Latent infection of CD4+ T cells provides a mechanism for lifelong persistence of HIV-1, even in patients on effective combination therapy
Diana Finzi1, Joel Blankson1, Janet D. Siliciano1, Joseph B. Margolick2, Karen Chadwick2, Theodore Pierson1, Kendall Smith3, Julianna Lisziewicz4, Franco Lori4, Charles Flexner1, Thomas C. Quinn1, 5, Richard E. Chaisson1, Eric Rosenberg6, Bruce Walker6, Stephen Gange7, Joel Gallant1
& Robert F. Siliciano1
1 Department of Medicine, Johns Hopkins University School of Medicine, Baltimore Maryland 21205, USA
2 Department of Molecular Microbiology and Immunology, Johns Hopkins University School of Hygiene and Public Health, Baltimore Maryland 21205, USA
3 Department of Medicine, Cornell University Medical College, New York, New York 10021, USA
4 RIGHT and Georgetown University, Washington DC 20007, USA
5 National Institutes of Allergy and Infectious Diseases, NIH, Bethesda Maryland 20892, USA
6 Partners AIDS Research Center and Infectious Disease Division,
Massachusetts General Hospital and Harvard Medical School, Charleston, Massachusetts 02129, USA
7 Department of Biostatistics, Johns Hopkins University School of Hygiene and Public Health, Baltimore Maryland 21205, USA
Correspondence should be addressed to Robert F. Siliciano
Combination therapy for HIV-1 infection can reduce plasma virus to undetectable levels, indicating that prolonged treatment might eradicate the infection. However, HIV-1 can persist in a latent form in resting CD4+ T cells. We measured the decay rate of this latent reservoir in 34 treated adults whose plasma virus levels were undetectable. The mean half-life of the latent reservoir was very long (43.9 months). If the latent reservoir consists of only 1 105 cells, eradication could take as long as 60 years. Thus, latent infection of resting CD4+ T cells provides a mechanism for lifelong persistence of HIV-1, even in patients on effective anti-retroviral therapy.
Analysis of viral load in HIV-1 infection has contributed to our understanding of AIDS pathogenesis and has facilitated management of the disease1,
2,
3,
4,
5,
6,
7,
8,
9,
10. Virus replication continues throughout the disease, even during the asymptomatic phase between primary infection and the development of AIDS (refs. 2,3). Potent inhibitors of HIV-1 reverse transcriptase and protease produce a rapid exponential decrease in plasma virus levels, reflecting the short half-lives of free virus (t1/2, less than 6 hours) and of the cells that produce most of the plasma virus (t1/2, about 1 day)(refs. 11,12,13). After the first few weeks of therapy, a second, slower phase of viral decay becomes apparent (t1/2, about 14 days). This reflects the slower turnover of a distinct population of chronically infected cells14. After several weeks of combination therapy, plasma virus levels decrease to below the limit of detection in many patients15,
16, and it becomes difficult to culture virus from the blood. Viral burden in lymph node tissue is also correspondingly reduced17. Extrapolation of the second phase of decay to zero residual infected cells led to the suggestion that 2−3 years of fully suppressive therapy might be sufficient to achieve eradication14. However, this projection was made with the caveat that there might be other, more stable compartments in which the virus could persist.
One potential mechanism for viral persistence involves the establishment of a state of latent infection18,
19,
20,
21,
22,
23. HIV-1 replicates well in activated CD4+ T cells18,
24, and latent infection is thought to occur only in resting CD4+ T cells22. A potentially stable latent reservoir may form when productively infected CD4+ lymphoblasts survive the cytopathic effects of HIV-1, evade the immune system, and return to a resting memory state carrying integrated provirus21,
22,
23. In these cells, there is minimal transcription from the HIV-1 long terminal repeat because of the absence of necessary host factors that are present only in activated T cells18. Because the biological function of memory cells is to persist for long periods of time to allow responses to previously encountered antigens, and because the viral DNA in these cells is stably integrated, these latently infected memory cells can potentially serve as a long-term reservoir for HIV-1.
