Skip Navigation

The Effect of Commencing Combination Antiretroviral Therapy Soon after Human Immunodeficiency Virus Type 1 Infection on Viral Replication and Antiviral Immune Responses

  1. Martin Markowitz,
  2. Mika Vesanen,
  3. Klara Tenner-Racz,
  4. Yunzhen Cao1,
  5. James M. Binley,
  6. Andrew Talal,
  7. Arlene Hurley,
  8. Xia Ji,
  9. M. Rashid Chaudhry,
  10. Melody Yaman,
  11. Sarah Frankel,
  12. Margo Heath-Chiozzi,
  13. John M. Leonard,
  14. John P. Moore,
  15. Paul Racz,
  16. Douglas F. Nixon and
  17. David D. Ho
  1. Aaron Diamond AIDS Research Center, Rockefeller University, New York, New York; Bernhard-Nocht-Institut fur Tropenmedizin, Hamburg, Germany; Abbott Laboratories, Abbott Park, Illinois; Division of Retrovirology, Walter Reed Army Institute of Research, Rockville, Maryland; Department of Infectious and Parasitic Disease Pathology, Armed Forces Institute of Pathology, Washington, DC
  1. Reprints or correspondence: Dr. Martin Markowitz, Aaron Diamond AIDS Research Center, 455 First Ave., New York, NY 10016 (marty{at}adarc.org).

  1. Presented in part: 4th Conference on Retroviruses and Opportunistic Infections, Washington, DC, January 1997.

  • ↵1 Present affiliation: Department of Clinical Virology, National AIDS Prevention and Control Center, Beijing, Peoples' Republic of China.

 
Next Section

Abstract

Twelve subjects were treated with zidovudine, lamivudine, and ritonavir within 90 days of onset of symptoms of acute infection to determine whether human immunodeficiency virus type 1 (HIV-1) infection could be eradicated from an infected host. In adherent subjects, with or without modifications due to intolerance, viral replication was suppressed during the 24-month treatment period. Durable suppression reduced levels of HIV-1—specific antibodies and cytotoxic T lymphocyte responses in selected subjects. Proviral DNA in mononuclear cells uniformly persisted. The persistence of HIV-1 RNA expression in lymphoid tissues and peripheral blood mononuclear cells suggests that elimination of this residual pool of virus should be achieved before considering adjustments in antiretroviral therapeutic regimens. In addition, given the reduction in levels of virus-specific immune responses, it would seem prudent to consider enhancing these responses using vaccine strategies prior to the withdrawal of antiviral therapy.

Acute infection with human immunodeficiency virus type 1 (HIV-1) is characterized by high levels of viral replication [1, 2], often associated with a clinical syndrome that includes fever, rash, lymphadenopathy, headache, myalgia, and gastrointestinal and neurologic symptoms [3–5]. Probably because of a genetic bottleneck during or soon after transmission, the initial expansion of HIV-1 in the new host typically involves a relatively clonal virus population [6, 7]. With the onset of a cellular immune response, principally cytotoxic T lymphocytes (CTL), there is a partial, and at times marked, reduction in viremia [8, 9]. Characteristically, a quasi-steadystate (set point) is established by 9–12 months after the initial infection, when HIV-1 production and clearance are in approximate balance [10–12]. In characterizing 74 persons newly infected with HIV-1, Schacker et al. [13] showed that the inflection point in the reduction in plasma HIV-1 RNA levels occurs, on average, as early as 117 days after infection. Despite the presence of strong cellular and humoral immune responses during this period of chronic infection, viral replication remains high and drives a gradual destruction of the immune system [14, 15], dominated by an inexorable loss of CD4 T lymphocytes. Severe immunodeficiency and death ultimately ensue in nearly all untreated infected persons.

The level of HIV-1 in plasma at the set point is highly predictive of the subsequent clinical course [10–13, 16]. Between 93% and 99% of plasma virus is produced by rapid rounds of infection and destruction of activated CD4 T cells, with the rest produced by long-lived chronically infected cells, such as tissue macrophages and dendritic cells, latently infected lymphocytes, and release of trapped virions from lymphoid tissue [17]. Effective antiviral therapy rapidly interrupts the replication of HIV-1 in CD4 T cells and prevents the spread of infectious virus from the longer-lived cells; the half-lives of HIV-1 production in these two compartments are ∼1 day and ∼2 weeks, respectively [17]. Although quantitatively minor, a third viral compartment consisting of resting memory CD4 T cells harboring integrated, replication-competent HIV-1 has been identified and could serve as a persistent reservoir of latent virus despite prolonged antiviral therapy [18].

At the time this clinical trial was designed, having observed a 99% reduction in plasma HIV-1 RNA in subjects treated with protease inhibitor monotherapy [19, 20], we asked whether HIV-1 infection could be eradicated with a combination of the most active therapies available when applied in the best of clinical situations: a newly infected host with a drug-susceptible virus population and a minimally affected immune system.

We selected the combination of ritonavir, a potent protease inhibitor, with the reverse transcriptase (RT) inhibitors zidovudine and lamivudine. Used alone, ritonavir can cause ∼2 log10 reductions in plasma viremia [19, 20], and the combination of zidovudine with lamivudine also has a powerful antiviral effect [21]. We therefore anticipated that the triple combination would be highly effective at reducing HIV-1 replication and restoring CD4 T cells in vivo. Triple therapy with ritonavir (1200 mg/day), zidovudine (600 mg/day), and lamivudine (300 mg/day) was begun in July 1995 in 12 subjects within 90 days of symptoms of acute HIV-1 infection. Patient acceptance and tolerance of the regimen was poor, and subsequent experience with ritonavir clearly showed that the drug in the 600 mg twice-a-day dosage can be difficult to tolerate, particularly in combination with zidovudine (unpublished data). Despite these inherent limitations, we report here a detailed analysis of the virologic effects of this regimen in both peripheral blood and lymphoid tissue and the immunologic consequences of this regimen over the first 2 years of treatment.

Our more recent research and that of others indicate that eradication of HIV-1 from the infected host may require treatment regimens that are more sophisticated and of longer duration. This proof-of-concept study remains valuable nevertheless for assessing the virologic and immunologic effect of early intervention with highly active antiretroviral therapy. In addition, data generated from this and other studies [22, 23] using highly effective combination therapy have strongly influenced recent guidelines on treating HIV infection [24].

