Skip Navigation

Sustained CD4+ T Cell Response after Virologic Failure of Protease Inhibitor-Based Regimens in Patients with Human Immunodeficiency Virus Infection

  1. Steven G. Deeks1,
  2. Jason D. Barbour1,2,
  3. Jeffrey N. Martin1,
  4. Melinda S. Swanson1 and
  5. Robert M. Grant1,2
  1. 1University of California, San Francisco, and San Francisco General Hospital, San Francisco, California
  2. 2Gladstone Institute of Virology and Immunology, San Francisco, California
  1. Reprints or correspondence: Dr. Steven G. Deeks, UCSF AIDS Program, 995 Potrero Ave., San Francisco General Hospital, San Francisco, CA 94110 (sdeeks{at}php.ucsf.edu).

  1. Presented in part: 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Illinois, February 1999 (abstract 494) and at the 7th Conference on Retroviruses and Opportunistic Infections, San Francisco, California, February 2000 (abstract 236).

 
Next Section

Abstract

The relationship between plasma human immunodeficiency virus (HIV) RNA levels and peripheral CD4+ T cell counts was examined in 380 HIV-infected adults receiving long-term protease inhibitor therapy. Patients experiencing virologic failure (persistent HIV RNA >500 copies RNA/mL) generally had CD4+ T cell counts that remained greater than pretherapy baseline levels, at least through 96 weeks of follow-up. The CD4+ T cell response was directly and independently related to degree of viral suppression below the pretreatment baseline. For any given HIV RNA level measured 12 weeks after virologie failure, subsequent CD4+ T cell decline was slower in patients receiving a protease inhibitor-based regimen than in a historical control group of untreated patients. These observations suggest that transient or partial declines in plasma HIV RNA levels can have sustained effects on CD4+ T cell levels.

The goal of antiretroviral therapy for human immunodeficiency virus (HIV) infection is to lengthen and improve the quality of the patient's life. According to current clinical guidelines, the optimal way to achieve this goal is by use of combination therapy to reduce HIV RNA levels below the level of detection for as long as possible [1, 2]. In prospective clinical trials, combination therapy containing a protease inhibitor is associated with durable viral suppression, increased number of absolute CD4+ T cells, reconstitution of lost immune function, and decreased mortality [3–6]. Unfortunately, a significant number of patients who initiate combination therapy involving a protease inhibitor eventually experience virologie failure (generally defined as a persistent HIV RNA >500 copies RNA/mL) [5–8]. The long-term immunologie consequences of virologie failure with a protease inhibitor-based regimen have not been well described, partly because patients enrolled in prospective clinical trials tend to withdraw from study after experiencing virologie failure. To address this question, we examined the long-term CD4+ T cell response in patients who experienced a wide variety of virologie responses to protease inhibitor therapy.

Previous SectionNext Section

Methods

Subjects

All patients in this study received longitudinal care through the University of California, San Francisco AIDS program located at San Francisco General Hospital (SFGH). Patients were identified through an ongoing cohort study evaluating the effectiveness of protease inhibitor therapy in the primary care setting [7] or through an administrative database of participants enrolled in clinical trials. Inclusion criteria included the following: (1) >16 weeks of continuous therapy with a potent protease inhibitor-based regimen (ritonavir, indinavir, nelfinavir, and/or saquinavir soft-gel capsule); (2) an assessable HIV RNA and CD4+ T cell determination prior to the initiation of the protease inhibitor; and (3) initiation of protease inhibitor therapy before March 1997 (to allow at least 96 weeks of observation). Patients were excluded if they initiated therapy during primary infection or if they received interleukin-2 or hydroxyurea at the time therapy was initiated.

We used the San Francisco Men's Health Study (SFMHS) [9] as an untreated comparison group. In brief, the SFMHS was a population-based prospective cohort study of the natural history of HIV/AIDS that was conducted from 1984 to 1994, an era prior to highly active antiretroviral treatment. In 1984, at baseline, 1034 single men between the ages of 25 and 54 years were recruited from the areas of highest AIDS incidence in San Francisco. Four hundred men (all homosexual) were HIV-infected at baseline and were followed up at 6-month intervals.

Measurements

For patients who received primary care at SFGH, plasma HIV RNA quantification was performed by use of a branched DNA (bDNA) assay (Chiron Corp, Emeryville, CA). Before 1 July 1996, version 1.0 was used (lower limit of quantification 10,000 copies/mL). After 1 July 1996 and 1 August 1998, versions 2.0 (lower limit of quantification 500 copies/mL) and 3.0 (lower limit of quantification 50 copies/mL), respectively, were used. For patients enrolled in a clinical trial, HIV RNA determinations were generally performed by use of a reverse transcriptase (RT) polymerase chain reaction (PCR) assay (Amplicor HIV Monitor Test, Roche Diagnostic Systems, Branchburg, NJ; limit of quantification 200 copies RNA/mL). In the SFMHS, HIV RNA testing was also performed by use of the Amplicor assay. Results obtained with PCR-based assays were converted to bDNA 2.0 equivalents by use of the following equation: bDNA (version 2.0) = 0.2 × (RT-PCR)1.1 [10]. Results obtained with bDNA version 3.0 were multiplied by 0.57 to obtain version 2.0 equivalents (R. Kokka, personal communication). All HIV RNA values <500 copies RNA/mL were assigned a value of 200 copies RNA/mL.

