Gastrointestinal-associated lymphoid tissue immune reconstitution in a randomized clinical trial of raltegravir versus non-nucleoside reverse transcriptase inhibitor-based regimens.

Objectives: To examine immune restoration in duodenal tissue and correlates of reduction of immune activation in chronic HIV-infected patients randomized to different treatment regimens.

Design: Randomized clinical trial (RCT) comparing raltegravir to a non-nucleoside reverse transcriptase inhibitor-based regimen, both with fixed-dose tenofovir difumerate/emtricitabine.

Methods: Antiretroviral therapy (ART)-naive volunteers underwent upper endoscopy for duodenal biopsies before and after 9 months of therapy. Tissue was paraffin-embedded for immunohistochemistry or digested into single-cell suspensions for flow cytometry of lymphocyte subsets and activation phenotype. Plasma-soluble CD14 levels were measured as a surrogate for bacterial translocation.

Results: Sixteen HIV-positive and seven control individuals completed study procedures. Small increases in duodenal lamina propria CD4 T-cell numbers were observed, especially when viewed relative to populations in control volunteers, with no differences between treatment arms. The increase in CD4 T-cell percentage was due largely to declines in CD8 T-cell numbers, which were disproportionately increased compared to peripheral blood and controls. Patients randomized to the raltegravir arm had consistent declines in both sCD14 levels and CD8 T-cell numbers in the duodenal tissue lamina propria.

Conclusions: This first RCT of lymphocyte population restoration in duodenal tissue demonstrates more modest increases in CD4 T-cell numbers during the first 9 months of therapy than when considering CD3/CD4 percentages only. Although reduced after 9 months of ART, disproportional increased CD8 populations persist in duodenal gastrointestinal-associated lymphoid tissue (GALT). Local rather than systemic antigenic stimulation appears to be driving expanded CD8 T lymphocytes in GALT. Factors other than viral-induced CD8 expansion may be contributing to this local immunologic response.