot study at the NIH suggested IL-10 might reduce HIV replication, but a larger trial conducted by Canadian researchers (published in the journal AIDS two years ago) was unable to confirm these findings, and interest in the approach has waned.

At one time, there was considerable enthusiasm about the prospects for IL-12 as an IBT. This cytokine is known to promote the development of antiviral Th1-type CD4 cell responses and enhance the activity of natural killer cells. Test tube studies have found that IL-12 can sometimes improve defective CD4 cell responses to HIV and other antigens. Research plans were dealt a severe blow in 1995 when a phase II study of IL-12 for kidney cancer was halted due to severe toxicity. One participant died and ten others were hospitalized. Oddly, the IL-12 had appeared safe in a phase I trial, and a change in dosing and schedule of administration appears to have accounted for the sudden change in side effect profile. Eventually, the green light was given by the Food and Drug Administration (FDA) for a phase I dose-escalation study in HIV infection. The results, published in 2000 by Marc Jacobson from the University of California at San Francisco (UCSF), turned out to be inconclusive. No changes in viral load or CD4 cell counts were observed, and severe toxicities emerged at the highest doses. Some increases in gamma interferon and CD8 T-cell levels were seen, and the researchers are continuing to explore the role of low IL-12 doses in modifying the immune response.

G-CSF (granulocyte colony stimulating factor) is a cytokine that stimulates production of white blood cells called granulocytes. The synthetic form is approved for the treatment of AZT-induced neutropenia (low neutrophils) in people with HIV. However, G-CSF does not appear to influence HIV directly and is thus not considered an IBT. A related cytokine, GM-CSF (granulocytemacrophage colony stimulating factor), appears to have broader activity, and there have been several trials aimed at assessing its potential as an IBT. A recent phase III study investigated GM-CSF combined with antiretrovirals in people with less than 100 CD4 cells. No overall difference in the number of opportunistic infections was seen in those that received GM-CSF compared to placebo, but CD4 cell counts were slightly improved. People that entered the study with undetectable viral load were also more likely to stay undetectable if they received GM-CSF. Despite these hints of some positive effect, the manufacturer of GM-CSF (Immunex, which markets the drug under the trade name Leukine) has not sought FDA approval for the treatment of HIV infection.

Therapeutic Vaccines

The idea of enhancing the natural immune response to HIV through vaccination arose soon after the virus was discovered. At that time, the long asymptomatic period between HIV infection and the development of AIDS was thought to be a time of viral inactivity or "latency." Researchers theorized that immunizing with parts of HIV (HIV antigens) might extend the time before disease progression occurred. One of these researchers was legendary polio pioneer Jonas Salk, who emerged from semi-retirement in order to develop a candidate vaccine. Salk's approach involved an almost entire HIV virus, chemically killed and given with a vaccine-booster called an adjuvant. Strategies pursued by other researchers included subunit vaccines, which included only a portion of HIV (such as the gp120 and gp160 proteins from the viruses outer covering or envelope), and vector-based vaccines which used weakened viruses such as vaccinia (the smallpox virus) and canarypox (a bird virus) to deliver HIV proteins into the body.

Early studies of these vaccines provided some evidence that new immune responses against HIV could be triggered, but only in individuals with relatively high CD4 cell counts. A phase I trial of Salk's vaccine was begun in 1987 by UCLA researcher Alexandra Levine and found improvements in delayed-type hypersensitivity (DTH) responses to HIV antigens in 12 of 25 individuals studied. It was also claimed by Salk that these "responders" experienced fewer opportunistic infections during follow-up. But this interpretation was complicated by the association between developing DTH and having higher CD4 cells -- the individuals who responded to the vaccine may have been at lower risk for disease progression simply due to their healthier CD4 cell counts. A similarly positive spin was put on early results from studies of a gp160 protein vaccine (made by a company called MicroGenesys), in this case by army researcher Robert Redfield. At the 1990 International AIDS Conference in Amsterdam, Redfield presented data suggesting CD4 cell counts in vaccine recipients were preserved compared to a progressive decline in study participants assigned to placebo.

While these hints of promise raised hopes, the next few years saw the entire therapeutic vaccine effort unravel in a firestorm of controversy. Redfield's data were challenged by other researchers, including independent statistics experts who analyzed his results but found they did not support the claims made in Amsterdam. As if this were not troubling enough, MicroGenesys somehow managed to secure a specific $20 million congressional appropriation for an army phase III trial of their now-suspect gp160 product, to be led by Redfield. The AIDS community erupted and eventually managed to ensure that the money was diverted to the army's overall AIDS research effort.

