The evolution of antiretroviral therapy since 1987 has been driven by the goal of maximizing both the degree and duration of virologic suppression. However, there is increasing evidence that virologic suppression alone is necessary but not sufficient for long-term clinical success. Despite improvements over time, life expectancy among patients on antiretroviral therapy remains lower than in the general population1 but has been observed to normalize after 6 or more years of sustained CD4 levels above 500 cells/mm3.2 However, up to 25% of patients on virologically suppressive therapy fail to achieve a robust immune recovery.3 Poor CD4 recovery and on-treatment immunodeficiency are associated with a greater incidence of both AIDS-related and non-AIDS-related diseases and cancers,4,5 even among those with undetectable viremia.6 Thus, despite the innate link between virologic control and immune recovery, independent strategies to maximize CD4+ cell response may be needed to safeguard long-term outcomes.
Recent meta-analyses of clinical trial data have suggested that, for a given level of virologic suppression, patients receiving regimens containing CCR5 antagonist antiretrovirals experience greater increases in CD4+ cells than those receiving regimens without them.7 We have therefore undertaken a post hoc analysis to investigate CD4+ cell responses and associated clinical outcomes in two large phase III registrational trials of the CCR5 antagonist maraviroc (MVC).
Pooled 48-week data from the identically designed MOTIVATE 1 and 2 studies of MVC once (QD) or twice daily (BID) plus an optimized background therapy versus optimized background therapy alone in treatment-experienced patients with CCR5-tropic HIV-1 were analyzed. The design, patient demographics, and primary results from these studies have been fully described elsewhere.8
All CD4+ T-cell analyses used last-observation-carried-forward imputation of missing data, with baseline CD4 values defined as the mean of the screening and baseline (day 1) visits. Patients without baseline data (n = 2) were excluded. Longitudinal changes were analyzed using a repeated measures model (PROC MIXED in SAS, Carey, NC, USA) including terms for treatment (both MVC arms combined), week, week by treatment interaction, and whether a patient achieved and maintained less than 50 copies/mL through week 48. Week-48 changes from baseline were analyzed in a multivariate linear regression model using covariates of baseline CD4+ T-cell count, baseline HIV-1 RNA, week-48 RNA change, age, sex, and treatment. Cox proportional hazards modeling was used to evaluate the hazard ratios (HR) and 95% confidence intervals (CI) for the occurrence of AIDS-defining events using baseline HIV-1 RNA, time-dependent on-treatment CD4 counts, age, sex, and treatment as covariates. All analyses and summaries were performed using SAS (version 8.2, Carey, NC, USA).
There were 1047 patients in these analyses (413 MVC QD, 426 MVC BID, 208 placebo [PBO]) Most (89%) were male, median age was 45 years, median baseline HIV-1 RNA was 4.9 log10 copies/mL, and median baseline CD4+ T-cell count was 169 cells/mm3. Median CD4+ cell percentages were 12%, 12%, and 11% in the PBO, MVC QD, and MVC BID arms, respectively.
CD4+ Cell Increases
Overall, patients receiving MVC QD or BID experienced significantly larger and faster median CD4+ T-cell increases than patients receiving PBO, with most of the difference accruing within the first 12 weeks (Fig. 1A) when greater than 70% of patients remained on their assigned treatment. This difference was also observed in the subset of patients who achieved a viral load less than 50 copies/mL at any point (Fig. 1B), most of whom (74.4%) remained virologically suppressed through week 48 once less than 50 copies/mL was achieved. Furthermore, the week-48 CD4 benefit associated with MVC treatment also persisted when adjusted for other known influences9 on CD4 gain, including age, baseline CD4, baseline HIV-1 RNA, sex, and RNA change from baseline, by multivariate regression modeling. Adjusted for these other confounders, MVC treatment was still associated with a significant (P = 0.0015) 26-cell increase in CD4 count over PBO treatment at week 48. Similar results were observed when the same model was run for short-term (12-wk) CD4 changes.
Significantly more patients at high baseline risk of opportunistic infections (<200 CD4+ cells/mm3) achieved greater than 200 cells/mm3 at week 48 on MVC BID than PBO (47% vs. 23%; P < 0.0001), which was the only prespecified comparison. By week 48, median CD8+ cell counts had increased by 141 and 151 cells/mm3 in the MVC QD and BID arms, respectively, but only by 10 cells/mm3 among PBO recipients, resulting in week-48 increases in median CD4% of 3.2%, 3.3%, and 1.1% in the MVC QD, BID, and PBO arms, respectively.
Time to AIDS-Defining (Category C) Events
Time to first occurrence of a category C event was significantly longer among MVC recipients (Fig. 1C) in the univariate analysis. Multivariate regression modeling, as described, identified significant associations between risk of an event and baseline HIV-1 RNA (HR 2.3 per 1 log10 increase [95% CI 1.6-3.3]), absolute on-treatment CD4 levels (HR 0.8 per 25 cell/mm3 increase [95% CI 0.78-0.87]), and sex (HR 1.9 for women vs. men [95% CI 1.1-3.4]), whereas baseline CD4 and treatment were not significantly associated.