Plasma HIV RNA and CD4+ T cell determinations closest to each of the following time points were used in the analysis: baseline (most recent value prior to baseline), week 6 (± 3 weeks), week 12 (± 3 weeks), week 24 (± 6 weeks), week 36 (± 6 weeks), week 48 (± 6 weeks), week 72 (± 8 weeks), and week 96 (± 10 weeks). Data were considered missing if not available within the time range listed. We categorized patients into 4 groups on the basis of their HIV RNA response over 96 weeks of follow-up: complete virologic responders (HIV RNA consistently <500 copies/mL), partial virologic responders (HIV RNA >500 copies RNA/mL but at least 1.0 log copies/mL below pretherapy baseline), transient virologie responders (an initial HIV RNA decrease of >1.0 log copies/mL followed by a rebound to within 1.0 log copies/mL of baseline), and virologie nonresponders (no HIV RNA decrease >1.0 log copies/mL below baseline).

Statistical analysis

All patients who met our eligibility criteria were included in the final analysis, even if they subsequently modified or discontinued therapy. For patients who discontinued protease inhibitor therapy for ⩾16 weeks, we censored data at the time therapy was discontinued. Similarly, for patients who experienced virologie failure, and later switched to a successful salvage regimen (defined as achieving an HIV RNA level <500 copies/mL for ∼⩾16 weeks), we censored data at the time the successful salvage regimen was initiated.

The relationship between mean change in HIV RNA and mean absolute HIV RNA (both expressed as log10 copies RNA/mL) with mean change in CD4+ T cell count (the dependent variable) was assessed by means of bivariable and multivariable linear regression. We determined mean change in HIV RNA and CD4+ T cell values by comparing each time point with the pretherapy baseline. All data from week 6 to week 96 were used. Both independent variables were entered into a multivariable model, regardless of significance in bivariable regression. The relative contribution of each variable was determined by use of partial r2 values.

We determined the CD4 rate of change during periods of no antiretroviral treatment (SFMHS patients), viral suppression (SFGH complete responders), and virologie failure (SFGH partial responders, transient responders, and nonresponders). For the SFGH patients experiencing a complete virologie response, CD4+ T cell values from week 12 to week 72 of therapy were evaluated. We chose this period to exclude any earlier CD4+ T cell rises, as these may be due to redistribution [11]. For SFGH patients experiencing any type of virologie failure (partial responders, transient responders, nonresponders), CD4 values from 6 to 72 weeks after virologie failure were analyzed. Virologie failure was assumed to have occurred after the patient's last date of a successful response; this was generally the last date a patient had an HIV RNA level below the level of detection (<500 copies/mL) or the date of the patient's HIV RNA nadir (for patients who did not achieve an undetectable HIV RNA level). For the untreated SFMHS cohort, all available CD4 data were used. Once individual CD4+ cell rates were determined, patients were stratified on the basis of their absolute HIV RNA levels. Patients experiencing complete viral suppression were assumed to have an HIV RNA <500 copies RNA/mL. Patients experiencing virologie failure were stratified on the basis of their HIV RNA level 12 weeks after virologie failure. Untreated patients were stratified on the basis of their HIV RNA level when they entered the SFMHS.

CD4+ cell counts were log10 transformed before calculation of the rate of CD4+ T cell change. This was done because the proportionate rate of CD4+ T cell decline is dependent on the baseline CD4+ T cell count when expressed on an absolute scale, but not after logarithmic transformation [12]. Least-squares linear regression for each individual was used to estimate CD4 rate of change.

Nonparametric methods were used throughout (except in the regression analysis). All P values are 2-sided. All statistical analyses were performed by use of SAS version 6.12 (SAS Institute, Cary, NC) or SPSS for Windows 8.0.

Previous SectionNext Section

Results

Study population

A total of 380 patients met eligibility criteria. The majority of patients were middle-aged men (table 1). The median plasma HIV RNA level before initiation of protease inhibitor therapy was 4.55 log10 copies RNA/mL (interquartile range [IQR] 4.06–5.04) and the median CD4+ T cell count was 143 cells/mm3 (IQR 50–273). Only 17.1% of the patient population was treatment naive at the time a potent protease inhibitor was initiated. Indinavir was the most common initial protease inhibitor used; many patients added a protease inhibitor to a stable nucleoside analogue regimen.

