The immunologic effects of maraviroc intensification in treated HIV-infected individuals with incomplete CD4+ T-cell recovery: a randomized trial

Abstract

The CCR5 inhibitor maraviroc has been hypothesized to decrease T-cell activation in HIV-infected individuals, but its independent immunologic effects have not been established in a placebo-controlled trial. We randomized 45 HIV-infected subjects with CD4 counts <350 cells per mm(3) and plasma HIV RNA levels <48 copies per mL on antiretroviral therapy (ART) to add maraviroc vs placebo to their regimen for 24 weeks followed by 12 weeks on ART alone. Compared with placebo-treated subjects, maraviroc-treated subjects unexpectedly experienced a greater median increase in % CD38+HLA-DR+ peripheral blood CD8+ T cells at week 24 (+2.2% vs -0.7%, P = .014), and less of a decline in activated CD4+ T cells (P < .001). The % CD38+HLA-DR+ CD4+ and CD8+ T cells increased nearly twofold in rectal tissue (both P < .001), and plasma CC chemokine receptor type 5 (CCR5) ligand (macrophage-inflammatory protein 1β) levels increased 2.4-fold during maraviroc intensification (P < .001). During maraviroc intensification, plasma lipopolysaccharide declined, whereas sCD14 levels and neutrophils tended to increase in blood and rectal tissue. Although the mechanisms explaining these findings remain unclear, CCR5 ligand-mediated activation of T cells, macrophages, and neutrophils via alternative chemokine receptors should be explored. These results may have relevance for trials of maraviroc for HIV preexposure prophylaxis and graft-versus-host disease. This trial was registered at www.clinicaltrials.gov as #NCT00735072.

Figure 1

Enrollment, allocation, and follow-up for trial subjects. The outcomes of the 67 screened subjects are described in the Consolidated Standards of Reporting Trials (CONSORT) flow diagram.

Figure 2

Changes in low-level viremia during maraviroc intensification. Changes in low-level viremia (by SCA) were assessed with linear mixed models for maraviroc-treated (red) and placebo-treated (blue) subjects. Undetectable values were assigned a value equal to the lowest limit of detection for the assay. Estimated mean levels at each time point are plotted with 95% CIs. Mean changes over each indicated interval are also compared both within and between each arm, with P values provided for each comparison.

Figure 3

Changes in CD4+ and CD8+ T-cell counts in peripheral blood and rectal tissue during maraviroc intensification. Median changes from baseline in peripheral blood CD4+ (A) and CD8+ (B) T-cell counts are plotted over time for maraviroc-treated (red) and placebo-treated (blue) subjects. Vertical error bars represent IQR at each time point. The mean slopes of change over the indicated intervals (with P values) are also reported. Estimated mean changes (with 95% CI) in % CD4+ (C) and % CD8+ (D) T cells in rectal tissue (by flow cytometry on extracted cells), as well as the % CD4+ area (by immunohistochemistry) in both the lamina propria (E) and lymphoid aggregates (F) of rectal tissue, are also plotted over time for maraviroc-treated (red) and placebo-treated (blue) subjects. P values are provided for the changes within each indicated interval both within arms and between arms. Because only a subset of subjects had baseline rectal tissue with adequate representation of lamina propria and lymphoid aggregates, the number of subjects contributing to each time point is noted below the figure.

Figure 4

Changes in peripheral blood and rectal T-cell activation during maraviroc intensification. Estimated mean changes (with 95% CI) in the frequency of activated (CD38+HLA-DR+) CD4+ (A) and CD8+ (B) T cells in peripheral blood are plotted for placebo-treated (blue) and maraviroc-treated (red) subjects. Estimated mean changes (with 95% CI) in the frequency of activated CD4+ (C) and CD8+ (D) T cells in rectal tissue are also plotted for both arms. P values are provided for the changes within each indicated interval both within arms and between arms.

Figure 5

Changes in the frequencies of T-cell maturational phenotypes during maraviroc intensification. Estimated mean changes in naive (CD45RA+CCR7+) (A-B), central memory (CD45RA–CCR7+) (C-D), effector memory (CD45RA–CCR7–) (E-F), and terminally differentiated effector memory (TEMRA) (CD45RA+CCR7–) (G-H) CD4+ (A,C,E,G) and CD8+ (B,D,F,H) T-cell frequencies in peripheral blood (with 95% CI) are plotted for maraviroc-treated (red) and placebo-treated (blue) subjects over time. P values are provided for the changes within each indicated interval both within arms and between arms.

Figure 6

Changes in the frequency of CCR5+ T cells and plasma MIP-1β levels during maraviroc intensification. Changes in the % CCR5+ CD8+ T cells (A) and CD4+ T cells (B) in peripheral blood, the % CCR5+ CD4+ T cells in rectal tissue (C), and plasma levels of the CCR5 ligand MIP-1β (D) were assessed in placebo-treated (blue) and maraviroc-treated (red) subjects with linear mixed models. Estimated mean values are plotted over time with 95% CIs. P values are provided for the changes within each indicated interval both within arms and between arms.

Figure 7

Changes in monocyte activation, neutrophil levels, and plasma LPS levels during maraviroc intensification. Changes in plasma sCD14 (A), sCD163 levels (B), peripheral blood absolute neutrophil counts (C), rectal tissue myeloperoxidase density (a surrogate marker for neutrophil density) (D), and plasma LPS levels (E) are plotted over time for both maraviroc-treated (red) and placebo-treated subjects (blue). Mean changes from baseline (A-B) or estimated mean values (C-E) with 95% CI from linear mixed models are plotted. P values are provided for the changes within each indicated interval both within arms and between arms. Given more frequent observations, changes in peripheral blood neutrophil counts were modeled as a linear spline with a change point at week 24.