Recent studies have directly demonstrated that HIV-1 establishes a state of latent infection in resting memory CD4+ T cells in vivo22,
23. Using new culture methods, replication-competent HIV-1 can be recovered from these cells, even in patients on combination therapy who have no detectable plasma virus25,
26,
27. The half-life of this latent reservoir, an essential parameter in determining whether current therapy regimens can produce eradication, is unknown. Here we have used direct longitudinal measurements to assess the size and stability of the latent reservoir in patients who have been treated with current 'standard of care' anti-retroviral therapy. Our results challenge the idea that anti-retroviral therapy as it is now given can ever be expected to eradicate the infection.
We evaluated the latent reservoir by longitudinal and cross-sectional studies in 34 HIV-1-infected adults who had responded well to 'standard of care' anti-retroviral therapy28,
29. Patients were selected on the basis of strict patient- and physician-reported adherence to three to five drug regimens that met or exceeded current guidelines28,
29; viral load measurements that decreased to less than the limit of detection of a standard RT−PCR assay (200 copies/ml) within 1−4 months after initiation of therapy; and follow-up viral load measurements that were consistently less than 200 copies/ml. The patients were diverse in age, sex, race or ethnic group, risk factors and CD4 levels at the start of therapy (Table 1). Thus, this population reflects well the diversity of the epidemic. Various combination anti-retroviral regimens were used, all of which had at least one protease inhibitor (except for patient 7, who was on zidovudine (AZT), lamivudine (3TC) and nevirapine (NVP)).
To evaluate the latent reservoir, we isolated resting CD4+ T cells from the peripheral blood using a procedure that gives purities of greater than 99% (refs. 22,23). We evaluated the presence of cells with latent virus among these purified resting cells using a limiting dilution virus culture assay that detects only virus that is fully competent for replication22,
23.
Size of the latent reservoir We determined the frequency of latently infected resting CD4+ T cells with replication-competent HIV-1 as a function of time on combination therapy (Fig. 1, summary plot). Of 79 cultures of resting CD4+ T cells from 34 treated patients in whom plasma virus was less than 200 copies/ml, replication-competent virus was isolated 83% of the time (Fig. 1, filled symbols). This high rate of virus isolation was important because it allowed quantitative analysis of decay rates. Only 2 of these 34 patients (6%) were uniformly negative in repeated assays. Despite the wide dynamic range of the culture assay (more than 6 logs: 10,000−0.01 infectious units per million, IUPM), measurements showed a distinct propensity to cluster in the range of 0.1−1.0 IUPM (Fig. 1). The geometric mean frequency for all positive determinations after 3 months was 0.82 IUPM. These results confirm the persistence of HIV-1 in a small latent reservoir in most patients, including those with plasma virus levels less than 200 copies/ml.
Figure 1. Frequency of latently infected cells in 34 HIV-1-infected adults on combination therapy who had plasma virus levels less than 200 copies/ml.
Frequencies were measured using a culture assay that detects replication-competent HIV-1 persisting in resting CD4+ T cells. Results are expressed as infectious units per million (IUPM) resting CD4+ T cells. Colored lines, repeat measurements in individual patients (numbers in key, patient numbers); filled symbols, successful detection of resting CD4+ T cells with replication-competent HIV-1 at the indicated frequencies; open symbols, unsuccessful virus isolation (for these, the upper bound on the infected cell frequency, estimated based on the number of input cells, is plotted). Initial time points on patients 1−22 have been previously reported25 and are included here for reference and to provide initial values for the measurement of decay rates. Blue bar (right), dynamic range of the assay.