Figure 1
View larger version:
Figure 1

Change in human immunodeficiency virus (HIV) RNA and CD4+ T cell count over 96 weeks in patients meeting a definition of complete virologic responders, partial responders, transient responders, and nonresponders. Figure shows median CD4 changes (A) and median HIV RNA changes (B). The number of patients evaluated at each time point is shown.

Figure 2
View larger version:
Figure 2

Change in CD4 after virologic failure in patients who had either a transient virologic response (initial decrease in human immunodeficiency virus [HIV] RNA of at least 1 log with return toward baseline) or partial virologie response (durable decrease in HIV RNA of at least 1 log below baseline). Median CD4 changes (open diamonds) and median HIV RNA levels (filled diamonds) are shown. Day 0 was defined as the last date that the patient had had a successful response to therapy.

Figure 3
View larger version:
Figure 3

Change in CD4 (expressed as log10 cells/μL/year) in patients receiving protease inhibitor therapy (●) and in patients not receiving antiretroviral therapy (■). Treated patients experiencing a complete virologie response were assumed to have a human immunodeficiency virus (HIV) RNA level of <500 copies RNA/mL. Treated patients experiencing virologie failure were stratified on the basis of their HIV RNA level 12 weeks after the last date the patient had had a successful response (see text). Untreated patients (SFMHS) were stratified on the basis of their HIV RNA at the initial baseline visit. CD4 depletion rates for the patients experiencing virologie failure were calculated by use of CD4 values from week 6 to week 72 after the prefailure date. CD4 depletion rates for patients experiencing a complete virologie response (HIV RNA <500) were calculated on the basis of CD4 values from week 12 to week 72 of follow-up. CD4 depletion rates for the untreated patients were calculated by use of data at 6-month intervals from baseline to 24 months. Only patients who had at least 2 assessable data points were included in the analysis. P values were determined by means of the Wilcoxon rank sum test (treated vs. untreated patients). SFGH, San Francisco General Hospital; SFMHS, San Francisco Men's Health Study.

Table 1
View larger version:
Table 1

Baseline demographic and clinical characteristics according to virologic response through 96 weeks.

Of the 380 patients included in this analysis, 30 patients were lost to follow-up before week 96 and 18 patients died before week 96. An additional 19 patients were followed up through week 96 but did not have an HIV RNA level and CD4+ T cell count at or near week 96. Twenty-one patients were censored at the time successful salvage therapy was initiated, and 53 patients were censored at the time protease inhibitor therapy was discontinued for 16 weeks or more. Subjects who died or were lost to follow-up before week 96 did not differ from the entire cohort with respect to age, baseline HIV RNA, baseline CD4+ T cell count, or virologie response (responders vs. nonresponders).

HIV RNA and CD4+ T cell change after initiation of a protease inhibitor-based regimen

The relationship between the HIV RNA and CD4+ T cell response over time is illustrated in figure 1. Patients were classified on the basis of the virologie response as complete virologie responders (n = 171), partial virologie responders (n = 74), transient virologie responders (n = 86), and virologie nonresponders (n = 49). At week 48, the median change in CD4+ T cell count from baseline for each group was as follows: complete virologie responders, 129 cells/mm3 (IQR 50–219); partial virologie responders, 130 cells/mm3 (IQR 69–221); transient virologie responders, 94 cells/mm3 (IQR 32–142); and virologie nonresponders, 36 cells/mm3 (IQR 4–103). At week 96, the median change for the complete responders was 190 cells/mm3 (IQR 82–306); for the partial responders, 151 cells/mm3 (IQR 109–234); and for the transient virologie responders, 87 cells/mm3 (IQR 30–144) (there were insufficient numbers of virologie nonresponders at week 96). The change in the CD4+ T cell count from baseline in the complete versus partial virologie responders did not differ significantly through week 96 (P > .10 at most time points through week 96; Wilcoxon rank sum); the difference approached significance at week 36 (P = .06). In contrast, the difference in the change in CD4+ T cell levels from baseline for the complete versus transient virologie responders achieved statistical significance at each point after week 48 (P < .05). In pairwise comparisons, the CD4+ T cell change was lower in the virologie nonresponders than in each of the other groups at each time point after week 12 (P < .05).

To further define the relationship between the virologie and the CD4 response, we calculated a mean change in HIV RNA (log10 copies RNA/mL) from baseline and a mean change in CD4+ T cell count (cells/mm3) from baseline for each patient (using all available data points from week 6 to week 96). Across all patients, there was a weak but highly significant correlation between the mean change HIV RNA and the mean change CD4+ T cell count (Spearman rank correlation coefficient, ρ = −0.35, P = .0001). The correlation was stronger in patients experiencing virologie failure (ρ = −0.37, P = .0001) than in patients experiencing a complete virologie response (ρ = −0.26, P = .005). We then stratified patients on the basis of their mean log10 reduction in HIV RNA over 96 weeks of follow-up. As shown in table 2, there was a consistently greater CD4+ T cell increase for each additional mean log10 reduction in HIV RNA (P < .0001, Kruskal-Wallis).