Previous SectionNext Section

Methods

Study design

Twelve subjects newly infected with HIV-1 were recruited for this open-label, noncontrolled pilot study between July 1995 and February 1996. Entry criteria included plasma viremia detectable by the branched DNA (bDNA) assay (2.0; Chiron Diagnostics, Emeryville, CA), with a detection limit of 500 HIV-1 RNA copies/mL plasma, and either an initially negative test for HIV-1 antibodies followed by the appearance of ⩾2 new bands on a Western blot or a convincing clinical history for acute HIV-1 infection within the prior 90 days plus a documented negative antibody test within 120 days of screening. Subjects were treated with escalating doses of ritonavir over 7–10 days to achieve a maximum final dose of 1200 mg/day. Zidovudine and lamivudine were used from the outset at dosages of 600 mg/day and 300 mg/day, respectively. Therapy was initiated on an inpatient basis but continued on an outpatient basis. After the initial hospitalization of ∼7 days, subjects were seen weekly to week 4, biweekly to week 12, and then monthly, at which time blood was collected and plasma and cells separated and cryopreserved for the virologic and immunologic studies detailed below. Adherence to therapy was determined by drug reconciliation and patient history. Semen was collected from all subjects after 1 year of therapy. Biopsy of gastrointestinal tissue by flexible sigmoidoscopy to obtain gastrointestinal or gut-associated lymphoid tissue (GALT) was performed in all subjects after a minimum of 1 year of therapy. After 15 months of therapy, 1 patient (subject 9) consented to a repeat gut biopsy, simultaneous biopsies of cervical lymph node and tonsil, and a lumbar puncture. Additional subjects consented to biopsies of either cervical or inguinal node or tonsil after ∼24 months of therapy. Tissues were snap-frozen and stored at −70°C for subsequent polymerase chain reaction (PCR) studies, and some were formalin-fixed and paraffin-embedded for in situ examination.

Virologic measurements

Plasma HIV-1 RNA was quantified longitudinally using both the bDNA signal amplification assay [25] and the ultrasensitive RT-PCR assay [26] (Roche Diagnostics, Alameda, CA), which have detection limits of 500 and 50 RNA copies/mL, respectively. Infectious HIV-1 titers in peripheral blood and lymphoid tissue were determined by limiting-dilution cocultures and expressed as TCID50/106 mononuclear cells [27]. The coculture procedure of patient peripheral blood mononuclear cells (PBMC) was modified when infectious titers were persistently <0.1 TCID50/106 cells by directly stimulating donor and patient PBMC with phytohemagglutinin (PHA) and depleting them of CD8 T cells [28–30]. Semen was collected and processed within 6 h. After a 1: 4 dilution with RPMI, cells were separated from seminal plasma by centrifugation at 800 g. Cells were stored at −70°C for subsequent analysis. Levels of multiply spliced (MS) and unspliced (US) HIV-1—specific mRNA and proviral DNA in mononuclear cells from blood, genital secretions, and tissue were determined by PCR as described previously [31, 32]. In situ hybridization on formalin-fixed, paraffin-embedded lymphoid tissue was performed using HIV-1—specific antisense probes corresponding to >90% of the viral genome [33, 34]. GALT was histologically graded 1–4 as follows: 1, scattered lymphoid cells; 2, small lymphoid aggregate; 3, large well-defined aggregate; and 4, germinal center present.

Humoral and cellular immunology

CD4 and CD8 T cell counts were assessed longitudinally using standard techniques. Routine screening ELISA and Western blots were performed. Plasma antibody titers to HIV-1 gp120 and p24 proteins were measured as described elsewhere [35]. HIV-1—specific CTL precursor frequency (CTLp) was quantitated using autologous B lymphoblastoid cell targets infected with recombinant vaccinia viruses expressing HIV-1—specific antigens (gag, pol, env, or nef). Direct CTL effector activity was measured using freshly isolated PBMC, per published methods [8].

Previous SectionNext Section

Results

Study design, compliance, and adherence

All study subjects were homosexual men with symptomatic acute HIV infection whose treatment commenced, on average, 65 days (range, 45–90) from onset of symptoms. For subjects able to define when they became infected, their symptoms commenced a mean of 15 days (range, 6–20) after the reported high-risk sexual exposure. Seven of the original 12 participants have now remained on either the original regimen or a modified highly active regimen for up to 24 months (table 1). Gastrointestinal adverse events led 2 subjects (1 and 12) to discontinue drugs and withdraw early from the study at weeks 4 and 52. Subjects 4 and 10 voluntarily withdrew from the trial after 24 and 8 weeks, respectively, because of the rigorous protocol schedule. Participants were questioned at each clinic visit to detail the number of missed doses, if any, of the 3 study medications, to monitor for adherence to the prescribed drug regimen. At the month 20 visit, when results of plasma RNA measurements using a new ultrasensitive assay with a lower detection limit of 50 RNA copies/mL plasma were made available, subject 11 described reducing ritonavir doses by 50% after only 2 months of therapy and not taking any of the prescribed drugs about 25% of the time. These 5 subjects (1, 4, 10, 11, and 12) were formally removed from the trial, although some findings for subject 11 are presented below. Subject 8 briefly interrupted therapy after 19 months because of emotional distress unrelated to therapy and was promptly restarted on the identical regimen after a rapid virologic rebound was documented (see below).

In addition, at the month 21 visit, when a clear rebound in HIV-1 RNA was noted, subject 6 described altering his treatment regimen. He reduced his dose of ritonavir after ∼1 year to 50%, and ceased all 3 drugs 2 months thereafter (table 1). After recommencing therapy with the original regimen, he remained in the trial, and findings are presented below. All 5 remaining subjects (2, 3, 5, 7, and 9) were >95% adherent to the prescribed regimen, judged both by clinical history and drug reconciliation. However, 2 participants required replacements for ritonavir; 1 by indinavir at month 8 because of allergy (subject 3), the other by nevirapine at month 12 because of exacerbation of chronic hepatitis C infection (subject 5). Subject 2 had been maintained on a reduced (400 mg/day) dose of zidovudine because of grade 4 elevations in serum creatine phosphokinase levels beginning at week 12. Subject 9 received a slightly reduced (1000 mg/day) dose of ritonavir beginning at week 6 because of early grade 4, drug-related elevations in serum levels of transaminases on the original dose. We do not believe that the use of replacement therapies or dose alterations in these 4 subjects compromises the analyses, since HIV-1 replication in these subjects was successfully suppressed, consistent with the original intent of the clinical experiment.