Longitudinal analysis of the decay rate of the latent reservoir To address the essential issue of the decay rate of the latent reservoir, we did longitudinal analysis of the frequencies of latently infected cells in individual patients. Included in the data set (Fig. 1) are longitudinal measurements on 20 patients whose plasma virus levels had decreased to and remained less than 200 copies/ml (Fig. 2). In these patients, the frequencies of latently infected cells were measured two to seven times over the course of 7−21 months (mean follow-up, 14.4 months). Although there was some fluctuation in the levels in individual patients, consistent with the relatively wide 95% confidence intervals for individual determinations, the data demonstrate that this compartment has an extremely slow decay rate, with slopes nearly zero in most cases (Fig. 2).
Figure 2. Decay rates of the latent reservoir in individual patients.
Frequencies in IUPM are plotted on a log10 scale (vertical axes) as a function of time in combination therapy in months (horizontal axes). Black lines, the slope of a least-squares regression line (when more than two determinations were made). Acute, patients started on combination therapy during acute infection. For patients 2 and 4, the initial measurements were negative and only an upper bound on the infected cell frequency could be determined. Subsequent determinations were positive. The slope is therefore a minimal estimate, which may underestimate the actual half-life. Pt #, patient number; m, slope.
To provide a more accurate estimate of half-life, we did statistical analysis using a random-effects regression model30 for decay with first-order kinetics (Table 2). This analysis used all the positive longitudinal and cross-sectional measurements on patients who had plasma virus levels less than 200 copies/ml, taking into account correlation between repeated observations on the same patient. The result was an extremely slow mean decay rate, with a slope of −0.00686 log10 IUPM/month, consistent with a t1/2 of 43.9 months (Table 2). This slope is not statistically different from zero (P = 0.406). The lower 95% confidence bound for the slope (−0.0234) gave a half-life of 12.86 months. The upper confidence bound for the slope was positive; thus, the half-life could not be calculated. Using an estimate of 1 106 cells as the total size of the reservoir23 and the mean decay rate calculated using random effects regression model, we predict that 73 years of therapy would be required for eradication of this reservoir. Even using the lower 95% confidence bound for the slope, 21.4 years on therapy would be required. If the reservoir contains only 1 105 cells, 60.8 years of therapy would be required in the average case, and 17.8 years would be required, if the lower 95% confidence bound is used (Fig. 3). If the analysis is restricted to patients from whom multiple determinations are available, then the estimated decay rate is even slower (Table 2). These results conclusively demonstrate that in a patient population that reflects the diversity of the epidemic, this reservoir does not decay in a clinically relevant time frame with current 'standard of care' therapy.
Figure 3. Mean rate of decay of the latent reservoir.
Elimination of a reservoir of 1 105 cells will require 5 logs of decay, therefore decay data are plotted on an expanded axis (vertical). The mean slope (thick red line) and the 95% confidence intervals about the mean slope (thin red lines) were calculated as described in Table 2, Analysis A. IUPM, infectious units per million.
There was some variation in the slopes in individual patients, ranging from no decay at all (zero or positive slopes) in five patients (9, 11, 12, 17 and 23) to a fairly rapid decay rate in one patient (22, t1/2 = 3.5 months). Although decay rates were measured only in patients who had viral loads that were undetectable by a standard RT−PCR assay (less than 200 copies/ml), a subset of the patients (9, 10 and 21) had plasma RNA levels in the range of 20−200 copies/ml by an 'ultrasensitive' assay on more than one occasion. However, these 'spikes' did not seem to result in a slower decay rate. In two of these three patients, the decay rate was actually faster than the mean decay rate for the remaining patients, who consistently had less than 20 copies/ml.
In 3 of 34 patients (5, 15 and 34), virus isolation was sometimes or always unsuccessful. In the absence of culturable virus in the sample of resting cells (Fig. 1, open symbols), the value plotted represents an upper bound on the infected cell frequency. Virus isolation was most difficult in two patients (5 and 15) who were started on therapy late in the course of disease with very low CD4 nadirs. The only other patient in whom multiple attempts to detect latent HIV-1 were unsuccessful had a low level of latently infected cells before therapy began (data not shown) and is therefore also a 'special case'. Two attempts to detect these cells after the initiation of treatment were unsuccessful. Nevertheless, the persistence of virus in this patient is indicated by the fact that the patient experienced a 'rebound' in plasma virus after becoming noncompliant with therapy.