Table 2
View larger version:
Table 2

Mean change in human immunodeficiency virus (HIV) RNA and mean change in CD4+ T cell counts calculated for each patient (by use of all data from weeks 6–96) and stratified on the basis of the mean change in HIV RNA.

For patients receiving antiretroviral therapy, the CD4+ T cell response may be determined by the change in HIV RNA levels relative to pretreatment baseline, the absolute HIV RNA level achieved, or both. To examine this issue, we analyzed the mean change in HIV RNA relative to baseline (log10 copies RNA/mL), the mean absolute HIV RNA (log10 copies RNA/mL) and the mean CD4+ T cell change relative to baseline (using all available data points between week 6 and week 96). In a bivariable analysis, both the mean change in HIV RNA and the mean absolute HIV RNA were associated with the mean change in CD4+ T cell count (r2 = 0.13, P = .0001 and r2 = 0.08, P = .0001, respectively). In a multivariable regression analysis, only mean change in HIV RNA was significantly associated with mean change in CD4 (partial r2 = 0.13, P = .0001).

Effect of virologie failure on the subsequent CD4+ T cell response

To directly illustrate the effect of virologie failure on short-term and long-term CD4 T cell changes, we did further analysis of those patients who had evidence of an initial virologie response and who subsequently experienced virologie failure (transient and partial responders). As shown in figure 2, virologie rebound was not temporally associated with a significant change in the absolute CD4+ T cell level. After 48 weeks of documented virologie failure, the absolute CD4+ T cell count increased relative to the prefailure baseline (median increase of 34 cells/mm3, P = .002, signed ranked test). Notably, after virologie failure, plasma HIV RNA levels remained below the pretherapy baseline (see figure 2).

To more directly analyze the effect of virologie failure on CD4+ T cell changes, we compared change in HIV RNA from pretherapy baseline 24 weeks after virologie failure with the change in CD4+ T cell during the first 24 weeks of virologie failure. There was a significant but weak correlation between the degree of viral suppression below pretherapy baseline and the change in CD4+ T cell during 24 weeks of virologic failure (Spearman rank correlation coefficient, ρ=−0.23, P = .01, n = 120). In other words, the greater the degree of persistent partial viral suppression, the greater the continued increase in CD4+ T cell during virologie failure. This positive CD4+ T cell response diminishes as the plasma HIV RNA level returns to baseline.

Relationship between absolute HIV RNA and subsequent CD4+ T cell response in treated and untreated patients

Figure 3 shows the relationship between the rate of CD4+ T cell change (log10 transformed) and the absolute HIV RNA level among untreated and treated patients. As shown in figure 3, the higher the absolute HIV RNA level 12 weeks after virologie failure, the more rapid the subsequent rate of CD4+ T cell depletion; this became particularly evident as the HIV RNA level increased above 30,000 copies/mL. Nonetheless, among patients who failed to maintain undetectable plasma HIV RNA levels but remained below 30,000 copies/mL, there was no average decline in CD4+ T cell counts.

We next compared the rate of CD4+ T cell depletion in patients experiencing virologie failure on a protease inhibitor-based regimen with a group of untreated patients (SFMHS). Baseline CD4+ T cell counts were significantly lower in the SFGH-treated patient population (median, 143 cells/mm3) than in the SFMHS group (median, 648 cells/mm3; P = .0001, Wilcoxon rank sum). Within each HIV RNA stratum, there was a greater decline in CD4+ cell counts in patients who were not receiving antiretroviral therapy than in patients who were experiencing virologie failure with a protease inhibitor-based regimen. These differences reached statistical significance in most HIV RNA strata (figure 3).

Previous SectionNext Section

Discussion

Virologie failure of combination therapy is common in clinical practice [6–8]. Furthermore, patients failing an initial protease inhibitor-based regimen typically have a limited virologie response to salvage therapy, even when novel compounds are used in the salvage regimen [7, 13]. Recognizing that durable viral suppression will not occur in many patients, it is important to determine the long-term immunologie implications of virologie failure. In this study, we report that there is a significant relationship between the virologie response and the CD4+ T cell response. Patients who achieve durable viral suppression (persistent HIV RNA <500 copies RNA/mL) show a continued rise in the absolute CD4+ T cell count whereas patients who experience virologie failure have a blunted CD4+ T cell response. Furthermore, among all patients there was a clear relationship between the time-weighted changes in HIV RNA and the time weighted changes in CD4+ T cell counts. These observations are consistent with others [10, 14–16], and underscore the importance of viral replication as a critical determinant of CD4+ T cell levels.

