Print this page    •   Back to Web version of article

Immunotherapy of HIV: On the Horizon?

By Anne S. De Groot, M.D.*, Brown Medical School, Editor, HEPP News
(The assistance of Jesse Creel** with this article is gratefully acknowledged)

October 2001

Why should the correctional HIV provider care about T cells, T-cell epitopes, and HLA molecules? Because the use of Strategic Treatment Interruption (STI) and therapeutic vaccines for the treatment of HIV -- particularly acute HIV infection -- may be on the horizon for our patients. This article will describe the components of immune response to HIV and explain why STI is more likely to provide benefit in acute HIV infection and less likely to benefit patients who have chronic HIV infection, at least until more effective forms of STI, such as STI given in conjunction with effective therapeutic HIV vaccines, are developed. It should be noted that STI is not an FDA approved use of any of the HAART drugs, and that STI should never be attempted outside of well-established experimental protocols run by experts in the field.

HIV Immunology 101

The cellular immune response is the first arm of the immune system to respond to HIV infection (see "HEPPigram"). Cellular immune response is provided by T helper cells (also known as CD4+ cells or T4 cells) and cytotoxic T cells (also known as CTL, or CD8+ or T8 cells). Studies of acutely infected HIV patients have shown that each individual mounts a cellular immune response to HIV that determines the individual's "set point," the point of stabilization of viral load after acute infection has been controlled. This set point is closely linked to progression to AIDS.1 Patients whose immune systems do not successfully control HIV during acute infection have higher set points and more rapid progression to AIDS than patients whose immune systems do not perform as well during acute infection.

The ability to control HIV and obtain a lower set point is determined by several factors (illustrated in Figure 1). These factors include the initial dose of HIV during infection, the type of virus transmitted (i.e., the specific viral DNA and protein sequences), the HLA of the infected individual, and the number of T-cell epitopes that can be recognized by the individual's T cells.

Figure 1. Factors Determining "Set Point" and Viral Load
in the Setting of Acute HIV Infection

Factors Determining 'Set Point' and Viral Load in the Setting of Acute HIV Infection

Dose of HIV

Numerous studies have clearly demonstrated that the initial dose of infecting virus can determine outcome. For example, patients who received blood donations containing large amounts of HIV (such as hemophiliacs, prior to screening of factor VIII) appear to progress more rapidly than patients who are infected via sex or through minor intravenous exposures.2 Mother-to-child transmission is also dependent on viral load (particularly at the time of delivery). Individuals who are infected with smaller doses may have a lower viral burden at the outset, and perhaps more importantly, may be infected with fewer quasispecies or viral variants. Just as it is easier to control HIV with HAART when starting from a lower total viral load, it may be easier to obtain immune control of HIV when there are fewer viral variants.

HLA and Inheritance

The HLA molecule carries pieces of HIV viral proteins in its central "binding groove" to the surface of infected cells. When T cells see a piece of an HIV protein (a short piece called a peptide) on the surface of the cell, carried by the HLA protein, they react, either by killing the infected cell (the function of CTLs) or by secreting messages that amplify the immune response to HIV (T helper cells). HIV peptides that bind to HLA and turn on T cells are called epitopes. These epitopes tell the immune system that HIV is present, and it needs to begin to fight. So HIV epitopes are the "key" that turns on the immune defense against HIV.

There are many different types of HLA molecules (different alleles of the following HLA genes: HLA A, B, C, DR, DQ, DP). Different HLA molecules present different epitopes; the epitopes they are able to present are determined by the pattern of the amino acids in the HIV peptide. (If the HLA is a lock, only certain epitopes with a particular pattern fit the lock.) Since HLA are genetically encoded, some individuals may be able to use epitopes (that have the right pattern for their HLA) that others cannot (because the sequences don't match the pattern for their HLA).

New Information on the Role of HLA

One of the key findings in HIV over the past two years has been that the genetic background (HLA) of an HIV-infected individual can determine how well they are able to control their HIV. The role of HLA has been reported previously34 but the mechanism of interaction between the HLA and the outcome of infection was less clear until recent data was published in the New England Journal. Mary Carrington, one of the lead authors, demonstrated the link between selected HLA molecules (a variant of B35) and HIV virus sequence.5 She showed that the HLA variant in question did not effectively present a T-cell epitope. (To use our metaphor, the key did not fit the lock.) Other HLA were able to present the epitope. This epitope appeared to play a critical role in the immune response to HIV for these patients.