CD4+ cell increases over 48 weeks in the MOTIVATE studies were significantly greater in patients receiving MVC than in those who did not, with the difference remaining significant both when adjusted for other predictors of CD4 response (including viral load reduction) and when the patient set was restricted to those in any treatment arm who achieved undetectable viremia. These data are consistent with previous observations of large CD4 increases on MVC in non-R5 infections without significant virologic response to MVC10 and larger CD4 increases in treatment-naïve patients receiving MVC compared with efavirenz despite similar control of HIV replication.11 The data imply that the benefit observed is not solely a function of MVC's virologic potency. The mechanism of this finding is uncertain, but CCR5 antagonists have the potential to influence several known effectors of HIV-associated CD4+ cell decline, including reducing cytokine-mediated immune activation and CCR5-dependent cytolysis and apoptosis,12,13 which could contribute to these observations.
Although the MOTIVATE studies were not designed to find differences in clinical endpoints, we found that the time to a new AIDS-defining event was significantly longer for patients receiving MVC. In a multivariate hazard model, a significant protective effect on the incidence of new events was observed for greater time-dependent on-treatment CD4 count but not for baseline CD4 or treatment arm. The lack of a treatment effect is presumably a function of the clustering between greater CD4 increase and MVC use in this patient set and implies that the precise antiretrovirals used to increase CD4 count do not influence subsequent improvements in immune function as assessed by time to disease. Because time-dependent CD4 count was used in this model, it is not surprising that the most proximate CD4 was the strongest predictor of progression.
These post hoc analyses have certain limitations that we have attempted to minimize through the analytical design. A large part of the observed CD4 rises are potentially attributable to greater virologic suppression on MVC than PBO in these patients; however, as described, the CD4 difference remains when adjusted for postbaseline HIV-1 RNA reduction or assessed only in patients with undetectable viremia.
Furthermore, the large discontinuation rate in the PBO arm may have affected week-48 results by the last-observation-carried-forward imputation used here. However, week-48 findings were also borne out by early (wk-12) data as described, suggesting that the imputation method did not significantly influence these outcomes.
In conclusion, MVC-treated patients in the MOTIVATE trials experienced larger increases in CD4+ T-cell counts than those who received PBO, which included a MVC-specific component independent of its antiviral activity. This greater CD4+ cell increase was associated with a significantly longer time to new AIDS-defining events in MVC patients, but multivariate analysis did not identify an independent effect of MVC on the risk of a new event over and above its influence on CD4 changes.
1. ART Cohort Collaboration. Life expectancy of individuals on combination antiretroviral therapy in high-income countries: a collaborative analysis of 14 cohort studies. Lancet
2. Lewden C, Chene G, Morlat P, et al. HIV-infected adults with a CD4 cell count greater than 500 cells/mm3
on long-term combination antiretroviral therapy reach same mortality rates as the general population. J Acquir Immune Defic Syndr
3. Robbins GK, Spritzler JG, Chan ES, et al. Incomplete reconstitution of T cell subsets on combination antiretroviral therapy in the AIDS Clinical Trials Group protocol 384. Clin Infect Dis
4. Baker JV, Peng G, Rapkin J, et al. Poor initial CD4+
recovery with antiretroviral therapy prolongs immune depletion and increases risk for AIDS and non-AIDS diseases. J Acquir Immune Defic Syndr
5. Silverberg MJ, Abrams DI. AIDS-defining and non-AIDS-defining malignancies: cancer occurrence in the antiretroviral therapy era. Curr Opin Oncol
6. Moore DM, Hogg RS, Chan K, et al. Disease progression in patients with virological suppression in response to HAART is associated with the degree of immunological response. AIDS
7. Wilkin T, Ribaudo H, Gulick R.The relationship of CCR5 inhibitors to CD4 cell count changes: a meta-analysis of recent clinical trials [Abstract 800]. From the Proceedings of the 15th Conference on Retroviruses and Opportunistic Infections, Boston, MA, February 3-6, 2008.
8. Gulick R, Lalezari J, Goodrich J, et al. for the MOTIVATE Study Teams. Maraviroc for previously treated patients with R5 HIV-1 infection. N Engl J Med
9. Gandhi RT, Spritzler J, Chan E, et al. Effect of baseline- and treatment-related factors on immunologic recovery after initiation of antiretroviral therapy in HIV-1-positive subjects: results from ACTG 384. J Acquir Immune Defic Syndr
10. Saag M, Goodrich J, Fätkenheuer G, et al. A double-blind, placebo controlled trial of maraviroc in treatment-experienced patients infected with non-R5 HIV-1. J Infect Dis
11. Lazzarin A, Battegay M, Cooper DA, et al. CD4+
Cell increases at 48 weeks in the maraviroc treatment-naive MERIT trial [Abstract H-1248]. From the Proceedings of the 48th Interscience Conference on Antimicrobial Agents and Chemotherapy/46th Annual Meeting of the Infectious Diseases Society of America, Washington, DC, October 26-28, 2008.
12. Decrion AZ, Dichamp I, Varin A, et al. HIV and inflammation. Curr HIV Res
13. Lelievre JD, Petit F, Perrin L, et al. The density of coreceptors at the surface of CD4+
T cells contributes to the extent of human immunodeficiency virus type 1 viral replication-mediated T cell death. AIDS Res Hum Retroviruses