Although the degree of viral suppression was associated with the long-term CD4+ T cell response, CD4+ T cell counts remained elevated above the pretherapy baseline in the majority of patients experiencing virologie failure and continued to increase during the first 48 weeks of documented virologie failure. This phenomenon, in which CD4+ T cell counts remain elevated after virologie failure, has been observed by others [14, 17, 18], and has been referred to as “discordant” or “paradoxical” CD4+ T cell responses. Our data suggest that the apparent discordance derives partly from the definition of virologie failure as any rebound of viremia into the detectable range without regard to the level of virologie rebound. Patients experiencing virologie failure in this cohort had, on average, persistent viral suppression below pretreatment levels. Presumably this partial viral suppression contributed to the observed maintenance of CD4+ T cell increases (see figure 2). Similar sustained increases in CD4+ T cell counts was seen over 52 weeks in patients receiving partial viral suppression on dual nucleoside analogue regimens [19]. By avoiding simple dichotomies of virologie and CD4 responses into “failure” and “success,” the concordance of typical partial responses can be appreciated.

In an analysis of >1600 patients followed in the Multicenter AIDS Cohort Study (MACS), there was a significant relationship between the absolute HIV RNA level at baseline and the subsequent rate of CD4+ T cell decline [10]. We observed a similar relationship in a smaller group of untreated patients (the SFMHS) and in a group of patients experiencing virologie failure with a protease inhibitor-based regimen. However, when treated and untreated patients were compared, the curve appears to be shifted up for treated patients (figure 3); that is, for a given HIV RNA level, treated patients experiencing virologie failure with a protease inhibitor-based regimen had lower CD4+ T cell depletion rates than patients who did not receive antiretroviral therapy. These observations suggest that the natural history of HIV mediated CD4+ T cell depletion may be altered by the continued use of protease inhibitor-containing regimens.

There are several differences between the treated SFGH and the untreated SFMHS patients that may contribute to the different rates of CD4+ T cell depletion observed within any viral load stratum. First, and perhaps most importantly, SFGH patients were treated and had continued partial viral suppression below baseline, whereas patients followed in the SFMHS were not treated and presumably had increasing HIV RNA levels over time. Since the relationship between HIV RNA and CD4 levels is presumably a dynamic one, higher rates of CD4+ T cell depletion would be expected in patients whose HIV RNA levels are increasing [20]. As we did not have longitudinal HIV RNA data in the untreated SFMHS subjects, we were unable to directly compare the effect of HIV RNA slope on the rate of CD4+ T cell depletion. Second, treated patients at SFGH generally had more advanced disease than the untreated patients in the SFMHS. However, in our study, and others [12], baseline CD4+ T cell count did not correlate with the subsequent log 10 CD4 T cell rate of change, suggesting that the difference in disease stage between the 2 cohorts does not fully explain the difference in the rates of CD4+ T cell depletion. Third, the SFMHS patients were studied between 1984 and 1986, 12 years earlier than the SFGH patients. Although decreases in HIV pathogenicity over the course of the epidemic have not been described, an evolution toward less virulence is theoretically feasible [21].

Several mechanisms may underlie the preservation of CD4+ T cell numbers after virologic failure for patients receiving combination antiretroviral therapy. For example, small reductions in HIV RNA levels on a log10 scale signify large reductions on an absolute scale. Partial viral suppression may tip the balance between viral cytopathicity and cell regeneration in favor of T cell homeostasis; that is, a viral threshold may exist below which CD4 T cell production can exceed HIV-1 mediated CD4+ T cell depletion [22, 23]. Alternatively, drug-resistant virus may have reduced replicative capacity or altered host tissue range. Viruses containing protease inhibitor related mutations compete poorly against wild-type viruses in competitive cultures [24], have reduced replicative capacity in single cycle infectivity assays [25], and are less able to infect thymic tissue [26]. Similar reductions in virus fitness have been observed with in association with resistance to nucleoside reverse transcriptase inhibitors [27, 28]. Theoretically, decreased replication capacity in some or all tissues would result in reduced CD4+ T cell death or increased CD4+ T cell production. Consequently, more CD4+ T cells become available to support increased viral replication (“predator/prey” phenomenon). A new steady state is therefore achieved where, in the presence of a less fit virus, CD4+ T cell increases are maintained whereas HIV RNA levels return toward pretherapy baseline [29]. Finally, protease inhibitors may have direct effects on T cell function and survival, independent of their effect on viral replication. Preliminary data suggest that protease inhibitors may prevent CD4+ T cell apoptosis [30] and/or exert beneficial immunomodulatory effects [31]. Although there are no convincing evidence that the sustained CD4+ T cell response after virologic failure is a protease inhibitor-specific phenomenon, one recently published study suggests that there may be a class-specific effect. In Merck 039, indinavir monotherapy was compared with zidovudine/lamivudine among patients with advanced disease. Despite similar virologie responses, subjects receiving indinavir had more sustained increase in peripheral CD4+ T cell counts (through 24 weeks of follow-up) [32].

