点击显示 收起
Department of Medicine, San Francisco General Hospital, University of California at San Francisco
Veterans Affairs Medical Center, San Francisco, Department of Medicine, University of California at San Diego and San Diego Veterans Affairs Healthcare System, La Jolla
Gilead Sciences, Inc., Foster City, California
In human immunodeficiency virus (HIV)infected patients successfully treated with highly active antiretroviral therapy (HAART), a low level of HIV RNA persists in plasma at steady state for years and varies among patients. To understand predictors of residual viremia, we measured HIV RNA levels <50 copies/mL in patients after 1 year of treatment with efavirenz and lamivudine plus either tenofovir disoproxil fumarate (n = 55) or stavudine (n = 45), by use of an HIV RNA assay with a limit of detection of 2.5 copies/mL. The mean posttreatment HIV RNA levels were 0.58 log10 copies/mL (3.8 copies/mL) in the tenofovir arm and 0.61 log10copies/mL (4.1 copies/mL) in the stavudine arm (P = .24). Forty-seven percent of patients receiving tenofovir, compared with 29% of patients receiving stavudine, had undetectable residual viremia (P = .07). In multivariate analyses, we found that lower baseline HIV RNA levels in plasma, lower HIV DNA levels in peripheral blood mononuclear cells, and inclusion in the tenofovir arm each independently predicted undetectable residual viremia (P < .05). However, a level of residual viremia <50 copies/mL was not associated with CD4 cell count changes or risk of virologic rebound through 72 weeks of follow-up.
With initiation of highly active antiretroviral therapy (HAART), HIV RNA levels in plasma can decline rapidly to <50 copies/mL [14] and can reach steady-state levels after 69 months of therapy [5]. HIV RNA can be detected in the plasma of many of these patients for years by use of sensitive assays, and productive infection appears to contribute to residual viremia [69]. Although the latently infected pool of cells has been identified as the major obstacle to HIV eradication [1013], understanding the factors that contribute to residual viremia could assist in developing long-term treatment strategies. In a recent small study of patients receiving identical treatment regimens, the pretreatment level of HIV DNA in peripheral blood mononuclear cells (PBMCs) was the strongest predictor of residual viremia [5]. In the present study, we evaluated the predictors and consequences of residual viremia in a broader context, by examining host and virus characteristics in 100 patients receiving the combination treatment regimens of either (1) efavirenz, lamivudine, and tenofovir disoproxil fumarate (hereafter, "tenofovir") or (2) efavirenz, lamivudine, and stavudine.
PATIENTS AND METHODS
Patient population.
The study population was a subset of patients participating in the Gilead 903 trial (GS-99-903) in whom sustained virologic suppression had been achieved after 6 months of therapy. Study 903 is a study with an ongoing open-label component and a completed 144-week, randomized, double-blind study of tenofovir or stavudine in combination with efavirenz and lamivudine in treatment-naive, HIV-infected patients. Details of study 903 have been reported elsewhere [14]. For the present study, sustained virologic suppression was defined as HIV RNA levels <50 copies/mL of plasma from week 24 to week 72. PBMCs were collected from a total of 266 patients at North American sites in study 903. Of these eligible patients, 144 had completed 72 weeks of therapy and had HIV RNA levels <50 copies/mL from week 24 to week 72, without any elevations above this threshold (hereafter referred to as "blips") at the time of the analysis. The first 100 patients by randomization order who had available plasma and PBMC samples were included in this analysis.
HIV RNA levels.
The baseline HIV RNA level was determined using the geometric mean value of 2 measurements taken by use of the Roche Amplicor assay (version 1.0; Roche Molecular Diagnostics). Residual viremia was determined by measuring HIV RNA levels at weeks 48, 64, and 72, by use of a modification of the Roche Amplicor assay [9]. In brief, the modifications of the assay included the pelleting of virions from 2 mL of plasma, the resuspension of extracted HIV RNA in 50 L of diluent volume, and the input of the entire volume of resuspended RNA for reverse-transcriptase (RT) polymerase chain reaction (PCR). Details on the determination of the limit of detection (LOD) and the reproducibility of this modified assay [9], as well as the independence of assay results from HIV DNA in plasma [5], have been described elsewhere. In brief, at the nominal LOD of 2.5 copies/mL, the 2.5-copy RNA assay yielded a coefficient of variation of 37% in validation assays when patient plasma of defined HIV RNA level was used. However, on the basis of Poissonian statistics, as many as 12 of 100 samples might contain 2 HIV RNA copies and as many as 26 of 100 samples might contain 3 HIV RNA copies at the LOD and be judged to be "undetectable." Therefore, the outcome variable of undetectable residual viremia was defined as all 3 measures of HIV RNA level below the nominal LOD of 2.5 copies/mL. For a sample with 2.5 copies/mL, the likelihood of having all 3 samples be negative would be 1.5%. Follow-up HIV RNA levels from week 72 through week 144 were determined by use of the Roche Ultrasensitive Amplicor Monitor assay (version 1.0) at each available study visit.