Thus, a person's genetic background (HLA) determines how many epitopes their immune system is able to recognize. The ability to recognize some specific epitopes appears to be linked to control of infection. One of the HLA types that appears to be associated with more rapid progression so far is a variant of HLAB35.6 HLA Types that are associated with better outcomes include HLA B27 (a rare haplotype) and HLA B57.7 It may not be long before we begin HLA typing our HIV patients so as to predict the course of their infection.

T Cells and Epitopes

T-cell response to HIV epitopes has been shown to be a critical determinant of outcome: patients who have more T-cell responses to many different pieces of HIV (epitopes) appear to be more likely to contain HIV infection.89 Broad T-cell response to many HIV epitopes appears a major determinant of the HIV set point.8 Broad T-cell response is also seen in long-term non-progressors.10 Most of the epitopes that have been mapped are cytotoxic T-cell (CTL) epitopes. T helper cells (Th) also play an important role in early immune response to HIV, but since they are destroyed in chronic HIV infection, researchers are only beginning to understand their role in HIV. They appear to play an important role in early infection. Th cells that respond to HIV are destroyed or inactivated in the setting of chronic infection since it is these Th or CD4+ T cells that are the target of the virus. Rebuilding the supply of Th and CTL that respond to HIV epitopes is one of the goals of STI and therapeutic vaccination.

Virus Sequence and Quasispecies

Different HIVs have different DNA sequences, due to the error-prone RT enzyme whose function is to enable the virus to replicate in the host cell. The RT makes mistakes, and as a result, variants of the original HIV are created. Some of these variants are not viable, however, others are. In acute infection, the variability of the strain is minimal. In chronic infection, the variability of HIV infecting the individual host expands. The swarm of variants that develop in the course of HIV infection are called quasispecies. The number of quasispecies expands as infection progresses.11

The development of quasispecies appears to have an adverse impact on the ability of the immune system to contain the virus. In early infection, response to a few epitopes may effectively contain the virus. In late HIV infection, response to many different epitopes may still fail to contain infection, because of the number of variant viruses. Furthermore, if the patient has HLA that cannot present epitopes contained in the virus sequence, there will be only the minimum CTL response. Alternatively, as the HIV evolves into quasispecies, some isolates evolve in which the original epitopes recognized by the host immune system have been altered and no longer bind to the host HLA (these are called escape mutants).

The Theory Behind STI: Boost and Broaden T-Cell Responses

The goal of STI is to "broaden immune response." From the above discussions, it is clear that the type of immune response that needs to be broadened is the T-cell response. This can be achieved by vaccination, however, there only a few HIV vaccines under study in clinical trials (see "HIV 101"). In the absence of effective vaccines, proponents of STI have argued that allowing the patient to develop a response to his or her own virus, by "autologous vaccination" is a reasonable approach. Opponents of STI note that autologous vaccination involves allowing live, virulent autologous virus that has already been poorly controlled in the absence of drug treatment to replicate and potentially seed more sites in the body.

In general, patients who are undergoing STI are given HAART until viral replication has been completely suppressed below detectable levels (<50 copies per ml), for long durations of time (eight months in some cases). Then the HAART is stopped, and the patient's virus is allowed to rebound up to a certain level (up to 50,000 copies on one occasion or above 5,000 copies on two occasions, for example). This cycle of HAART treatment followed by treatment interruption is continued until there is sustained suppression of the viral load. Following several cycles of STI, in most cases, the viral rebound is smaller.

One potential benefit of STI that has been observed is that rebound virus tends to be "wild type," a strain lacking any viral mutations that had been developed in the course of HAART. However, when HAART is resumed in the course of STI, some researchers have observed that mutant drug resistant strains re-emerge (since they have only been archived in resting T cells). The same may be true for CTL escape mutants, however, data is lacking.