Although the clinical implications of our observations remain to be determined, it seems likely that a sustained increase in CD4+ T cell counts after virologie failure will translate into a sustained clinical benefit. In the Swiss HIV Cohort Study (a large prospective observational study), patients who had an initial virologie response and later experienced a rebound (transient responders in our study) had clinical progression rates that were similar to that seen in patients who maintained an undetectable viral load (after controlling for baseline CD4 and age) [6]. In a second observational study (EuroSIDA), clinical progression was rare in protease inhibitor-treated patients whose number of CD4+ T cells increased to >200 cells/mm3 [33]. Taken as a whole, our data and that of the European cohort studies suggest that there may be a delay between virologie rebound and disease progression. These observations also raise concerns about the emphasis on HIV RNA levels as a surrogate marker in clinical trials [1, 2]. For heavily pretreated patients with limited therapeutic options, antiretroviral strategies that result in sustained CD4+ T cell increases independent of continued viral suppression successfully prevent clinical disease and death.

The management of virologie failure with a protease inhibitor-based regimen is unclear, particularly for patients who have limited therapeutic options [1, 2]. Continuing a “failing” regimen in the presence of ongoing viral replication may result in continued viral evolution and the further selection of resistance mutations. Over time, high-level cross-resistance to other drugs within the same therapeutic class may occur, thus limiting future options. If complete viral suppression is no longer achievable, then stopping therapy until more effective antiretroviral agents are available may be considered. However, if the sustained CD4+ T cell response after virologie failure is due to partial viral suppression, altered viral fitness, or direct drug effects, then discontinuing therapy should result in an accelerated loss of CD4+ T cells. Notably, dramatic decreases in CD4+ T cell counts after therapy was discontinued have been reported in other observational studies [17]. Whether to modify or discontinue antiretroviral therapy in patients with limited therapeutic options is a common dilemma that cannot be fully addressed with our observations.

Our study has several limitations. First, some patients died or were lost to follow-up prior to week 96; since long-term data are not available for these patients, our analysis may be biased toward those patients who were doing well. To address this concern, we compared patients who died or were lost to follow-up prior to week 96 with the entire cohort and found no significant differences in their virologie response or in baseline characteristics (age, baseline CD4+ T cell count, baseline HIV RNA levels, and duration of nucleoside analogue therapy prior to the initiation of a protease inhibitor). Second, during the 96 weeks of follow-up, many patients who experienced virologie failure modified their antiretroviral therapy. Transient periods of viral suppression may have contributed to the observed CD4+ T cell responses. To truly determine the natural history of CD4+ T cell changes after virologie failure, we would need patients to remain on a stable regimen indefinitely; such studies have not been feasible. Notably, when our data were analyzed only up to the time of first switch, results were similar (data not shown).

In conclusion, CD4+ T cell counts can remain elevated in patients experiencing virologic failure, even in those patients who exhibited a clear rebound in HIV RNA levels back toward baseline. Although this increase appears to be partially explained by incomplete but persistent viral suppression, the rate of CD4+ T cell decline after virologic failure (for any given HIV RNA level) is slower than that observed in untreated patients. Additional studies are necessary to determine whether this “disconnect” between virologie failure and CD4+ T cell change is simply due to partial viral suppression or whether alternative mechanisms should be considered. Further, it is unclear whether continued therapy with protease inhibitors is necessary to maintain these benefits and whether these benefits will persist for more than 2 years of treatment.

Previous SectionNext Section

Acknowledgments

We would like to acknowledge the following for assistance in data collection: Richard Loftus, Shana Chin, George Beatty, Nina Simmonds, Susan Huang, and Peter Cohen.

Previous SectionNext Section

Footnotes

  • This study was approved by the institutional review board of the University of California, San Francisco.

  • Financial support: National Institutes of Health, University of California San Francisco-Gladstone Institute of Virology and Immunology Center for AIDS Research (P30 MH59037) and the Universitywide AIDS Research Program, University of California San Francisco AIDS Clinical Research Center (CC95-SF-123).

  • Received August 11, 1999.
  • Revision received November 17, 1999.
Previous Section
 