HIV DNA levels.
PBMC DNA quantitation was performed using a Roche PCR-based system with colorimetric detection. The PCR primers detect only late-stage or fully reverse-transcribed HIV DNA only. On the basis of genomic DNA input into PCRs, the lower limit for reliable quantitation was 5 HIV DNA copies/g of PBMC DNA. HIV DNA levels presented were quantified per microgram of PBMC DNA. Results based on an alternative normalization of HIV DNA levels, either by blood volume based on blood lymphocyte and monocyte counts or by calculated CD4 lymphocyte DNA levels, were also analyzed.
Statistical analyses.
All statistical analyses were performed using SAS (version 8.1; SAS Institute). Bivariate and multivariate logistic regression analyses were performed to evaluate the impact of randomized treatment assignment, baseline plasma HIV RNA level, baseline PBMC HIV DNA level, baseline CD4 cell count, and other baseline demographic parameters on achieving undetectable residual viremia. Undetectable residual viremia was defined as each HIV RNA level at week 48, 64, and 72 being <2.5 copies/mL, the LOD of the modified assay. For the multivariate analysis, a stepwise method was applied in model selection, with a significance level of .10 for entry and staying in the model. P values for all models are from the 2 test. All other P values are from 2-sided Fisher's exact, Wilcoxon rank sum, or Wilcoxon signed rank tests, as indicated.
RESULTS
Selection of 100 subjects with HIV RNA levels <50 copies/mL from week 24 to week 72 yielded 55 patients treated with efavirenz, lamivudine, and tenofovir and 45 patients treated with efavirenz, lamivudine, and stavudine. The baseline characteristics of patients in these 2 treatment arms were similar, with a mean pretreatment HIV RNA level of 4.76 log10 copies/mL in each arm (table 1). The baseline CD4 cell count ranged from 5 to 885 cells/mm3, with a mean of 320 cells/mm3 and no significant difference between the arms. Undetectable residual viremia was achieved in 39% of the study subjects. This classification required all 3 plasma samples collected between weeks 48 and 72 of therapy to be below the LOD of the modified Roche Amplicor assay. The remainder (61%) of the subjects had at least 1 detectable HIV RNA level, but the levels were <50 copies/mL, per the substudy inclusion criteria. Within this group, the distribution of subjects with 1, 2, or 3 detectable HIV RNA levels appeared to be evenly distributed (n = 17, 19, and 25, respectively). When the treatment arms were compared in a univariate analysis, residual viremia was detectable in 29/55 subjects (53%) in the tenofovir arm and in 32/45 subjects (71%) in the stavudine arm (P = .07). The analysis evaluating posttreatment HIV RNA levels revealed no difference between the 2 treatment arms, with mean levels of 0.58 log10 copies/mL (3.8 copies/mL) in the tenofovir arm and 0.61 log10 copies/mL (4.1 copies/mL) in the stavudine arm (P = .24). Of note, in this analysis, a large proportion of the measurements were imputed as 0.38 log10copies/mL (2.4 copies/mL) if they were below the LOD of the HIV RNA assay. Reductions in HIV DNA were also similar between the treatment arms (table 2). After week 72, the mean reductions in HIV DNA levels for the tenofovir and stavudine arms were 0.56 and 0.48 log10copies/g, respectively (P = .32).
In the bivariate logistic-regression model analysis, 7 potential predictors of residual viremia were analyzed. Of these 7 predictors, 3 were significant: baseline HIV RNA level (in log10 copies/mL), baseline HIV DNA level (in log10 copies/g), and treatment arm (table 2). Age, sex, race, and baseline CD4 cell count were not significant predictors. When a stepwise multivariate regression analysis was applied, higher baseline HIV RNA levels (P = .016), higher baseline HIV DNA levels (P = .048), and assignment to the stavudine arm (P = .049) were all associated with higher risk of detectable residual viremia (table 2).
