Boosting Immunity to HIV--Can the Virus Help?

See allHide authors and affiliations

Science  03 Nov 2000:
Vol. 290, Issue 5493, pp. 946-949
DOI: 10.1126/science.290.5493.946

The human immunodeficiency virus (HIV), the cause of AIDS, invades certain types of host immune cells, in particular CD4+ T helper 1 (TH1) lymphocytes that are progressively eliminated as the virus replicates. These CD4+ T cells, along with CD8+ cytotoxic T lymphocytes, are essential for mounting a coordinated immune attack against HIV. In the absence of antiretroviral drug therapy, the intensity and diversity of CD4+ and CD8+ T cell responses against HIV increase as the amount of virus in the blood (viral load) decreases during primary HIV infection (1, 2). Treating newly infected HIV patients with antiretroviral triple drug therapy (HAART) provides a rapid reduction in the initial burst of virus replication and helps to preserve the HIV-specific CD4+ TH1 cell population from rapid elimination (2); unfortunately, this strategy also lessens the intensity of the HIV-specific CD8+ cytotoxic T cell response (3). This calls into question the actual benefit of early therapeutic intervention, which does not completely eradicate the virus and leaves the patient with an incomplete antiviral immune response. In an effort to solve this dilemma, researchers have interrupted drug treatment during primary (acute) HIV infection. Although the benefits of such therapeutic manipulations remain controversial, anecdotal evidence from patients who have had drug therapy interrupted suggests that periodic and transient increases in the viral load restimulate the waning HIV-specific CD8+ cytotoxic T cell response (4, 5). Now, in a recent issue of Nature, Walker and co-workers (6) report that administering HAART to patients within 72 hours of diagnosing HIV infection, and then discontinuing treatment after 1 to 2 years, helps to establish durable immune control of the virus.

Walker and colleagues treated eight acutely infected HIV patients with HAART, and then, after complete suppression of the viral load for at least a year, interrupted drug therapy. Before treatment was stopped, the CD4+ TH1 (but not CD8+ cytotoxic T cell) response to HIV could be detected in all eight patients. After discontinuing treatment, the virus rebounded and boosted the numbers of virus-specific cytotoxic CD8+ lymphocytes in all patients (see the figure). In three cases, the amount of virus rapidly reached a low steady state—below 10,000 copies of viral RNA per milliliter (ml) of plasma, the level at which drug therapy would have to be reinstated. In the other five cases, the intensity of the initial virus rebound led to a transient decrease in the TH1 cell response and drug therapy was immediately reintroduced. The subsequent reduction of HIV in these patients allowed the TH1 cells to return to baseline levels. A second interruption in drug treatment then caused a weaker rebound of virus to a viral load below 5000 RNA copies/ml of plasma, and did not result in a decrease in the TH1 response (see the figure). In addition, it boosted and broadened the HIV-specific CD8+ cytotoxic T cell response, eventually leading to a low, steady-state viral load. Altogether, five out of eight patients remained off therapy for 6.5 months (median), with their viral loads remaining below 500 RNA copies/ml of plasma. This encouraging finding contrasts with the viral load in untreated HIV patients in the Multicenter AIDS Cohort Study (7), whose viral loads remained consistently above 5000 copies/ml.

Therapy interrupted.

Immune responses to HIV after interrupted drug therapy in patients with primary or chronic HIV infection. Rapid introduction of antiretroviral drug therapy (yellow areas) during the earliest phase of primary infection reduces the initial virus burst (pink line) and allows numbers of HIV-specific TH1 cells (blue dots) to be maintained, but does not allow a strong HIV-specific CD8+ cytotoxic T cell response (turquoise squares). Discontinuation of drug treatment allows the virus to rebound to levels just below the virus peak. (A) The plasma viral load rapidly reaches a steady state below 5000 RNA copies/ml (dashed line) after a single interruption to drug treatment. (B) If the intensity of the initial virus rebound is too great, there is a transient decrease in TH1 cells and drug therapy must be reintroduced. The subsequent reduction in virus restores TH1 cells to baseline levels. A second interruption in drug therapy results in a weaker viral rebound (with the viral load remaining below 5000 RNA copies/ml), no deleterious effects on TH1 cells, and stimulation of HIV-specific cytotoxic CD8+ T cells. Eventually, the virus is maintained at low steady-state levels. (C) In chronically infected HIV patients, treatment interruption is rapidly followed by a strong rebound in virus, which only weakly and transiently stimulates both TH1 and CD8+ cytotoxic T cells, even when repeated (11). (D) Boosting immunity to HIV with vaccines (rather than with the virus itself) before interrupting drug treatment (arrows) may allow the virus to be maintained in a low steady state both in primary and in chronic HIV infection.