References

  1. 1.↵
    Department of Health and Human Services and Henry J. Kaiser Family Foundation. Guidelines for the use of antiretroviral agents in HIV infected adults and adolescents. MMWR Morb Mortal Wkly Rep 1998;47(RR-05):43-82.
  2. 2.↵
    1. Carpenter CC,
    2. Cooper DA,
    3. Fischl M,
    4. et al
    . Antiretroviral therapy in adults: updated recommendations of the International AIDS Society-USA Panel. JAMA 2000;283:381-90.
    Abstract/FREE Full Text
  3. 3.↵
    1. Gulick R,
    2. Mellors J,
    3. Havlir D,
    4. et al
    . Simultaneous vs. sequential initiation of therapy with indinavir, zidovudine and lamivudine for HIV-1 infection: 100 week follow-up. JAMA 1998;280:35-41.
    Abstract/FREE Full Text
  4. 4.
    1. Li TS,
    2. Tubiana R,
    3. Katlama C,
    4. Calvez V,
    5. Mohand HA,
    6. Autran B
    . Long lasting recovery of CD4 T-cell function and viral-load reduction after highly active antiretroviral therapy in advanced HIV-1 disease. Lancet 1998;351:1682-6.
    CrossRefMedlineWeb of Science
  5. 5.↵
    1. Hammer SM,
    2. Squires KE,
    3. Hughes MD,
    4. et al
    . A controlled trial of two nucleoside analogues plus indinavir in persons with human immunodeficiency virus infections and CD4 cell counts of 200 per cubic millimeter or less. N Engl J Med 1997;337:725-33.
    CrossRefMedlineWeb of Science
  6. 6.↵
    1. Ledergerber B,
    2. Egger M,
    3. Opravil M,
    4. et al
    . Clinical progression and virological failure on highly active antiretroviral therapy in HIV-1 patients: a prospective cohort study. Lancet 1999;353:863-8.
    CrossRefMedlineWeb of Science
  7. 7.↵
    1. Deeks SG,
    2. Hecht FM,
    3. Swanson M,
    4. Elbeik T,
    5. Loftus R,
    6. Cohen PT,
    7. Grant RM
    . Virologie outcomes with protease inhibitor therapy in an urban AIDS clinic: relationship between baseline characteristics and response to both initial and salvage therapy. AIDS 1999;13:F35-F44.
    CrossRefMedlineWeb of Science
  8. 8.↵
    1. Lucas GM,
    2. Chaisson RE,
    3. Moore RD
    . Highly active antiretroviral therapy in a large urban clinic: risk factors for virologie failure and adverse drug reactions. Ann Intern Med 1999;131:81-7.
    Abstract/FREE Full Text
  9. 9.↵
    1. Winkelstein W Jr..,
    2. Lyman DM,
    3. Padian N,
    4. et al
    . Sexual practices and risk of infection by the human immunodeficiency virus. The San Francisco Men's Health Study. JAMA 1987;257:321-5.
    Abstract/FREE Full Text
  10. 10.↵
    1. Mellors JW,
    2. Munoz A,
    3. Giorgi TV,
    4. et al
    . Plasma viral load and CD4+ lymphocytes as prognostic markers of HIV-1 infection. Ann Intern Med 1997;126:946-54.
    Abstract/FREE Full Text
  11. 11.↵
    1. Pakker NG,
    2. Notermans DW,
    3. de Boer RJ,
    4. et al
    . Biphasic kinetics of peripheral blood T cells after triple combination therapy in HIV-1 infection: a composite of redistribution and proliferation. Nat Med 1998;4:208-14.
    CrossRefMedlineWeb of Science
  12. 12.↵
    1. Hughes MD,
    2. Stein DS,
    3. Gundacker HM,
    4. Valentine FT,
    5. Phair JP,
    6. Volberding PA
    . Within-subject variability in CD4 lymphocyte count in asymptomatic human immunodeficiency virus infection: implications for patient monitoring. J Infect Dis 1994;169:28-36.
    Abstract/FREE Full Text
  13. 13.↵
    1. Gulick RM,
    2. Hu XJ,
    3. Fiscus SA,
    4. et al
    . Program and abstracts: 7th Conference on Retroviruses and Opportunistic Infections (San Francisco). Alexandria, VA: Foundation for Retrovirology and Human Health; 2000. Salvage therapy with saquinavir SGC (SQV) in combination with ritonavir (RTV) or nelfinavir (NFV) plus delavirdine (DLV), adefovir dipivoxil (ADV) or both: ACTG 359 [abstract 235].
  14. 14.↵
    1. Renaud M,
    2. Katlama C,
    3. Mallet A,
    4. et al
    . Determinants of paradoxical CD4 cell reconstitution after protease inhibitor-containing antiretroviral regimen. AIDS 1999;13:669-76.
    CrossRefMedlineWeb of Science
  15. 15.
    1. Staszewski S,
    2. Miller V,
    3. Sabin C,
    4. et al
    . Determinants of sustainable CD4 lymphocyte count increases in response to antiretroviral therapy. AIDS 1999;13:951-6.
    CrossRefMedlineWeb of Science
  16. 16.↵
    1. Marschner IC,
    2. Collier AC,
    3. Coombs RW,
    4. et al
    . Use of changes in plasma levels of human immunodeficiency virus type 1 RNA to assess the clinical benefit of antiretroviral therapy. J Infect Dis 1998;177:40-7.
    Abstract/FREE Full Text
  17. 17.↵
    1. Kaufmann D,