To determine possible consequences of on-therapy residual viremia between weeks 48 and 72, the frequency of HIV RNA level blips after 72 weeks, the occurrence of late therapy virologic failure, and the change in CD4 cell counts were examined through week 144 of the clinical trial. HIV RNA level blips >50 copies/mL after week 72 occurred in 20 of 100 patients in this substudy. However, there was no association between the occurrence of these blips and either treatment arm or on-therapy residual viremia (P .8). In all cases, the blips occurred at single time points, with the plasma samples collected afterward again showing HIV RNA levels <50 copies/mL. Virologic failure after week 72 occurred infrequently in this cohort; there were 2 cases in each treatment arm, and 1 case in each treatment arm was associated with detectable residual viremia. There were 4 confirmed cases of rebound to HIV RNA levels of >400 copies/mL. Finally, changes in CD4 cell counts were evaluated. At weeks 48, 96, and 144, there were no significant differences from baseline in CD4 cell responses in patients with residual viremia, compared with those without residual viremia (P .11). Within this substudy, the median follow-up duration was 72 weeks after the week 72 visit, with 94% of the substudy patients having completed week 144 of the clinical trial.
DISCUSSION
Rates of virologic suppression achieved among patients receiving the current generation of HAART regimens exceed those of initial multidrug therapies for a number of reasons, including drug potency, ease of administration, and tolerance. In recent clinical trials, virologic suppression was achieved for prolonged periods in >70% of patients. Although achieving HIV RNA levels <50 copies/mL has repeatedly been identified as a critical factor in the prediction of the risk of virologic rebound, the clinical significance or benefit of achieving even lower HIV RNA levels has not yet been established. The ability to address this question has been limited, until recently, by the availability of assays sensitive enough to measure plasma HIV RNA levels <50 copies/mL [6, 8, 9].
In this study, we evaluated residual viremia by applying an assay with an LOD of 2.5 copies/mL. We were interested both in the predictors and in the consequences of residual viremia in successfully treated patients. We specifically excluded patients in whom plasma HIV RNA levels <50 copies/mL were not achieved, because of the increased risk of virologic failure associated with the nadir reached after treatment initiation [15]. In a previous study of patients in whom sustained suppression of HIV RNA levels to <50 copies/mL for 5 years had been achieved, we identified HIV DNA level as the strongest predictor of residual viremia [5]. This observation was limited to a small number of patients receiving efavirenz and 3-times-daily indinavir, a regimen not currently recommended for initial HIV therapy [16]. In the present study, we have confirmed our initial finding that HIV DNA level is an independent predictor of residual viremia and have provided support for a model in which there is a gradient of residual viremia in patients receiving identical treatment regimens that is predicted independently by both baseline pretreatment HIV RNA level in plasma and HIV DNA level in PBMCs. We have also demonstrated that there is a difference between successful regimens in their antiviral potency. On the basis of the reductions in HIV RNA level observed in monotherapy studies, a tenofovir regimen would be expected to achieve higher levels of viral suppression than an otherwise identical stavudine regimen [17, 18]. Using a sensitive assay and a multivariate analysis, we did find that higher rates of undetectable viremia were achieved in patients receiving tenofovir than in those receiving stavudine, even when baseline HIV DNA levels in PBMCs and HIV RNA levels in plasma were accounted for.
Our finding that pretreatment HIV DNA level in PBMCs is an independent predictor of residual viremia in patients receiving HAART adds to accumulating data supporting a prognostic value of this parameter. In untreated patients, HIV DNA level in PBMCs has predicted disease progression independent of HIV RNA level in 3 separate studies [1921]. HIV DNA level in PBMCs has also predicted disease progression and even mortality independent of HIV RNA level in early treatment studies using dual-nucleoside RT inhibitors [22]. One hypothesis that could explain the association between HIV DNA level and residual viremia proposes that latently infected cells are the primary source of virus for permissive cells that maintain a steady state of productive infection [5, 23]. Our study cannot directly address this hypothesis, since we did not distinguish replication-competent provirus from defective DNA. An alternative explanation for the observation that pretreatment HIV DNA level is a predictor of residual viremia is that HIV DNA levels detected during viremia are a surrogate for viral or host factors that influence the magnitude and kinetics of viral replication or cell infectivity.