What can we learn from the observations of Walker and colleagues? First, gentle or progressive exposure of the immune system to HIV, instead of abrupt and intense stimulation, might better facilitate the generation of protective immune responses. It is already known that a short exposure to the virus is sufficient for HIV-specific TH1 cells to be generated and maintained for long periods (2). The new findings confirm that virus-specific TH1 cells—in the absence of a stronger or more prolonged exposure to HIV—are not capable of stimulating an intense, broad, and durable CD8+ cytotoxic response. Indeed, lessons from animal models infected with lymphocytic choriomeningitis virus show that the magnitude of the effector CD8+ T cell response to an acute viral infection is proportional to the initial burst of virus replication, and determines the intensity of the memory T cell response, which in turn controls the virus at steady-state levels (8). Therefore, a tight, dynamic equilibrium between the virus, CD4+ TH1 cells, and CD8+ cytotoxic T cells must be reached to establish efficient immune control. Walker and co-workers demonstrate that a low, quasi-steady state viral load can be achieved when HAART is discontinued during primary HIV infection. In most of their patients, however, the initial immune responses were too weak to ensure immediate protection after discontinuation of drug therapy. Thus, the immune response had to be boosted by the rising viral load that followed discontinuation of drug therapy.

Despite the encouraging findings of the Walker study, several important questions remain unanswered. Which factors of the host immune system govern the kinetics of establishing protective immunity, and how durable will this protection be? How many treatment interruptions will acutely infected patients require to achieve a steady-state viral load, and will these treatment manipulations be manageable in real life? Although there are no definitive answers to these questions as yet, and the number of patients in the Walker study is small, the results suggest that appropriate therapeutic intervention in patients during the earliest days of HIV infection should help to induce some immune control of HIV.

One intriguing question is whether the apparent benefits of interrupted therapy can be extrapolated to any stage of HIV infection—anecdotal data suggest that this is not the case. It is already known that patients with chronic HIV infection relapse 2 to 3 weeks after a single interruption to drug therapy and require the immediate resumption of treatment (see the figure) (9, 10). In these patients, the HIV-specific CD8+ T cell response can be boosted by interrupting drug therapy, but the immune response is limited compared with the broad repertoire of CD8+ cytotoxic T cells observed in the Walker study patients with primary HIV infection (11, 12). This lack of cytotoxic T cell diversity might reflect the existence of an immune footprint or “antigenic sin” as proposed for B cell immunity—a footprint of immune responses is established during exposure to a virus and is automatically triggered against the same antigens when the immune system encounters the virus again. Alternatively, the lack of cytotoxic T cell diversification in chronic HIV infection might also reflect the crippled HIV-specific TH1 response, which can barely be detected in these patients (2). Although they can be restimulated when exposed again to the virus, these weak HIV-specific CD4+ TH1 cells cannot provide efficient help to CD8+ cells. They are thus rapidly eliminated in the face of vigorous and poorly controlled virus production after each treatment interruption (11). So far, there is no evidence for virus control in chronic HIV infection after structured treatment interruptions. Therefore, any attempts to apply the observations of Walker and co-workers to chronically infected HIV patients should be discouraged until there is conclusive evidence of efficacy.

The encouraging results obtained in patients with a primary HIV infection support the rationale for combining interrupted drug therapy with immunological interventions such as immunization. Indeed, the heavy cost and toxicity of the current drug regimens frequently lead to unsupervised cessation of treatment. To quickly obtain a low, steady-state viral load when treatment is withdrawn, we should ideally restore strong and diverse TH1 and cytotoxic T cell responses against HIV before, and not after, the virus rebounds. In addition, we should not be content with a deleterious immunogen such as HIV itself; instead, nonpathogenic antigenic formulations (such as those in candidate vaccines) should be administered to patients even though their immunogenicity might still be limited. Several clinical trials, involving about 200 acutely infected and 500 chronically infected patients, are currently combining antiretroviral drug regimens with immunization using various HIV immunogens. The response of these patients to cessation of drug treatment after the completion of therapeutic immunization may provide us with the best large-scale evidence yet that, when given a fighting chance, the immune system can indeed control HIV.


View Abstract

Stay Connected to Science

Navigate This Article