Effects of triple-drug therapy on plasma viremia

Longitudinal plasma viremia (HIV-1 RNA levels) measured by both signal amplification (bDNA) assays with a detection limit of 500 copies/mL and PCR-based assays with a detection limit of 50 copies/mL are shown in figure 1A for the adherent subjects (2, 3, 5, 7, and 9) and in figure 1B for the nonadherent subjects (6, 8, and 11).

Figure 1
View larger version:
Figure 1

Longitudinal plasma HIV-1 RNA copies and CD4 cells for 8 subjects on therapy ⩾12 months. Open red circles, HIV-1 RNA by branched DNA (lower limit, 500 copies/mL); + and −, detectable and undetectable (<50 copies/mL) plasma HIV-1 RNA by ultrasensitive polymerase chain reaction, respectively; green dots, absolute CD4 cell count.

All 12 participants initially responded to triple-drug therapy. An undetectable level of HIV RNA in plasma, as determined by bDNA (<500 RNA copies/mL), was reached after a mean of 28.8 days and a median of 9.5 days (range, 4–97). The mean maximum reduction in plasma RNA was 2.2 log10 (data not shown). In 7 subjects with baseline plasma RNA values >10,000, a typical biphasic pattern of HIV-1 RNA decay was observed. Mean and median half-lives for the initial rapid phase were 2.4 and 1.9 days, respectively (range, 1.2–5.3). In the 6 patients in whom a second slower phase was measurable, the mean and median half-lives were 22.4 and 19.2 days, respectively (range, 112–46.2). At the time they discontinued therapy, subjects 1 and 10 both had undetectable plasma viremia by the bDNA assay. However, viremia in subjects 4 and 12 was rebounding at discontinuation, presumably reflecting the difficulties they had in following the therapeutic protocol (data not shown).

Plasma viremia remained persistently below the detection threshold of 50 copies/mL in compliant subjects 2, 3, 7, and 9 (figure 1A). Sporadic increases in plasma viremia to levels between 50 and 500 copies/mL were detected in subject 5 around month 12, when ritonavir was replaced by nevirapine because of complications with chronic hepatitis C infection (figure 1A). An unexplained further sporadic increase in plasma viremia was noted near month 18, but no sustained viral replication has since been documented in this person. Plasma viremia in nonadherent subject 11 was never suppressed to <50 copies/mL but still remained below 500 copies/mL despite his less-than-optimal therapy. Furthermore, despite complete discontinuation of all medications at month 18, the level of HIV RNA detectable in plasma was never above 500 copies/mL for 6 months but was detectable by the ultrasensitive RNA-PCR assay (figure 1B). Although subject 6 completely discontinued therapy at month 14 after 2 months of ritonavir at a reduced dose, his plasma viremia remained low (50–500 copies/mL) for 4 months before rebounding to >500 copies/mL (figure 1B). Conversely, subject 8 chose to discontinue therapy shortly after his month 19 visit, and his plasma viremia level rebounded rapidly (figure 1B).

Virus isolation and detection of HIV-1 in tissues

The results of cross-sectional analyses of HIV-1 infectivity and viral mRNA in PBMC and GALT of 5 adherent and 3 nonadherent subjects are presented in table 2, as are tonsil and lymph node tissue findings of subjects 2, 3, and 9. To assess whether infectious HIV-1 could be isolated from PBMC, qualitative cocultures were performed on CD8-depleted, PHA-stimulated cells after therapy for 19–24 months. No infectious HIV-1 was detected (<0.1 TCID50/106 CD4 T cells) in subjects 2, 3, and 7, consistent with their undetectable plasma viremia. No infectious HIV-1 was detected in subject 8 at a time that he was both adherent and aviremic. However, unexpectedly, the coculture was positive for subject 9, whose plasma viral RNA was consistently <50 copies/mL. In keeping with their low but persistent plasma viremia, the cocultures for subjects 5, 6, and 11 were positive.

We next used a PCR assay to measure MS and US HIV-1 mRNA in subjects' PBMC. The presence of MS mRNA is indicative of ongoing viral replication, whereas the detection of US mRNA could be indicative of the presence of virions regardless of their infectivity [31, 32]. As shown in table 2, both MS and US mRNA species were detected in the PBMC of subject 11. For the other subjects, no MS mRNA was found in PBMC, although US mRNA was detectable in some subjects (2, 3, 5, and 6). No HIV-1 mRNA of either form was detectable in PBMC from subjects 7, 8 (with a brief exception), and 9, 3 of the subjects who were most able to adhere to the original treatment protocol up to the time these analyses were done. HIV-1 DNA was uniformly detectable in PBMC of every subject, regardless of the degree of suppression of plasma viremia (table 2). PBMC isolated from semen after 12 months of therapy in subjects 2, 3, 5, 6, 7, 8, 9, and 11 were all positive for proviral DNA and negative for both MS and US mRNA (data not shown). HIV RNA was undetectable by the ultrasensitive RNA-PCR method in the cerebrospinal fluid obtained by lumbar puncture from subject 9 after 15 months of therapy (data not shown).

Results of analyses of the HIV-1 content of GALT samples during months 12–17 are shown in table 2. Mononuclear cells were isolated from biopsy specimens by tissue homogenization followed by ficoll-hypaque density gradient before placement in quantitative cocultures to detect infectious HIV-1. In each case, GALT obtained during months 12–17 of treatment were negative for infectious virus (i.e., <0.1–1.0 TCID50/106 mononuclear cells). Adequate numbers of CD4 T cells could not be isolated from GALT, precluding more sensitive qualitative co-culture, as described for PBMC.