Meanwhile, Salk's vaccine was heading for trouble of its own. Results from a phase II/III trial involving 103 participants were slated to be presented at the 1993 International AIDS Conference in Berlin, provoking immense hype and hoopla in the media. Rather than present the data during a conference session, Salk's company -- The Immune Response Corporation -- foolishly decided to hold a press event instead. Conference participants without press credentials were left loitering in the corridors, while Levine presented to an audience of journalists and Wall Street analysts. The results of the trial, once known, only compounded the bad blood created by Salk's apparent grandstanding. After a year of follow-up, there were no significant differences in CD4 cell counts between the vaccine and placebo groups. Some improvements in DTH responses were reported, and using a precursor to the viral load test, Levine claimed a slower increase in HIV DNA (as opposed to the RNA measured by current tests) levels in vaccine recipients compared to those receiving placebo. The general audience reaction, however, was disappointment tainted with anger.

Although surrounded by less controversy, the other therapeutic vaccine candidates all met similar fates. The use of vaccinia as a vaccine vector was abruptly halted when an individual with AIDS developed disseminated vaccinia infection and died. Candidates using gp120, p24 and p17 components of HIV all showed some ability to trigger new immune responses in people with high CD4 cell counts, but these responses were not associated with any clear health benefit.

These failures and controversies cast a pall over therapeutic vaccine research for many years afterward. Only with the advent of HAART has the idea of boosting HIV-specific immune responses once again found favor. Even Salk's vaccine (now known as Remune) experienced something of a revival in 1998, when Fred Valentine of New York University showed it could induce new HIV-specific lymphoproliferative responses (see "Measuring Memory T-Cell Responses") in people on HAART. This vaccine has been haunted by its history, however, and trials were recently halted when the latest financial backer -- Pfizer -- decided it no longer wanted to support Remune's development. The hardiest survivor has been the canarypox vector produced by Aventis-Pasteur, which is currently in therapeutic trials at the ACTG.

Immune Suppressants

At the opposite end of the spectrum from stimulating immuneresponses, researchers have also studied drugs that might suppress the immune system in people with HIV. The rationale is based on HIV's preference for replicating in activated, dividing CD4 cells. Certain drugs can inhibit T-cell activation and thus may be able to reduce HIV viral load by indirect means. Among the drugs that have been studied are cyclosporine (CsA), prednisone, cyclophosphamide and methotrexate. These drugs are normally used to reduce potentially harmful immune activation, which occurs in autoimmune disease (when the immune system attacks body tissue) and during transplant rejection (when the immune system attacks transplanted tissue).

CsA was first tried in 1985, when French researchers reported temporary CD4 cell count increases in six people with AIDS treated in a week-long pilot study. A group of Canadian researchers followed up on this lead, treating eight people with CsA for about two months. Results were less encouraging, and significant toxicities including pain, fatigue, loss of appetite, weight loss and progression of KS were reported. A larger study by the original French group then gave CsA to 27 asymptomatic people with HIV for about a year, and claimed to show stabilization of CD4 cell counts. However, seven participants withdrew from the study due to declining T-cell levels, and most of the remaining individuals eventually had to stop taking CsA due to severe side effects. Although these initial results were inconclusive, studies of low-dose CsA are continuing even today. ACTG 334 is evaluating the drug in people with CD4 cell counts above 500, while Swiss researchers led by Guiseppe Pantaleo are combining CsA with HAART for the treatment of acute (very recent) HIV infection.

The immune suppressant prednisone has also received attention as a possible IBT. In 1992, the same French research team that studied CsA tried prednisone in 44 people with CD4 cell counts above 200. The dose was 0.5 mg/kg of body weight for six months, followed by 0.3 mg/kg for the remainder of the year-long trial. The result was an average CD4 cell count increase of 119 cells, but the lack of a control group caused the data to be greeted with caution. A subsequent NIH study using a higher dose was stopped before results could be analyzed and, as Mark Milano reports in this issue, interest in prednisone has faded despite lingering questions about its potential.

Two other immune suppressants, methotrexate and cyclophosphamide, underwent trials as HIV treatments in the mid-nineties. Unfortunately, neither approach produced results that were deemed worthy of publication in the scientific literature.

Thymus-Derived Therapies

Several thymus-derived substances were the subject of IBT trials from the late eighties to the mid-nineties. These included thymomodulin, thymosin alpha-1, THF (thymic humoral factor), TP-5 (thymopentin), thym-uvocal, and thymostimulin. Some were "natural" thymic extracts (taken from calves, for example), while others were made synthetically. Despite early hints that some of these products might improve CD4 cell counts, research in this area has not persisted into the HAART era. The largest trial of any of these approaches was a 352-person, placebo-controlled study of TP-5 combined with AZT. The results (eventually published in 1995) showed no clear benefit from TP-5 treatment and, like so many IBTs before them, thymus-derived therapies subsequently disappeared from the scene.