According to STI theory, subjects who are undergoing HAART treatment are able to recover immune function during the period of viral suppression. Stopping HAART allows autologous virus to rebound, and expands the number of T cells that respond to the virus. If the viral load reaches a dangerous level, HAART is restarted and the viral load returns to baseline (below 50 copies in most cases). Opponents of STI have pointed out that HAART may allow CD4 T-cell counts to increase but few, if any, of these T cells are programmed to fight HIV.12 Suppression of viral rebound may be due to broader CTL response, but it may also be due to assistance by T helper cells, particularly in early HIV infection.13

The role of CD8+ T cells (CTL) in containing HIV infection was recently confirmed in an animal model.14 In this model, acutely infected rhesus macaques who were treated with tenofovir (the new Gilead drug) and demonstrated subsequent suppression of a highly virulent strain of the SIV (simian HIV) also showed substantial resistance to subsequent intravenous rechallenge with homologous and highly heterologous SIV isolates, up to more than one year later, despite the absence of effective (neutralizing) antibody directed against the virus. However, when CD8 T cells were depleted by the researchers, the amount of viremia rose by as much as 100,000 fold. Viremia returned to low levels as soon as CD8+ cells were restored. This study showed a positive effect of STI in early infection and confirmed the critical role of CD8 T cells (CTL) in controlling infection after STI.

STI and Acute Infection

The most successful studies of STI have focused on the treatment of acutely infected HIV subjects. Bruce Walker and colleagues (Boston, MA) have reported on 14 subjects who underwent repeated cycles of STI.15 Of the first 14 subjects, seven were able to achieve persistent control of viremia. Four achieved control after a single treatment cessation, two after a second STI, and one after a third STI. Several other studies have been reported where treatment was ceased entirely (treatment cessation, not equivalent to STI), after initial optimal or suboptimal treatment of acute infection. Many of the subjects in these two studies appeared to be able to control viremia (4 out of 10 in one study,16 2 out of 7 in another study17). Whether the number of patients who controlled infection following treatment cessation is any different than the number that would be able to control infection without therapy is unknown at this point.

STI and Chronic Infection

In general, studies of the use of STI in chronically infected patients have not shown any benefit. One reason may be that HAART does not as effectively suppress HIV replication in chronic HIV. In a large study of 128 subjects,18 only 54 subjects were able to completely suppress HIV replication during the cycles of HAART therapy, only 9 of these had viral loads lower than 5,000 copies after completion of the STI cycles. Chronically infected patients also have greater viral diversity, so boosting immune responses to the virus by "autologous immunization" may not have the same effect as this treatment in the setting of acute infection, where viral variability is less and response to the same number of CTL epitopes may be able to contain infection.

Vaccines as Immunotherapy

HIV-1 vaccines in development fall roughly into three categories: recombinant proteins, vectored vaccines, and "replicons" (highly engineered viral vectors).

The vaccine that has been used most often in the setting of immunotherapy is Remune, a vaccine developed by the Salk Institute and now licensed to a subsidiary of Agouron (Immune Response Corp). Remune is an inactivated, protein-depleted, HIV virus. Several cycles of inactivation are carried out so as to assure the safety of the vaccine. Studies of Remune in the treatment of chronic infection have been recently published. Although CD4 T-cell counts were significantly higher, no impact on viral load, or clinical course could be demonstrated.19 The vaccine may have a role when used in conjunction with STI (see below). Thus far, studies of therapeutic vaccination of chronically infected patients with any of the vaccines approved for human use have failed to demonstrate substantial benefits.

Vaccination as an Adjunct to STI

Investigators from the Aaron Diamond AIDS Research Center have explored the potential for therapeutic vaccination in 14 subjects started on HAART within 90 days of HIV infection. This study was reported at the 8th CROI in Chicago, 2001, and reviewed in HEPP News (February 2001). Subjects received 4 doses of a vaccine cocktail that included the canary pox vector vCP1452, expressing HIV gag, pol, env, and nef genes, as well as a recombinant HIV protein, gp160. This was the first study to incorporate a vaccine protocol into STI; while the results weren't altogether exciting, the procedure appeared to be safe.20

A second study was carried out combining STI and Remune vaccination. This study was reported at the AIDS Vaccine 2001 meeting in September of this year.21 The slope of the initial rise in plasma vRNA was significantly slower (0.16 vs. 0.21 log10 copies/day) in the Remune vs. control group (p <0.05). The frequency of cells that produce IFN-gamma when stimulated with several HIV proteins was significantly increased in the Remune group. Therapeutic immunization with Remune appeared to increase both CD4 and CD8 T-cell immunity to HIV antigens and altered the kinetics of viral rebound during the treatment. Viral load and T cells returned to baseline after cessation of HAART, so the long term benefits of this approach are not yet clear.22