Messenger RNA was extracted from the total GALT RNA pool for detection of HIV-1 MS and US mRNA. As summarized in table 2, no MS mRNA was detected in any case, whereas US mRNA was detectable in nearly all subjects. Proviral DNA was routinely detected in the mononuclear cells derived from GALT of all subjects. Tonsil and cervical lymph node biopsies were performed on subject 9 after 15 months of therapy. Analyses of these tissues yielded results similar to those derived from GALT: no culturable virus and undetectable MS mRNA, but detectable US mRNA and proviral DNA. At 24 months, patients 2 and 3 underwent tonsillar biopsy and inguinal lymph node biopsies, respectively. These tissues were negative in viral coculture and contained no detectable US or MS mRNA.

In situ hybridization was also performed to detect HIV-1 RNA expression in lymphoid tissues (figure 2). An optimal biopsy from an HIV-1—infected untreated control with a large well-defined lymphoid nodule is shown in figure 2A. The large amount of trapped viral RNA and the intense signal from some productively infected cells from the untreated control are evident in panels 2B and 2C, respectively. Gastrointestinal biopsy samples from 7 subjects (2, 3, 5–8, and 11) who remained on therapy ⩾12 months are shown in panels 2D–2K. Although only a limited number of biopsies could be performed on each subject, HIV-1 RNA-expressing cells could not be detected in samples from any of these subjects. In 3 GALT biopsy sections from subjects 2, 6, and 7 in which lymphoid follicles were observed, no trapped HIV-1 was detected. A more extensive series of biopsies was performed on subject 9 (figure 2L–2Q); HIV-1—specific RNA was rarely detected in individual cells in rectal, tonsil, and cervical node tissue (figure 2N, 2P, 2Q). Germinal centers in tonsil and lymph node were free of trapped virus (figure 2L, 2M), and the occasional RNA-positive cell had relatively few grains (7–37) compared with untreated controls (figure 2C), in which there were too many grains to count. In situ hybridization of the tonsil and inguinal node tissue from subjects 2 and 3 obtained after 24 months of therapy were negative for trapped virus in the germinal centers, and no RNA-expressing cells were seen in >30 sections examined (data not shown).

Figure 2
View larger version:
Figure 2

Lymphoid tissue histology and in situ hybridization (ISH). A, B, Rectal biopsy in untreated control with submucosal lymphoid nodule (hematoxylin-eosin stain, ×25); ISH with large amounts of HIV-1 RNA (blue-green). C, Sigmoid biopsy in positive control with multiple HIV-1 RNA cells (×50). D, Subject 2, grade 4 gut-associated lymphoid tissue (GALT) germinal center (×100), ISH-negative for HIV RNA in cells and follicular dendritic cell (FDC) network. E, Subject 3, grade 2 GALT, no HIV RNA by ISH (×50). F Subject 5, grade 3 GALT, no HIV RNA by ISH (×50). G and H, Subjects 6 and 7, respectively, grade 4 GALT, no HIV RNA by ISH (×50). J and K, Subjects 8 and 11, respectively, grade 2 GALT, no HIV RNA by ISH (×50). L, M, Subject 9, germinal centers in tonsil and lymph node, no FDC-associated HIV RNA (L, ×100; M, ×157.5). N, P, Q, Subject 9, rectum, lymph node, and tonsil tissue with rare HIV RNA by ISH (N, P, ×100; Q, ×157.5). (D–Q, Giemsa stain.)

Cellular and humoral immunology

Lymphocyte subset analyses are presented longitudinally in figure 1 for 8 subjects. At 18 months, mean CD4 cells had increased significantly by 290 cells/mm3 (P <.001), whereas mean CD8 cells fell by 129 cells/mm3 (P = .3).

The binding antibody responses (midpoint titers) to the HIV-1 gp120 and p24 proteins are presented for subjects 2, 3, 5–9, and 11 in figure 3. In general, antibody titers remained substantially lower than the titers of ⩾ 104 typically found in chronically infected persons [35]. In addition, antibody titers to gp120 were 1–2 logs higher than those to p24, which were often undetectable. Among subjects with consistently undetectable levels of plasma HIV-1 RNA, only subject 9 had a rising anti-gp120 titer, and the increase was modest. In subject 6, there was a rapid increase in antibody titers about the time he discontinued therapy; of note, the increase in his anti-gp120 titer preceded the rise in plasma viremia (figure 1B). Subject 11 had the highest antibody titers to both gp120 and p24, probably because of the relatively incomplete suppression of HIV-1 replication in this noncompliant subject.

Figure 3
View larger version:
Figure 3

Longitudinal midpoint antibody titers to env (gp120) (red dots) and gag (p24) (green triangles) in 8 subjects on triple therapy >12 months.

“Fresh” (unstimulated) CTL responses to env, gag, pol, and nef antigens were very low (<10% specific lysis) at all times in each trial participant and did not change in response to reductions in plasma viremia (data not shown). CTLp analyses revealed three different patterns of response in the 7 persons studied (figure 4). No CTLp were detectable in subject 5 at baseline, on day 28, or at month 20. Subjects 9 and 11 had either a broadening or an increase in CTLp levels over time, suggesting that HIV-1 replication was still occurring in some tissue compartment. Of note was the late broadening in CTLp frequencies in subject 9, about the time his anti-gp120 titers were gradually rising, despite the absence of detectable plasma viremia and his sustained compliance with the therapeutic protocol (figures 3, 4). However, subjects 3, 6, 7, and 8 all had undetectable or significantly reduced CTLp levels at the end of the analytical period.

Figure 4
View larger version:
Figure 4

Cytotoxic T lymphocyte precursor (CTLp) levels/106 peripheral blood mononuclear cells (PBMC) to HIV env (green), gag (blue), pol (yellow), and nef (red) in 6 subjects treated with antiretroviral therapy >12 months.