Establishment of the latent reservoir in primary HIV-1 infection Given the extremely slow decay rate of the latent reservoir in resting CD4+ T cells, we also determined whether early treatment could affect the size and stability of the reservoir. Included in the data set are four patients (17, 20, 23 and 24) who were started on combination therapy during primary HIV-1 infection and were then followed for 16−22 months, during which time they had suppression of viremia to less than 200 copies/ml. Latently infected cells were detected in each (Figs. 1 and 2), despite the fact that in two of these patients, treatment was started even before seroconversion (patients 23 and 24). For patient 24, therapy was started 48 hours after the patient presented with acute retroviral syndrome, but persistent, latently infected cells were nevertheless readily detectable. Longitudinal studies in three patients (17, 23 and 24) showed only minimal evidence for decay (two of three slopes were zero; Fig. 2).
Discussion Recent advances in anti-retroviral therapy have proven to be very effective in reducing viral load in patients with HIV-1-infection, leading to demonstrated reductions in morbidity and mortality31 and providing a basis for the first attempts, at predicting HIV-1 eradication since the beginning of the epidemic14. Despite this progress, our study indicates that latent HIV-1 in resting CD4+ T cells will be a principal obstacle to virus eradication. Through direct longitudinal analysis of the decay rate of the latent reservoir, we have demonstrated that the reservoir is very stable, with a half-life of more than 43 months in the average patient on current 'standard of care' therapy. Even with conservative estimates of the total body number of latently infected cells, an average of at least 60.8 years of treatment will be required to eradicate this compartment in most patients. Our findings are based on the isolation from this reservoir of viruses that are fully replication-competent and are therefore likely to be capable of 'rekindling' the infection in patients who stop therapy. The extremely slow decay rate of this reservoir raises the disturbing prospect that in some patients, the time required for HIV-1 eradication with current combination regimens may be so long that other intervening problems, such as cumulative toxicities of anti-retroviral drugs32, may make eradication
difficult.
In interpreting the data presented here, the strengths and limitations of the culture assay used must be considered. The 95% confidence intervals about individual determinations are +/- 0.7 logs; therefore random fluctuations could either overestimate or underestimate the decay rate on an individual level if only a few observations are used. To overcome this variability, we have collected observations on many patients over a long time and analyzed the data using a random-effects regression model. Based on the number of patients measured here, and the time that they have been followed, we have provided both a point estimate of the half-life (43.9 months) and a 95% confidence interval for the half-life. Although our study does not have sufficient statistical 'power', with these data, to discern small differences in decay rates, such as whether the half-life is closer to 40 months or 50 months, this study is able to exclude very rapid decay. The lower bound of the 95% confidence interval (12−13 months) indicates that our data would not be consistent with a more rapid mean decay rate (for example, less than 12 months) for the patient population studied. The upper bound on the mean slope is positive, consistent with no decay at all.
Because the assay involves actual virus isolation, it directly measures the form of the virus that is of the most concern: replication-competent virus persisting in a latent form in resting CD4+ T cells. Thus, when virus is isolated from a culture of 1 million resting CD4+ T cells, there is no other explanation than that the virus has persisted in latently infected cells present at a frequency of at least 1 latently infected cell per million. Therefore, the results presented here argue in a very definitive way for long-term persistence of the virus.