In untreated patients, HIV DNA includes linear unintegrated HIV DNA, circular episomes, and integrated provirus [24]. At baseline, a large majority of the HIV DNA is likely to be in a linear unintegrated form, as the product of recent infection [12, 25, 26]. Higher baseline HIV DNA levels before suppression of virus replication could, therefore, indicate an increased susceptibility of host cells to infection or a higher viral replication capacity that is independent of the pathogenic factors underlying baseline plasma HIV RNA levels and could thereby serve to independently predict levels of ongoing residual replication during treatment. Findings from other studies establishing the prognostic value of HIV DNA for disease progression [20] and for the kinetics of rebound of HIV RNA during structured treatment interruption [27] are consistent with this hypothesis.
The clinical application of the finding that a gradient of residual viremia exists among successfully treated patients and is predicted by baseline HIV RNA level, baseline PBMC DNA level, and treatment regimen has yet to be established. In this study, achieving higher levels of viral suppression did not confer greater CD4 cell count increases, which is consistent with the results of prior studies showing no differences in CD4 cell count increases between lopinavir plus ritonavirbased and nelfinavir-based regimens [28]. Likewise, higher levels of viral suppression did not predict reduced rates of virologic failure. Although the certainty of this conclusion is limited by the overall low rates of virologic failure and duration of follow-up for a population that may require treatment for decades, this finding is consistent with those of one of our earlier studies showing that patients with transient increases in HIV RNA level (i.e., blips) and median HIV RNA levels of 30 copies/mL were at no greater risk for rebound than those with HIV RNA levels of 3 copies/mL [9]. Outside of clinical trials, however, in which adherence rates are often lower, a regimen with greater antiviral potency (if we assume no greater toxicity) may have a clinical advantage, and that hypothesis merits testing. Likewise, longer-term studies that take into account the evolution of drug resistance and drug sequencing strategies may shed light on the potential benefits of higher levels of viral suppression.
References
1. Ho DD, Neumann AU, Perelson AS, Chen W, Leonard JM, Markowitz M. Rapid turnover of plasma virions and CD4 lymphocytes in HIV-1 infection. Nature 1995; 373:1236. First citation in article
2. Wei X, Ghosh SK, Taylor ME, et al. Viral dynamics in human immunodeficiency virus type 1 infection. Nature 1995; 373:11722. First citation in article
3. Gulick RM, Mellors JW, Havlir D, et al. Treatment with indinavir, zidovudine, and lamivudine in adults with human immunodeficiency virus infection and prior antiretroviral therapy. N Engl J Med 1997; 337:7349. First citation in article
4. Hammer SM, Squires KE, Hughes MD, et al. A controlled trial of two nucleoside analogues plus indinavir in persons with human immunodeficiency virus infection and CD4 cell counts of 200 per cubic millimeter or less. AIDS Clinical Trials Group 320 Study Team. N Engl J Med 1997; 337:72533. First citation in article
5. Havlir DV, Strain MC, Clerici M, et al. Productive infection maintains a dynamic steady state of residual viremia in human immunodeficiency virus type 1-infected persons treated with suppressive antiretroviral therapy for five years. J Virol 2003; 77:112129. First citation in article
6. Yerly S, Kaiser L, Perneger TV, et al. Time of initiation of antiretroviral therapy: impact on HIV-1 viraemia. The Swiss HIV Cohort Study. AIDS 2000; 14:2439. First citation in article
7. Dornadula G, Zhang H, VanUitert B, et al. Residual HIV-1 RNA in blood plasma of patients taking suppressive highly active antiretroviral therapy. JAMA 1999; 282:162732. First citation in article
8. Palmer S, Wiegand AP, Maldarelli F, et al. New real-time reverse transcriptase-initiated PCR assay with single-copy sensitivity for human immunodeficiency virus type 1 RNA in plasma. J Clin Microbiol 2003; 41:45316. First citation in article
9. Havlir DV, Bassett R, Levitan D, et al. Prevalence and predictive value of intermittent viremia with combination HIV therapy. JAMA 2001; 286:1719. First citation in article