Previous SectionNext Section

Discussion

The original goal of this clinical experiment was to completely suppress viral replication in a cohort of newly infected subjects and thereby attempt virus eradication. This represents a major scientific challenge. The initial triple combination regimen was difficult to maintain; 4 subjects withdrew, dosage adjustments were made in 2 others, and poor compliance was noted in an additional 3 subjects. We believe this was due to the use of ritonavir in daily doses of 1200 mg. Subsequent studies by our group using alternative protease inhibitor—containing regimens (e.g., indinavir, 2400 mg/day, or ritonavir, 800–1200 mg/day, in combination with saquinavir, 800–1200 mg/day) with double-nucleoside RT inhibitor therapy have resulted in improved patient acceptance and adherence while achieving virologic results similar to those documented above [36, 37]. This supports the feasibility of the use of combination therapy in the acute and early stages of HIV-1 infection.

Because the use of replacement therapies or dose alterations did not compromise suppression of HIV-1 replication in the study subjects, the virologic and immunologic data collected reflect the original intent of the clinical experiment. The initial virologic response to the triple-drug regimen was robust, and the dynamics of the response comparable to our previous reports [15, 17]. Comprehensive virologic studies of blood, tissue, and genital secretions for up to 24 months of therapy suggest complete suppression of ongoing productive HIV replication in subjects 2, 3, 7, and possibly 9. A tonsillar biopsy on patient 2 and an inguinal node biopsy on patient 3 after 24 months of therapy support these conclusions; there was an absence of follicular dendritic cell (FDC)—associated HIV RNA in multiple germinal centers, and HIV RNA-expressing cells were absent from up to 30 sections of tonsillar or lymphoid tissue examined by in situ hybridization (data not shown). The demonstration of plasma HIV RNA levels below detection by an ultrasensitive assay (50 RNA copies/mL) and the absence of MS mRNA expression in blood and tissue, combined with low to undetectable levels of HIV-1—specific immune responses, supports this conclusion.

The virologic data were somewhat equivocal in subject 5, with two isolated instances of measurable levels of HIV RNA, albeit at very low levels. In addition, infectious virus could be isolated from 1.5 × 107 CD8-depleted, PHA-stimulated CD4 cells. It is unclear whether this reflects drug failure due to emerging resistance or the occasionally visible expression of virus from slowly decaying infected cellular compartments in blood and tissue. Our genotypic analysis of the culturable virus showed no evidence of drug-resistant mutations (data not shown), a finding consistent with the lack of persistent measurable plasma viremia in this patient. Despite undetectable plasma HIV-1 RNA levels, subject 9 had both rising antibody titers to env and broadening of the CTL response, both unexpected findings. Virus-positive, CD8-depleted, PHA-stimulated CD4 cell cocultures and persistent HIV-1 RNA expression in biopsied tissue may in this case represent early escape from the therapeutic regimen. However, genotypic analysis of the cultured virus revealed wild type sequences that were more consistent with the virus emerging from cells latently infected prior to the initiation of therapy. Moreover, results from in situ hybridization studies of a tonsillar biopsy done 4 months after the observed increase in both antibody levels and the broadening of the CTL response revealed neither RNA expression nor FDC-trapped virus (data not shown). Levels of MS and US mRNA species as determined by PCR from this tonsil biopsy were below the level of detection (table 2), and the plasma HIV-1 RNA level also remained undetectable. Clearly, it is possible that a viral compartment persists somewhere in this patient, allowing sufficient HIV-1 replication to stimulate specific immune responses, but insufficient in quantity to allow detection by current methodologies. Alternatively, since subject 9 appeared to be as completely suppressed as subjects 2, 3, and 7 (described above), the immune phenomena observed may reflect viral persistence and expression without ongoing viral replication.

Table 1
View larger version:
Table 1

Baseline characteristics of the study subjects.

Table 2
View larger version:
Table 2

Cross-sectional analyses of HIV-1 infectivity and viral mRNA in peripheral blood mononuclear cells (PBMC) and gut-associated lymphoid tissue (GALT) of adherent and nonadherent study subjects.

Cross-sectional measurements revealing low levels of US mRNA expression in the blood of apparently well-suppressed persons suggests the possibility of very low level ongoing viral replication. Alternatively, US mRNA may arise from activation of latently infected cells resulting in RNA expression without de novo infection of the susceptible cell population. In lymphoid tissue, detectable US mRNA may similarly arise from chronically infected long-lived cells or may represent low-level residual virions trapped on follicular dendritic cells. Given the presence of detectable proviral DNA in blood and tissue, it is likely that detectable US mRNA species arise in part from a pool of latently infected resting CD4 lymphocytes stochastically activated to produce virus, a portion of which may be infectious.

The size of the persistent latent pool harboring infectious provirus is indeed critical. Our inability to culture virus from 1.5 × 107 circulating CD8-depleted PHA-stimulated CD4 T cells suggests that the pool size of these latently infected resting cells harboring infectious provirus may be ⩽3 × 104. Vesanen et al. [38] measured levels of proviral DNA from this cohort and estimated the average half-life of this pool as ∼3 months (decay constant of ∼0.0007/day). Thus, on average, ⩽210 cells would be activated to produce infectious virus each day. Whether this low level of potential viral replication in vivo can be controlled in the absence of antiviral drugs by either HIV-specific immunologic factors (CTL or antibody) or nonspecific antiviral factors (chemokines, interleukins, soluble factors, or cytokines) is unknown. However, the persistence of proviral DNA in genital secretions and the unanswered question of viral persistence in the central nervous system, both immunologically privileged sites, raises additional concerns of the feasibility of immunologic control of viral replication in the absence of antiviral therapy.

The virologic and immunologic profiles observed in subject 6 after drug discontinuation are of particular interest. That there was a 4-month lag between drug cessation and clear viral rebound with an apparently drug-sensitive virus population is a provocative finding. Following drug discontinuation, a new equilibrium was established. After a lag period, the HIV RNA level climbed precipitously, only to drop again with reinstitution of the original regimen. Perhaps the early phase of limited viral replication reflects host factors, presumably immunologic, that initially substantially restrict viral replication. The rebound of a drug-sensitive virus population could have resulted from de novo mutation resulting in immunologic escape. Alternatively, the rebound viremia might represent the activation of latent virus from a cell infected before the initiation of therapy, producing a drug-sensitive yet antigenically distinct virus population.