The decay rates measured here are best thought of as the observed decay rates for this compartment in patients treated with current 'standard of care' therapy, in contrast to the true intrinsic decay rate of the reservoir. The considerable stability of the latent reservoir is consistent with the fact that the reservoir is composed at least in part of memory T cells carrying integrated HIV-1 DNA (23). The biological function of memory T cells is to persist and provide protection against previously encountered microorganisms. The half-life of memory T cells in normal humans has not been studied extensively, but some estimates of an intermitotic half-life in the range of 5−6 months have been reported33. The mean half-life of the latent reservoir is much longer than the reported mean intermitotic half-life of memory CD4+ T cells, indicating that the latent reservoir may be renewed by occasional proliferation of the infected cells or by entry of new cells into the reservoir as a result of a low level of ongoing viral replication34. A subset of our patients had occasional 'spikes' in plasma RNA levels into the range of 20−200 copies/ml. The importance of this very low level of ongoing replication is unclear, and in two of three patients, the decay rate was actually faster than that seen in patients who had consistent suppression of viral replication to less than 20 copies/ml. Additional studies will probably be needed to determine whether suppression to less than 20 copies/ml gives a faster decay than suppression to less than 200 copies/ml. If ongoing viral replication contributes to the stability of the reservoir, then the true intrinsic decay rate may be more rapid than that measured here. In this situation, the development of even more potent anti-retroviral regimens may stop all residual replication and show the true intrinsic decay rate of the latent reservoir, permitting eradication if this rate is sufficiently rapid and if there are not other substantial reservoirs.
If low-level ongoing replication contributes new infected cells to the latent pool, then selection for drug-resistant virus in this pool may eventually become evident. Sequencing of the reverse transcriptase and protease regions of the pol genes of viruses isolated from the latent reservoir has so far shown little evidence for the evolution of drug resistance25,
27. Thus, our results so far are consistent with the idea that the long half-life of the reservoir is mainly due to its intrinsic stability. Nevertheless, it is possible that low-level ongoing replication is occurring and may eventually result in the emergence of drug resistance26,
35 during the extremely long time required for decay of the latent reservoir.
Given the considerable stability of the latent reservoir, there has been interest in whether early treatment can prevent the establishment of the reservoir, limit its size or accelerate its decay. The answer to the first question seems to be no. Latently infected cells are present in patients who have been treated during or shortly after seroconversion, as demonstrated here and in previous reports25,
26. In the acute seroconvertors studied here, the size and decay rate of the latent reservoir were similar to those seen in patients who started therapy at later stages. However, it is possible that the size of the latent reservoir may be reduced by early treatment with new combination therapy regimens that include drugs like hydroxyurea that target the host cells (F.L., J.L, D.F. and R.F.S., unpublished results). In addition, a recent study indicates a more rapid decay rate in a subset of acute seroconvertors who started combination therapy within 90 days of infection (Zhang et al., manuscript submitted).
In the analysis of 34 patients on combination therapy who had plasma HIV-1 RNA levels of less than 200 copies/ml, replication-competent virus was isolated from the latent reservoir in all but two. In one (patient 34), the persistence of a latent reservoir is indicated by the fact that the patient eventually failed therapy because of problems with adherence. These results emphasize that negative results in an assay for latently infected cells cannot be used to demonstrate that eradication has occurred. The lower end of the dynamic range of the assay depends on the number of cells from the patient that can be cultured. For patients with negative culture assays, it is very likely that virus isolation would be possible if more cells could be cultured. The only other patient from whom virus was never isolated despite repeated attempts was patient 5, who was in advanced stages of AIDS with a CD4+ T-cell count of 10 when treatment was begun. It is possible that the few residual infected cells in this patient were simply 'diluted out' by newly generated CD4+ T cells appearing after the initiation of therapy. Nevertheless, it is encouraging that even in very late-stage patients, immune reconstitution can occur without the production of detectable numbers of new latently infected cells.