10. Finzi D, Hermankova M, Pierson T, et al. Identification of a reservoir for HIV-1 in patients on highly active antiretroviral therapy. Science 1997; 278:1295300. First citation in article
11. Wong JK, Hezareh M, Gunthard HF, et al. Recovery of replication-competent HIV despite prolonged suppression of plasma viremia. Science 1997; 278:12915. First citation in article
12. Chun TW, Carruth L, Finzi D, et al. Quantification of latent tissue reservoirs and total body viral load in HIV-1 infection. Nature 1997; 387:1838. First citation in article
13. Siliciano JD, Kajdas J, Finzi D, et al. Long-term follow-up studies confirm the stability of the latent reservoir for HIV-1 in resting CD4+ T cells. Nat Med 2003; 9:7278. First citation in article
14. Gallant JE, Staszewski S, Pozniak AL, et al. Efficacy and safety of tenofovir DF vs stavudine in combination therapy in antiretroviral-naive patients: a 3-year randomized trial. JAMA 2004; 292:191201. First citation in article
15. Kempf DJ, Rode RA, Xu Y, et al. The duration of viral suppression during protease inhibitor therapy for HIV-1 infection is predicted by plasma HIV-1 RNA at the nadir. AIDS 1998; 12:F914. First citation in article
16. US Government. Guidelines for the use of antiretroviral agents in HIV 1-infected adults and adolescents. Available at: http://aidsinfo.nih.gov/guidelines/adult/AA_102904.html. Accessed 23 February 2005. First citation in article
17. Pollard RB, Tierney C, Havlir D, et al. A phase II randomized study of the virologic and immunologic effect of zidovudine + stavudine versus stavudine alone and zidovudine + lamivudine in patients with >300 CD4 cells who were antiretroviral naive (ACTG 298). AIDS Res Hum Retroviruses 2002; 18:699704. First citation in article
18. Barditch-Crovo P, Deeks SG, Collier A, et al. Phase i/ii trial of the pharmacokinetics, safety, and antiretroviral activity of tenofovir disoproxil fumarate in human immunodeficiency virus-infected adults. Antimicrob Agents Chemother 2001; 45:27339. First citation in article
19. Kostrikis LG, Touloumi G, Karanicolas R, et al. Quantitation of human immunodeficiency virus type 1 DNA forms with the second template switch in peripheral blood cells predicts disease progression independently of plasma RNA load. J Virol 2002; 76:10099108. First citation in article
20. Gottlieb GS, Sow PS, Hawes SE, et al. Equal plasma viral loads predict a similar rate of CD4+ T cell decline in human immunodeficiency virus (HIV) type 1 and HIV-2infected individuals from Senegal, West Africa. J Infect Dis 2002; 185:90514. First citation in article
21. Rouzioux C. Early HIV-1 DNA level predicts disease progression and death independently of HIV-RNA level and CD4 cell count [abstract 23]. In: Program and abstracts of the 1st International AIDS Society Conference (IAS) on HIV Pathogenesis and Treatment (Buenos Aires, Argentina). Stockholm: IAS Headquarters, 2001. First citation in article
22. Tierney C, Lathey JL, Christopherson C, et al. Prognostic value of baseline human immunodeficiency virus type 1 DNA measurement for disease progression in patients receiving nucleoside therapy. J Infect Dis 2003; 187:1448. First citation in article
23. Grossman Z, Polis M, Feinberg MB, et al. Ongoing HIV dissemination during HAART. Nat Med 1999; 5:1099104. First citation in article
24. Chun TW, Finzi D, Margolick J, Chadwick K, Schwartz D, Siliciano RF. In vivo fate of HIV-1-infected T cells: quantitative analysis of the transition to stable latency. Nat Med 1995; 1:128490. First citation in article
25. Blankson JN, Finzi D, Pierson TC, et al. Biphasic decay of latently infected CD4+ T cells in acute human immunodeficiency virus type 1 infection. J Infect Dis 2000; 182:163642. First citation in article
26. Zack JA, Arrigo SJ, Weitsman SR, Go AS, Haislip A, Chen IS. HIV-1 entry into quiescent primary lymphocytes: molecular analysis reveals a labile, latent viral structure. Cell 1990; 61:21322. First citation in article
27. Perrin L. Evolution of HIV DNA during therapy. In: Program and abstracts of the XI Symposium on HIV Infection (Toulon, France), 2001. First citation in article
28. Walmsley S, Bernstein B, King M, et al. Lopinavir-ritonavir versus nelfinavir for the initial treatment of HIV infection. N Engl J Med 2002; 346:203946. First citation in article