The rapid rebound after the temporary discontinuation of antiviral therapy in subject 8 highlights the importance of the persistence of proviral DNA despite successful suppression of viral replication in peripheral blood and tissue (GALT) for ⩽19 months. Although control of viral replication was easily regained with the reintroduction of the same drug regimen, the rapidity of the rebound suggests that the multiple rounds of viral replication necessary to sustain significant levels of plasma HIV RNA were easily reignited, presumably from a population of cells infected prior to the initiation of the original therapy.

We observed that early combination antiretroviral therapy can result in arrest of the maturation, and subsequent decay, of HIV-1—specific immune responses, both humoral and cellular. We have documented reduction in both HIV-specific and-nonspecific B cell responses after initiation of antiretroviral therapy in cohorts of both newly and chronically HIV-infected persons [39]. Reduction in HIV-specific B cell activity is likely due to the reduction in viral antigen load. The observed reduction in hypergammaglobulinemia and the frequency of circulating IgG antigen—secreting cells associated with antiviral therapy may reflect reduction in cellular activation dependent on continuous HIV-1 replication, although this explanation remains hypothetical.

Evidence has accumulated that the activity of HIV-specific CTL is important in controlling HIV-1 replication. The resolution of high levels of viremia observed during primary infection is temporally associated with the appearance of HIV-specific CTL [8, 9]. We recently reported an inverse correlation between CTL activity and plasma HIV RNA in untreated HIV-1—infected subjects [40]. In a study of 33 persons with untreated primary HIV-1 infection, Musey et al. [41] reported the persistence of broad CTL activity in association with control of plasma viremia. In our treated subjects, we frequently observed the opposite phenomenon. Early pharmacologic control of viremia was generally associated with decreasing levels of CTL activity. We hypothesize that the reduction in these HIV-specific cellular responses results from a reduction in antigen load, below a level necessary to sustain immunologic stimulation at a maximal level. However, a direct inhibitory effect of antiviral therapy on CTL activity could not be ruled out. While these findings are encouraging in indicating the extent of virus suppression mediated by the therapy, it may be prudent to enhance HIV-specific immune responses with vaccination strategies before stopping the antiviral regimen. This patient population, that is, those treated early in infection, is ideal for such manipulations, since HIV-1—specific CD4 T helper activities are likely to be well-preserved [42]. What remains less clear is the most appropriate HIV-1 vaccine to use for effective immunization strategies [43].

In summary, combination therapy with multiple drugs targeting both HIV-1 RT and protease can potently and durably suppress HIV-1 replication in vivo. However, it is equally important to note that while this regimen was highly effective virologically, it was not well-tolerated. The use of antiretroviral agents in this setting must therefore include a consideration of issues surrounding patient adherence and tolerability. When applied early after infection, control of viral replication is associated with preservation of normal absolute CD4 cells, increases in the CD4-to-CD8 cell ratio, preservation of lymphoid architecture, and a dramatic reduction in HIV-1 in blood and tissue, along with waning HIV-1—specific immune responses. However, multiple lines of evidence indicate that replication-competent HIV-1 remains in persons whose infection has been suppressed for a prolonged period. This conclusion is consistent with recent reports that, after prolonged combination therapy, replication-competent HIV-1 remains detectable in PBMC [28–30], probably resident in resting CD4 T cells. Thus, the eradication of HIV-1 from infected persons will not be simple [44]. Nevertheless, this study provides virologic and immunologic insights to aid in planning future proof-of-concept clinical trials. New therapeutic approaches must also include strategies to immunologically activate resting CD4 T cells that latently harbor residual HIV-1. Also, it may be necessary to stimulate HIV-1—specific immunity with vaccine strategies before contemplating the withdrawal of antiviral therapy.

Previous SectionNext Section

Acknowledgments

We thank Rosemary Schluger, Zhayao Zheng, Peter Kuebler, Caroline Roberts, William Kakimoto, Brian Clas, Gudrun Grobschupff, Birgit Raschdorff, Johanna Andersen, and Olga Ford for clinical and laboratory assistance; Richard Fogler for surgical assistance; Amy Keller, Norman Mustafa, and Lynn Smiley (Glaxo/Wellcome) for zidovudine and lamivudine; Chiron and Roche Diagnostics for viral load measurements; W. Chen for assistance with graphics; Sarah Beatrice, Amado Punsalang, and William Oleszko (New York City Department of Health, Bureau of Laboratories, Retrovirology, and Immunobiology) for Western blots; Project Achieve, New York; referring physicians for identifying study participants; and the study participants for their cooperation and dedication.

Previous SectionNext Section

Footnotes

  • ↵M.H.-C. and J.M.L. are employees of Abbott Laboratories.

  • Informed consent was obtained from all subjects, and human experimentation guidelines of the US Department of Health and Human Services and of the Rockefeller University institutional review board were strictly followed in the conduct of this research.

  • Grant support: NIH (RR-00102, AI-41534, and AI-40387), German Ministry of Education and Research (BMBF 01-K1-9469), Korber Foundation (Hamburg), Vincent P. Beltosky, Jr., Foundation, and Abbott Laboratories.

  • Received June 11, 1998.
  • Revision received September 25, 1998.
Previous Section
 