    2. Panteleo G,
    3. Sudre P,
    4. Telenti A
    . CD4 T cell count in HIV-1 individuals remaining viremic with highly active antiretroviral therapy (HAART): Swiss HIV Cohort Study [letter]. Lancet 1998;351:723-4.
    CrossRefMedlineWeb of Science
  18. 18.↵
    1. Mezzaroma I,
    2. Carlesimo M,
    3. Pinter E,
    4. et al
    . Long-term evaluation of T-cell subsets and T-cell function afte r HAART in advanced stage HIV-1 disease. AIDS 1999;13:1187-93.
    CrossRefMedlineWeb of Science
  19. 19.↵
    1. Eron JJ,
    2. Benoit SL,
    3. Jamsek 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
  20. 20.↵
    1. Lyles R,
    2. Munoz A,
    3. Bazmi H,
    4. et al
    . Program and abstracts: 6th Conference on Retroviruses and Opportunistic Infections (Chicago). Alexandria, VA: Foundation for Retrovirology and Human Health; 1999. Natural history of HIV-1 viremia after seroconversion in the multicenter AIDS Cohort Study [abstract 273].
  21. 21.↵
    1. Lenski RE,
    2. May RM
    . The evolution of virulence in parasites and pathogens: reconciliation between two competing hypotheses. J Theor Biol 1994;169:23-265.
    CrossRefMedlineWeb of Science
  22. 22.↵
    1. Perelson AS,
    2. Neumann AU,
    3. Markowitz M,
    4. Leonard JM,
    5. Ho DD
    . HIV-1 dynamics in vivo: virion clearance rate, infected cell life-span, and viral generation time. Science 1996;271:582-6.
    FREE Full Text
  23. 23.↵
    1. Hellerstein M,
    2. Hanley MB,
    3. Cesar D,
    4. et al
    . Directly measured kinetics of circulating T lymphocytes in normal and HIV-infected humans. Nat Med 1999;5:83-9.
    CrossRefMedlineWeb of Science
  24. 24.↵
    1. Martinez-Picado J,
    2. Savara AV,
    3. Sutton L,
    4. D'Aquila RT
    . Replicative fitness of protease inhibitor-resistant mutants of human immunodeficiency virus type 1. J Virol 1999;73:3744-52.
    Abstract/FREE Full Text
  25. 25.↵
    1. Zennou V,
    2. Mammano F,
    3. Paulous S,
    4. Mathez D,
    5. Clavel F
    . Loss of viral fitness associated with multiple gag and gag-pol processing defects in human immunodeficiency virus type 1 variants selected for resistance to protease inhibitors in vivo. J Virol 1998;72:3300-6.
    Abstract/FREE Full Text
  26. 26.↵
    1. Stoddart C,
    2. Mammano F,
    3. Moreno M,
    4. et al
    . Program and abstracts: 6th Conference on Retroviruses and Opportunistic Infections (Chicago). Alexandria, VA: Foundation for Retrovirology and Human Health; 1999. Lack of fitness of protease inhibitor-resistant HIV-1 in vivo [abstract 4].
  27. 27.↵
    1. Harrigan PR,
    2. Bloor S,
    3. Larder BA
    . Relative replicative fitness of zidovudine-resistant human immunodeficiency virus type 1 isolates in vitro. J Virol 1998;72:3773-8.
    Abstract/FREE Full Text
  28. 28.↵
    1. Kosalaraksa P,
    2. Kavlick MF,
    3. Maroun V,
    4. Le R,
    5. Mitsuya H
    . Comparative fitness of multi-dideoxynucleoside-resistant human immunodeficiency virus type 1 (HIV-1) in an In vitro competitive HIV-1 replication assay. J Virol 1999;73:5356-63.
    Abstract/FREE Full Text
  29. 29.↵
    1. Bonhoeffer S,
    2. Coffin JM,
    3. Nowak MA
    . Human immunodeficiency virus drug therapy and virus load. J Virol 1997;71:3275-78.
    Abstract/FREE Full Text
  30. 30.↵
    1. Strack PR,
    2. Frey MW,
    3. Rizzo CJ,
    4. et al
    . Apoptosis mediated by HIV protease is preceded by cleavage of Bcl-2. Proc Natl Acad Sci USA 1996;93:9571-6.
    Abstract/FREE Full Text
  31. 31.↵
    1. Andre P,
    2. Groettrup M,
    3. Klenerman P,
    4. et al
    . An inhibitor of HIV-1 protease modulates proteasome activity, antigen presentation, and T cell responses. Proc Natl Acad Sci USA 1998;95:13120-4.
    Abstract/FREE Full Text
  32. 32.↵
    1. Hirsch M,
    2. Steigbigel R,
    3. Staszewski S,
    4. et al
    . A randomized, controlled trial of indinavir, zidovudine, and lamivudine in adults with advanced human immunodeficiency virus type 1 infection and prior antiretroviral therapy. J Infect Dis 1999;180:659-65.
    Abstract/FREE Full Text
  33. 33.↵
    1. Miller V,
    2. Mocroft A,
    3. Reiss P,
    4. et al
    . Relations among CD4 lymphocyte count nadir, antiretroviral therapy and HIV-1 disease progression: results from the EuroSIDA study. Ann Intern Med 1999;130:570-77.
    Abstract/FREE Full Text
| Table of Contents

This Article

  1. J Infect Dis. (2000) 181 (3): 946-953. doi: 10.1086/315334
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

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