Although decay of the latent reservoir is a very slow process, it is possible is that total eradication of the latent reservoir may not be necessary. In certain conditions, the immune system may be capable of controlling the small amount of virus released from the latent reservoir. Rare patients do not experience a rebound in plasma virus after stopping therapy even though they continue to harbor replication-competent virus in resting CD4+ T cells (J.L. et al., manuscript submitted). Long-term nonprogressors have persistent polyclonal HIV-1-specific CD4+ T-cell responses that effectively orchestrate antiviral immune responses36. New immunotherapeutic approaches such as HIV-1 vaccination or cytokine therapy might enhance virus-specific immunity sufficiently to allow containment of any virus released from this very stable reservoir or might directly mobilize the reservoir37,
38. However, the results presented here demonstrate that without specific interventions, the persistence of latently infected cells will probably represent a major barrier to virus eradication.
Methods Patient population. The patients included 21 patients from a previously described cohort25 who maintained long-term suppression of viral replication on combination therapy. These patients are numbered 1−22 in keeping with the previous numbering; patient 6 in this group failed therapy and was not considered. Thirteen additional patients selected according to the above criteria (patients 23−34) were also studied. In 17 patients, multiple viral load measurements made with an 'ultrasensitive' assay (sensitivity, 20 copies/ml) were available. These measurements were generally negative, except in patients 9, 10 and 21, who had more than one measurement in the range of 20−200 copies/ml. In most patients, substantial increases in CD4 counts with therapy were noted. All patients gave informed consent for phlebotomy.
Isolation of resting CD4+ T cells. Resting CD4+ T cells were purified from peripheral blood mononuclear cells by bead depletion and flow cytometry as described22,
23.
Quantitation of latently infected cells. Latently infected cells were measured using a limiting dilution culture assay22,
23,
25. The culture assay is based on the standard virus culture assay used by the AIDS Clinical Trials Group and the data are analyzed using their standard method. However, our assay has an important modification: a specific step to activate the resting cells. Resting cells were activated with the mitogen phytohemagglutinin and a more than tenfold excess of irradiated HIV-negative donor peripheral blood mononuclear cells, in conditions that induce activation of CD4+ T cells with high efficiency, allowing rescue of infectious virus from latently infected cells by co-culture22,
23,
25. Growth of virus was detected by measuring p24 antigen in culture supernatants by ELISA. The assay was set up as a duplicate fivefold-dilution series, from as many as 25 106 cells/well to as few as 320 cells/well. Control wells with no patient cells were assessed in each assay and were invariable negative. Sequence analysis of pol and env genes from virus isolates demonstrated that the isolates from each patient were unique and distinct from all reported isolates and laboratory strains25. Infected cell frequencies were determined by the maximum likelihood method39 and were expressed as infectious units per million (IUPM) resting CD4+ T cells. The confidence intervals for individual determinations were +/- 0.7 log IUPM. Random-effects regression models30 were used to estimate the mean decline in log IUPM over follow-up time from initiation of therapy. Half-life estimates were calculated assuming first-order decay kinetics. Only data obtained more than 3.5 months after the initiation of therapy were used in the analysis to avoid detection of a labile pre-integration form of latency observed in patients with high viral loads (refs. 23,40, 41, and J.B. and R.F.S., unpublished data).
Measurement of plasma virus. Standard viral load measurements were made by an RT−PCR assay with a sensitivity of 200 copies/ml. 'Ultrasensitive' measurements of plasma virus were made using a modification of the Roche HIV-1 Monitor test.
Harper, M.E., Marselle, L.M., Gallo, R.C., & Wong-Staal F. Detection of lymphocytes expressing human T-lymphotropic virus type III in lymph nodes and peripheral blood from infected individuals by in situ hybridization. Proc. Natl. Acad. Sci. USA83,772-776 (1986). | PubMed | ChemPort |
Coombs, R.W. et al. Plasma viremia in human immunodeficiency virus infection. N. Engl. J. Med.321, 626-1631 (1989).