References

  1. 1.↵
    1. Daar ES,
    2. Moudgil T,
    3. Meyer RD,
    4. Ho DD
    . Transient high levels of viremia in patients with primary human immunodeficiency virus type 1 infection. N Engl J Med 1991;324:961-4.
    MedlineWeb of Science
  2. 2.↵
    1. Clark SJ,
    2. Saag MS,
    3. Decker WD,
    4. et al
    . High titers of cytopathic virus in plasma of patients with symptomatic primary HIV-1 infection. N Engl J Med 1991;324:954-60.
    MedlineWeb of Science
  3. 3.↵
    1. Cooper DA,
    2. Gold J,
    3. Maclean P,
    4. et al
    . Acute AIDS retrovirus infection: definition of a clinical illness associated with seroconversion. Lancet 1985;1:537-40.
    MedlineWeb of Science
  4. 4.
    1. Ho DD,
    2. Sargadharan MG,
    3. Resnick L,
    4. DiMarzo-Veronese F,
    5. Rota TR,
    6. Hirsch MS
    . Primary human T lymphotropic virus type III infection. Ann Intern Med 1985;103:880-3.
    Abstract/FREE Full Text
  5. 5.↵
    1. Schacker T,
    2. Collier AC,
    3. Hughes J,
    4. Shea T,
    5. Corey L
    . Clinical and epidemiologic features of primary HIV infection. Ann Intern Med 1996;125:257-64.
    Abstract/FREE Full Text
  6. 6.↵
    1. Zhu T,
    2. Mo H,
    3. Wang N,
    4. et al
    . Genotypic and phenotypic characterization of HIV-1 in patients with primary infection. Science 1993;261:1179-81.
    Abstract/FREE Full Text
  7. 7.↵
    1. Zhang LQ,
    2. MacKenzie P,
    3. Cleland A,
    4. Holmes EC,
    5. Leigh Brown AJ,
    6. Simmonds P
    . Selection for specific sequences in the external envelope protein of human immunodeficiency virus type 1 upon primary infection. J Virol 1993;67:3345-56.
    Abstract/FREE Full Text
  8. 8.↵
    1. Koup RA,
    2. Safrit JT,
    3. Cao Y,
    4. et al
    . Temporal association of cellular immune response with the initial control of viremia in primary HIV-1 syndrome. J Virol 1994;68:4650-5.
    Abstract/FREE Full Text
  9. 9.↵
    1. Borrow P,
    2. Lewicki H,
    3. Hahn B,
    4. Shaw G,
    5. Oldstone M
    . Virus-specific CD8+ cytotoxic T-lymphocyte activity associated with control of viremia in primary human immunodeficiency virus type 1 infection. J Virol 1994;68:6103-10.
    Abstract/FREE Full Text
  10. 10.↵
    1. Mellors JW,
    2. Kingsley LA,
    3. Rinaldo CR Jr.,
    4. et al
    . Quantitation of HIV-1 RNA in plasma predicts outcome after seroconversion. Ann Intern Med 1995;122:573-9.
    Abstract/FREE Full Text
  11. 11.
    1. O'Brien TR,
    2. Blattner WA,
    3. Waters D,
    4. et al
    . Serum HIV-1 RNA levels and time to development of AIDS in the multicenter hemophilia cohort study. JAMA 1996;276:105-10.
    Abstract/FREE Full Text
  12. 12.↵
    1. Ho DD
    . Viral counts count in HIV infection. Science 1996;272:1124-5.
    CrossRefMedlineWeb of Science
  13. 13.↵
    1. Schacker T,
    2. Hughes JP,
    3. Shea T,
    4. Coombs RW,
    5. Corey L
    . Biologic and virologic characteristics of primary HIV infection. Ann Intern Med 1998;128:613-20.
    Abstract/FREE Full Text
  14. 14.↵
    1. Wei X,
    2. Ghosh SK,
    3. Taylor ME,
    4. et al
    . Viral dynamics in human immunodeficiency virus type 1 infection. Nature 1995;373:117-22.
    CrossRefMedline
  15. 15.↵
    1. Ho DD,
    2. Neumann AU,
    3. Perelson AS,
    4. Chen W,
    5. Leonard JM,
    6. Markowitz M
    . Rapid turnover of plasma virions and CD4 lymphocytes in HIV-1 infection. Nature 1995;373:123-6.
    CrossRefMedline
  16. 16.↵
    1. Mellors JW,
    2. Rinaldo CR Jr.,
    3. Gupta P,
    4. White RM,
    5. Todd JA,
    6. Kingsley LA
    . Prognosis in HIV-1 infection predicted by the quantity of virus in plasma. Science 1996;272:1167-70.
    Abstract
  17. 17.↵
    1. Perelson AS,
    2. Essunger P,
    3. Cao Y,
    4. et al
    . Decay characteristics of HIV-1—infected compartments during combination therapy. Nature 1997;387:188-91.
    CrossRefMedline
  18. 18.↵
    1. Chun TW,
    2. Carruth L,
    3. Finzi D,
    4. et al
    . Quantification of latent tissue reservoirs and total body viral load in HIV-1 infection. Nature 1997;387:183-8.
    CrossRefMedline
  19. 19.↵
    1. Danner SA,
    2. Carr A,
    3. Leonard JM,
    4. et al
    . A short-term study of the safety, pharmacokinetics, and efficacy of ritonavir, an inhibitor of HIV-1 protease. N Engl J Med 1995;333:1528-33.
    CrossRefMedlineWeb of Science
  20. 20.↵
    1. Markowitz M,
    2. Saag M,
    3. Powderly WG,
    4. et al
    . A preliminary study of ritonavir, an inhibitor of HIV-1 protease, to treat HIV-1 infection. N Engl J Med 1995;333:1534-9.
    CrossRefMedlineWeb of Science
  21. 21.↵
    1. Eron JJ,
    2. Benoit SL,
    3. Jemsek J,
    4. et al
    . Treatment with lamivudine, zidovudine, or both in HIV-positive patients with 200 to 500 CD4+ cells per cubic millimeter. N Engl J Med 1995;333:1662-9.
    CrossRefMedlineWeb of Science
  22. 22.↵
    1. Gulick R,
    2. Mellors J,
    3. Havlir D,
    4. et al
    . Treatment with indinavir, zidovudine, and lamivudine in adults with HIV infection and prior antiretroviral therapy. N Engl J Med 1997;337:734-9.
    CrossRefMedlineWeb of Science
  23. 23.↵
    1. Hammer S,
    2. Squires K,
    3. Hughes M,
    4. et al
    . A controlled trial of two nucleoside analogues plus indinavir in persons with HIV infection and CD4 cell counts of 200 per cubic millimeter or less. N Engl J Med 1997;337:725-33.
    CrossRefMedlineWeb of Science
  24. 24.↵
    1. Carpenter CC,
    2. Hammer SM,
    3. Hirsch MS,
    4. et al