Correctional HIV patients frequently experience what may be called unstructured treatment interruptions (USTI) because of numerous barriers to effective care that may exist in prisons today. Our patients more often experience interruptions because they experience side effects or have comorbidities such as liver disease that are exacerbated by HAART. Furthermore, these patients may be mentally ill and stop taking prescriptions or choose to drinking or taking drugs again and this interferes with their adherence. Bad care while incarcerated does occur but is just one of many reasons that patients have USTI. How does USTI compare to STI? Poorly. For example, many of our patients do not achieve suppression of viral replication of long enough duration. Secondly, the duration of HAART cessation may occur over periods of months (sometimes years), much longer than the duration of treatment interruption evaluated in STI studies. Furthermore, our patients frequently resume therapy, when they are reincarcerated, with viral loads vastly exceeding the limits used in STI studies. In addition, our patients usually resume drug use when they interrupt their HAART. Resumption of injection drug use and the associated life circumstances may adversely impact the immune response to "autologous immunization." In short, USTI has none of the features that may contribute to the success of STI in acute HIV infection. Experts in the field caution against the application of STI until more information can be obtained on the optimal duration of the individual cycles, the role of HLA, the role of T-cell epitopes, and the role of adjunctive treatments such as vaccines.

In conclusion, STI is a treatment that is on the horizon, particularly for modulation of acute HIV infection. Much remains to be learned about the optimal use of this strategy. Investigations related to these new treatments have expanded our understanding of immune response to HIV, and therefore may help us understand why some patients do well, and others worse, in the setting of a standardized approach to HIV treatment. Providers practicing in correctional settings probably have much greater experience with unstructured treatment interruption (USTI) than providers practicing in with non-incarcerated populations do. We may be in an excellent position to perform observational studies and teach our non-correctional peers about the outcomes of USTI. The next potential venue for this type of report would be the 9th CROI to be held in Seattle, Washington ( On the topic of treatment interruptions we have much to learn from our academic colleagues, and perhaps, even more to contribute.


T cell: Lymphocyte possessing either helper properties, assisting other cells in the body to respond to infection (a.k.a. T helper cells) or lytic properties (a.k.a. killer T cells, or Cytotoxic T cells/CTL).

MHC: Short for the Major Histocompatability Complex. This is the molecule present on the outside of cells that presents peptides to the immune system. It acts as the carrier for pieces of the HIV virus (or any other virus or bacteria in the cell). Each person has six MHC genes (and may be heterozygous at all six loci). These molecules determine transplant rejection (hence the MHC name) and response to infections.

HLA: Short for Human Leukocyte Antigen. These proteins are genetically determine molecules that sit on the surface of antigen presenting cells. They play a key role in the presentation of immune stimuli to T cells. Essentially, HLA is the name that refers to the human MHC.

T-cell epitope: Only a small piece of a virus, a peptide known as an epitope, will bind in the binding groove of each MHC. Mutations of the peptide can make binding, and hence immune response, impossible to detect (escape mutations). Alternatively, the patient can possess a variant of HLA to which critical HIV epitopes cannot bind. (See NEJM.)

Viral quasi species: The HIV virion's error-prone RT makes mistakes, and as a result, variants of the original infecting isolate of HIV are created. The swarm of variants that develop in the course of HIV infection are called quasi species. The number of quasispecies expands as infection progresses.

Escape mutation: HIV isolates that have evolved in which the original epitopes recognized by the host immune system have been altered or replaced and the new peptides no longer bind to the host HLA or are not recognized by the T cells.

Treatment cessation: Arresting HAART after initiation. (Without further cycles of intermittent therapy.)

Structured Treatment Interruption (STI): Cycles of HAART followed by cycles of no treatment. Viral load is closely monitored and treatment is resumed when viral load rises to a detrimental level.

Unstructured Treatment Interruption (USTI): Not an investigational regimen, however one that is observed for many incarcerated individuals, due to failure to resume therapy (for any number of reasons) after release from prison or jail. Cycles of HAART of inconsistent duration followed by cycles of interruption of variable duration. Viral loads often achieve very high levels before treatment is resumed.