Ho, D.D., Moudgil, T. & Alam, M. Quantitation of human immunodeficiency virus type 1 in the blood of infected persons. N. Engl. J. Med.321, 1621-1625 (1989). | PubMed | ISI | ChemPort |
Psallidopoulus, M.C. et al. Integrated proviral human immunodeficiency virus type 1 is present in CD4+ peripheral blood lymphocytes in healthy seropositive individuals. J. Virol.63, 4626-4631 (1989). | PubMed |
Schnittman, S.M. et al. The reservoir for HIV-1 in human peripheral blood is a T cell that maintains expression of CD4. Science245, 305-308 (1989). | PubMed | ISI | ChemPort |
Daar, E.S., Moudgil, T., Meyer, R.D. & Ho, D.D. Transient high levels of viremia in patients with primary human immunodeficiency virus type 1 infection. N. Engl. J. Med.324, 961-964 (1991). | PubMed | ISI | ChemPort |
Pantaleo, G. et al. Lymphoid organs function as major reservoirs for human immunodeficiency virus. Proc. Natl. Acad. Sci. USA88, 9838-9842 (1991). | PubMed | ChemPort |
Pantaleo, G. et al. HIV infection is active and progressive in lymphoid tissue during the clinically latent stage of disease. Nature362, 355-358 (1993). | Article | PubMed | ISI | ChemPort |
Piatak, M.Jr. et al. High levels of HIV-1 in plasma during all stages of infection determined by competitive PCR. Science259, 1749-1754 (1993). | PubMed | ISI | ChemPort |
Mellors, J.W., Rinaldo, Jr., C.W., Gupta, P., White, R.M., Todd, J.A. & Kingsley, L.A. Prognosis in HIV-1 infection predicted by the quantity of virus in plasma. Science272, 1167-1170 (1996). | PubMed | ISI | ChemPort |
Wei, X. et al. Viral dynamics in human immunodeficiency virus type 1 infection. Nature373, 117-122 (1995). | Article | PubMed | ISI | ChemPort |
Ho, D.D., Neumann, A.U., Perelson, A.S., Chen, W., Leonard, J.M. & Markowitz, M. Rapid turnover of plasma virions and CD4 lymphocytes in HIV-1 infection. Nature373, 123-126 (1995). | Article | PubMed | ISI | ChemPort |
Perelson, A.S. et al. Decay characteristics of HIV-1-infected compartments during combination therapy. Nature387, 188-191 (1997). | Article | PubMed | ISI | ChemPort |
Gulick, R.M. et al. Treatment with indinavir, zidovudine, and lamivudine in adults with human immunodeficiency virus infection and prior antiretroviral therapy. N. Engl. J. Med.337, 734-739 (1997). | Article | PubMed | ISI | ChemPort |
Hammer, S.M. et al. A controlled trial of two nucleoside analogues plus indinavir in persons with human immunodeficiency virus infection and CD4 cell counts of 200 per cubic millimeter or less. AIDS Clinical Trials Group 320 Study Team. N. Engl. J. Med.337, 725-733 (1997). | Article | PubMed | ISI | ChemPort |
Harris, M. et al. Correlation of virus load in plasma and lymph node tissue in human immunodeficiency virus infection. J. Infect. Dis.176, 1388-1392 (1997). | PubMed | ISI | ChemPort |
Nabel, G. & Baltimore, D. An inducible transcription factor activates expression of human immunodeficiency virus in T cells. Nature326, 711-713 (1987). | Article | PubMed | ISI | ChemPort |
Folks, T.M. et al. Tumor necrosis factor alpha induces expression of human immunodeficiency virus in a chronically infected T-cell clone. Proc. Natl. Acad. Sci. USA86, 2365-2368 (1989). | PubMed | ChemPort |
Pomerantz, R.J., Trono, D., Feinberg, M.B., & Baltimore D. Cells nonproductively infected with HIV-1 exhibit an aberrant pattern of viral RNA expression, a molecular model for latency. Cell61, 1271-1276 (1990). | Article | PubMed | ISI | ChemPort |
Garcia-Blanco, M.A. & Cullen, B.R. Molecular basis of latency in pathogenic human viruses. Science254, 815-820 (1991). | PubMed | ChemPort |