Cell-Transfer Therapies

A number of research groups have explored the potential for manipulating T-cells in the laboratory and then re-infusing them into people with HIV. The leading approach involves expanding HIV-specific CD8 T-cells in the lab and then administering them as an IBT. At least three separate studies have been completed and published since 1997. The results were fairly consistent, with some short-term reductions in HIV replication noted without obvious toxicity. One study, by Stan Riddell and Phil Greenberg from the University of Washington in Seattle, found that the infused CD8 T-cells localized at sites of HIV replication in the lymph nodes but appeared unable to sustain a prolonged antiviral effect. Advances in the understanding of immunology have now led to an appreciation that HIV-specific CD4 T-cells are required to maintain the activity of HIV-specific CD8 T-cells, perhaps explaining the limited success of these early experiments. Studies are now underway using infusions of both T-cell types, a method that has already shown success in the treatment of active CMV (cytomegalovirus) infection.

Variations on the cell transfer theme have also undergone evaluation. In one Australian trial, CD4 cells taken from individuals with early HIV infection were frozen, preserved and then re-infused later in the course of disease. Results from a 12-person pilot trial were published in 1998, demonstrating CD4 cell count increases in seven participants. Other studies have tried transferring cells between HIV- and HIV+ twins, genetically modifying transfused cells to render them resistant to HIV, or expanding the most functional-seeming T-cells from the lymph nodes of people with HIV and then re-infusing them into the same individual. Results from all of these differing approaches have been mixed, and the overall expense and impracticality of cell transfer therapies make their usefulness questionable, particularly now that HAART is available.

Antibody-Based IBTs

HIV has proven particularly capable of avoiding antibody responses, rendering it resistant to traditional antibody-based vaccine approaches. Similar problems have confronted antibody-based therapeutics. Passive immunotherapy utilized antibodies taken from individuals with early, asymptomatic HIV infection as a treatment for individuals for later stage disease and AIDS. Heavily promoted by Dr. Abraham Karpas from Oxford in the United Kingdom, passive immunotherapy was tried in several trials from the late eighties onwards. Results indicated that there might be some benefit in terms of reduced incidence of opportunistic infections, but a group of French researchers reported that disease could progress rapidly once the antibody infusions were stopped. Oddly, many of the studies also noted an apparent stabilization of CD4 cell counts in the individuals donating the antibodies, suggesting that they may also have experienced some benefit. This observation has never been followed up on. The practical difficulties in preparing and standardizing passive immunotherapy treatment, along with the arrival of HAART, have limited its potential for further development.

Another antibody-based treatment is actually approved for the prevention of bacterial infections in children with HIV. This preparation is called intravenous immune globulin (IVIG, trade name Gamimune) and contains a purified, concentrated mixture of antibodies. A trial in 394 children with HIV (under 13 years old) found that IVIG reduced the incidence of serious bacterial infections by 41%, leading the FDA to approve the drug for this indication in December 1993. Results from a small study in 18 adults with less than 100 CD4 cells, presented in 1997, suggested that IVIG might also reduce the rate of infections in older individuals, but the drug has never been approved for this use.

Hormonal and Herbal IBTs

The definition of an IBT can be blurry, and many different experimental therapies, including some considered "alternative," might fall into this category. DHEA is a synthetic version of a natural testosterone-related hormone that has been proposed to improve the health of the immune system. Spurred by the observation that DHEA levels progressively decline over the course of infection, dose-escalating supplementation trials were initiated in people with HIV in the late eighties. Doses up to 2,250 mg daily appeared safe, but no effect on CD4 cell count was noted. Later studies produced similarly inconclusive results, lessening interest in DHEA as an IBT (although it is still a popular supplement). A company called Hollis Eden is continuing to study a DHEA-derivative (HE2000) as a possible therapy, based on its theoretical ability to improve the functionality of impaired T-cell responses.

Various Chinese herbal compounds have also been described as immune boosters, but have never been thoroughly studied. In many cases, these compounds contain immune stimulants known as polysaccharides which might theoretically promote HIV replication (by causing immune activation) rather than help block it. The herb echinacea is one such example, since it appears to increase levels of the inflammation-associated cytokine TNF-alpha. While most doctors feel that short-term use of herbs like echinacea (e.g. to treat a cold) is unlikely to be harmful, these types of broad immune stimulants are no longer being pursued as IBTs for HIV infection.

Will There Ever Be an IBT for HIV?

Despite this litany of failure and uncertainty, the success of HAART has refocused attention on IBTs as the "final frontier" for HIV research. As Tracy Swan reports, IL-2 is in its final phase of testing and might conceivably be approved within the next few years. New cytokines and chemokines are constantly being discovered and investigated for their potential as IBTs, including IL-15, IL-16 and IL-18. Several major pharmaceutical companies have recently announced active therapeutic vaccine programs for HIV, including Merck and GlaxoSmithKline (see Jeff Gustavson's article). Unlike historical efforts, the latest IBT research is being informed by rapidly accumulating breakthroughs in the scientific understanding of the human immune system. Many researchers are confident the era of shots in the dark is over, and it's only a matter of time before an IBT finally hits the target.

Richard Jefferys is Basic Science Project Director with Treatment Action Group.