* Consultant & Speaker's Bureau: Agouron Pharmaceuticals, DuPont, Merck, Roche, Boehringer-Ingelheim/Roxane Laboratories.

** Nothing to disclose.


  1. Updated HHS guidelines on the Web at

  2. CDC. MMWR 1998; 47 (RR-17):3.

  3. Saah A.J., Hoover D.R., Weng S., Carrington M., Mellors J., Rinaldo C.R. Jr., Mann D., Apple R., Phair J.P., Detels R., O'Brien S., Enger C., Johnson P., Kaslow R.A. AIDS. 1998 Nov 12;12(16):2107-13.

  4. Gao X., Nelson G.W., Karacki P., Martin M.P., Phair J., Kaslow R., Goedert J.J., Buchbinder S., Hoots K., Vlahov D., O'Brien S.J., Carrington M. N Engl J Med. 2001 May 31;344(22):1668-75.

  5. Musey L., Hughes J., Schacker T., Shea T., Corey L., McElrath M.J. N Engl J Med. 1997 Oct 30;337(18):1267-74.

  6. Carrington M., Nelson G.W., Martin M.P., Kissner T., Vlahov D., Goedert J.J., Kaslow R., Buchbinder S., Hoots K., O'Brien S.J. Science. 1999 Mar 12;283(5408):1748-52.

  7. Bender T.J., Tang J., Rivers C., Muligan M.J., Kaslow R.A. Program and abstracts of the 8th Conference on Retroviruses and Opportunistic Infections, February 4-8, 2001, Chicago Illinois. Abstract 193.

  8. Walker, B.D. On Medscape at

  9. Koup R.A., Safrit J.T., Cao Y., Andrews C.A., McLeod G., Borkowsky W., Farthing C., Ho D.D. J Virol. 1994 Jul;68(7):4650-5.

  10. Shankarappa R., Margolick J.B., Gange S.J., Rodrigo A.G., Upchurch D., Farzadegan H., Gupta P., Rinaldo C.R., Learn G.H., He X., Huang X.L., Mullins J.I. J Virol. 1999 Dec;73(12):10489-502.

  11. Ruiz L., Carcelain G., et al. AIDS-Hagerstown, June 15 2001; 15 (9): F19-F27.

  12. Norris P.J., Rosenberg E.S. AIDS. 2001 Feb;15 Suppl 2:S16-21.

  13. Lifson J.D. J Virol. 2001 Nov; 75 (21): 10187-10199.

  14. Rosenberg E.S., Altfeld M., Poon S.H., et al. Nature. 2000; 407:523-526.

  15. Reported by Bruce Walker in his talk: Structured treatment interruption: novel strategy or oxymoron? State-of-the-art lecture and summary. 8th CROI; February 4-8, 2001. Abstract 266.

  16. Yang O.O., Kalams S.A., Rosenzweig M., et al. J Virol. 1996;70:5799-5806.

  17. Tsomides T.J., Walker B.D., Eisen H.N. Proc Natl Acad Sci USA. 1991;88:11276-11280.

  18. Yang O.O., Walker B.D. Adv Immunol. 1997;66:273-311.

  19. Kahn J.O., Cherng D.W., Mayer K., Murray H., Lagakos S. JAMA. 2000 Nov 1; 284 (17): 2193-2202.

  20. Jin X., Ramanathan M. Jr., Barsoum S., et al. 8th CROI; February 4-8, 2001. Abstract 21.

  21. AIDS Vaccine 2001, Sept 5th-8th, 2001, Philadelphia PA. See review by Nicodemus, Sbai, and De Groot on AIDScience, October 2001,, for more information on the meeting.

  22. Bucy R.P., Moss R., Gersten M., and the 806A Study Team. AIDS Vaccine 2001 Abstract 243.

Back to the HEPP News October 2001 contents page.

This article was provided by Brown Medical School. It is a part of the publication HEPP News. You can find this article online by typing this address into your Web browser:

General Disclaimer: is designed for educational purposes only and is not engaged in rendering medical advice or professional services. The information provided through should not be used for diagnosing or treating a health problem or a disease. It is not a substitute for professional care. If you have or suspect you may have a health problem, consult your health care provider.