    . Antiretroviral therapy for HIV infection in 1997. Updated recommendations of the International AIDS Society-USA panel. JAMA 1997;277:1962-9.
    Abstract/FREE Full Text
  25. 25.↵
    1. Cao Y,
    2. Ho DD,
    3. Todd J,
    4. et al
    . Clinical evaluation of branched DNA signal amplification for quantifying HIV type 1 in human plasma. AIDS Res Hum Retroviruses 1995;11:353-61.
    MedlineWeb of Science
  26. 26.↵
    1. Mulder J,
    2. McKinney N,
    3. Christopherson C,
    4. Sninsky J,
    5. Greenfield L,
    6. Kwok S
    . Rapid and simple PCR assay for quantification of HIV-1 RNA in plasma: application to acute retroviral infection. J Clin Microbiol 1994;32:292-300.
    Abstract/FREE Full Text
  27. 27.↵
    1. Ho DD,
    2. Moudgil T,
    3. Alam M
    . Quantitation of human immunodeficiency virus type 1 in the blood of infected persons. N Engl J Med 1989;321:1621-5.
    MedlineWeb of Science
  28. 28.↵
    1. Wong JK,
    2. Hezerah M,
    3. Gunthard H,
    4. et al
    . Recovery of replication-competent HIV despite prolonged suppression of plasma viremia. Science 1997;278:1291-5.
    Abstract/FREE Full Text
  29. 29.
    1. Finzi D,
    2. Hermankova M,
    3. Pierson T,
    4. et al
    . Identification of a reservoir for HIV-1-infected patients on highly active antiretroviral therapy. Science 1997;278:1295-300.
    Abstract/FREE Full Text
  30. 30.↵
    1. Chun TW,
    2. Stuyver L,
    3. Mizell SB,
    4. et al
    . Presence of an inducible HIV-1 latent reservoir during highly active antiretroviral therapy. Proc Natl Acad Sci USA 1997;94:13193-7.
    Abstract/FREE Full Text
  31. 31.↵
    1. Saksela K,
    2. Stevens CE,
    3. Rubinstein P,
    4. Taylor PE,
    5. Baltimore D
    . HIV-1 messenger RNA in peripheral blood mononuclear cells as an early marker of risk for progression to AIDS. Ann Intern Med 1995;123:641-8.
    Abstract/FREE Full Text
  32. 32.↵
    1. Vesanen M,
    2. Markowitz M,
    3. Cao Y,
    4. Ho DD,
    5. Saksela K
    . HIV-1 mRNA splicing pattern in infected persons is determined by the proportion of newly infected cells. Virology 1997;236:104-9.
    CrossRefMedlineWeb of Science
  33. 33.↵
    1. Fox C,
    2. Tenner-Racz K,
    3. Racz P,
    4. Firpo A,
    5. Pizzo P,
    6. Fauci A
    . Lymphoid germinal centers are reservoirs of HIV-1 RNA. J Infect Dis 1991;164:1051-7.
    Abstract/FREE Full Text
  34. 34.↵
    1. Embretson J,
    2. Zupancic M,
    3. Beneke J,
    4. et al
    . Analysis of human immunodeficiency virus-infected tissues by amplification and in situ hybridization reveals latent and permissive infections at single-cell resolution. Proc Natl Acad Sci USA 1993;90:357-61.
    Abstract/FREE Full Text
  35. 35.↵
    1. Binley JM,
    2. Klasse PJ,
    3. Cao Y,
    4. et al
    . Differential regulation of the antibody responses to gag and env proteins of human immunodeficiency virus type 1. J Virol 1997;71:2799-809.
    Abstract/FREE Full Text
  36. 36.↵
    1. Markowitz M,
    2. Cao Y,
    3. Vesanen M,
    4. et al
    . 4th Conference on Retroviruses and Opportunistic Infections: program and abstracts (Washington, DC). Alexandria, VA: Infectious Diseases Society of America; 1997. Recent HIV infection treated with AZT, 3TC, and a potent protease inhibitor; p. LB8.
  37. 37.↵
    1. Markowitz M,
    2. Talal A,
    3. Vesanen M,
    4. et al
    . Durable suppression of virus replication with saquinavir in combination with ritonavir and double nucleoside inhibitor therapy. Program and abstracts: XII World AIDS Conference (Geneva). 1998.
  38. 38.↵
    1. Vesanen M,
    2. Cao Y,
    3. Hurley A,
    4. Schluger R,
    5. Ho DD,
    6. Markowitz M
    . HIV-1 proviral DNA decay rate in patients treated with potent antiretroviral regimens. Program and abstracts: International Workshop on HIV Drug Resistance, Treatment Strategies, and Eradication (St. Petersburg, Florida). 1997.
  39. 39.↵
    1. Morris L,
    2. Binley J,
    3. Clas B,
    4. et al
    . HIV-1 antigen-specific and-nonspecific B cell responses are sensitive to combination antiretroviral therapy. J Exp Med 1998;188:233-45.
    Abstract/FREE Full Text
  40. 40.↵
    1. Ogg G,
    2. Jin X,
    3. Bonhoeffer S,
    4. et al
    . Quantitation of HIV-specific cytotoxic T lymphocytes and plasma load of viral RNA. Science 1998;279:2103-6.
    Abstract/FREE Full Text
  41. 41.↵
    1. Musey L,
    2. Hughes J,
    3. Schacker T,
    4. Shea T,
    5. Corey L,
    6. McElrath MJ
    . Cytotoxic T-cell responses, viral load, and disease progression in early HIV-1 infection. N Engl J Med 1997;337:1267-74.
    CrossRefMedlineWeb of Science
  42. 42.↵
    1. Rosenberg E,
    2. Billingsley JM,
    3. Caliendo AM,
    4. et al
    . Vigorous HIV-1—specific CD4+ T cell responses associated with control of viremia. Science 1997;278:1447-50.
    Abstract/FREE Full Text
  43. 43.↵
    1. Burton DR,
    2. Moore JP
    . Why do we not have an HIV vaccine and how can we make one? Nat Med 1998;4 suppl 5:495-488.
    CrossRefMedlineWeb of Science
  44. 44.↵
    1. Ho DD
    . Toward HIV eradication or remission: the tasks ahead. Science 1998;280:1866-7.
    Abstract/FREE Full Text
| Table of Contents

This Article

  1. J Infect Dis. (1999) 179 (3): 527-537. doi: 10.1086/314628
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free
  4. A correction has been published

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 205 (5)
  1. Journal of Infectious Diseases
  1. Alert me to new issues

Most