Chun, T.-W. et al. Fate of HIV-1-infected T cells in vivo, Rates of transition to stable latency. Nature Med.1, 1284-1290 (1995). | Article | PubMed | ISI | ChemPort |
Chun, T.-W. et al. Quantitation of latent tissue reservoirs and total body load in HIV-1 infection. Nature387, 183-188 (1997). | Article | PubMed | ISI | ChemPort |
Folks, T. et al. Susceptibility of normal human lymphocytes to infection with HTLV-III/LAV. J. Immunol.136, 4049-4053 (1986). | PubMed | ISI | ChemPort |
Finzi, D. et al. Identification of a reservoir for HIV-1 in patients on highly active antiretroviral therapy. Science278, 1295-1300 (1997). | Article | PubMed | ISI | ChemPort |
Chun, T.W. et al. Presence of an inducible HIV-1 latent reservoir during highly active antiretroviral therapy. Proc. Natl. Acad. Sci. USA94, 13193-13197 (1997). | Article | PubMed | ChemPort |
Wong, J.K. et al. Reduction of HIV-1 in blood and lymph nodes following potent antiretroviral therapy and the virologic correlates of treatment failure. Proc. Natl. Acad. Sci. USA94, 12574-12579 (1997). | Article | PubMed | ChemPort |
Carpenter, C.C. et al. Antiretroviral therapy for HIV infection in 1998, undated recommendation of the International AIDS Society - USA Panel. J. Am. Med. Assoc.280, 78-86 (1998). | Article | ISI | ChemPort |
Bartlett, J.G. in Medical Management of HIV Infection (Johns Hopkins University, Baltimore, Maryland, 1998).
Palella, F.J. et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. N. Engl. J. Med.338, 853-860 (1998). | Article | PubMed | ISI |
Flexner, C. HIV-1 protease inhibitors. N. Engl. J. Med.338, 1281-1292 (1998). | Article | PubMed | ISI | ChemPort |
Mclean, A.R. & Michie, C.A. In vivo estimates of division and death rates of human T lymphocytes. X. Proc. Natl. Acad. Sci. USA92, 3707-3711 (1995). | PubMed | ChemPort |
Natarajan, V. et al. HIV-1 replication in patients with undetectable plasma virus receiving HAART. Highly active antiretroviral therapy. Lancet353, 119-120 (1999). | Article | PubMed | ISI | ChemPort |
Gunthard, H.F. et al. Human immunodeficiency virus replication and genotypic resistance in blood and lymph nodes after a year of potent antiretroviral therapy. J. Virol.72, 2422-2428 (1998). | PubMed | ISI | ChemPort |
Rosenberg, E.S. et al. Vigorous HIV-1-specific CD4+ T cell responses associated with the control of viremia. Science278, 1447-1450 (1997). | Article | PubMed | ISI | ChemPort |
Jacobson, E.L., Pilaro, F., & Smith, K.A. Rational interleukin 2 therapy for HIV positive individuals, Daily low doses enhace immune fuction without toxicity. Proc. Natl. Acad. Sci. USA93, 10405-10410 (1996). | Article | PubMed | ChemPort |
Chun, T.-W. et al. Effect of interleukin-2 in diminution of a pool of latently infected, resting CD4+ T cells in HIV-1 infected patients receiving highly active antiretroviral therapy (6th Conference on Retroviruses and Opportunistic Infections, Chicago, Illinois, 1999).
Acknowledgments We thank C. Raines, S.Barnett, B. Perdue-Sabundayo and J. Keruly for coordinating patient visits and help with data analysis. We also thank L. Carruth, J. L'Esperance and C. Murray for help with the experiments. We thank Y. Afacan for providing patients and R. Brookmeyer for advice on statistical analysis of decay rates. This work was supported by NIH grant AI43222 to R.F.S.
105 cells, eradication could take as long as 60 years. Thus, latent infection of resting CD4+ T cells provides a mechanism for lifelong persistence of HIV-1, even in patients on effective anti-retroviral therapy.
