Abstract Body

Background: Despite peripheral immunological recovery, incomplete restoration of the gut associated lymphoid tissue (GALT) during treated HIV infection is likely a major determinant of disease progression. The association between blood and GALT T-cell changes following ART initiation, and whether these changes correlate across the different gastrointestinal tract segments remain unknown. Methodology: Peripheral blood (PB), duodenal and rectal biopsies were obtained from 11 HIV- controls and 33 ART naive HIV+ subjects participating in a randomized clinical trial comparing three different ART initiation strategies, at baseline and after 9 months of treatment. Tissue was digested into single-cell suspensions for measurements of T-cell subsets and activation phenotype by immunophenotypin Results: Twenty-six HIV+ subjects completed the follow-up. Median age was 37 years [25-42] vs 34 [28-43] in the control group, and median CD4 T-cell count 435 cell/mL [283-572]. Eight subjects received EFV, 10 MRV and 8 MVC+RAL, each in combination with FTC/TDF, and all achieved viral suppression. Overall, ART resulted in a significant repletion of CD4 T-cell in GALT at month 9, greater in rectum (24% to 38%) than in duodenum (12% to 17%) (all P<0.01). Also, CD8 T-cells decreased in rectum (62% to 53%) and duodenum (81% a 70%) (all P<0.01). However, the levels of both CD4 and CD8 T-cells remained very impaired relative to controls (beyond 2 SD). Changes in %CD4 in PB (24% to 40%, P<0.01) predicted %CD4 changes in colon (Beta 0.54, P=0.023) but not in duodenum (Beta -0.19, P=0.272). The %HLADR+CD38+ of CD8 T-cells was significantly higher at baseline in PB and rectal mucosa in HIV+ compared to HIV- subjects, but not in duodenum. CD8 T-cell activation significantly decreased in blood (57% to 28%, P<0.001), duodenum (65% to 52%, P=0.015) and rectum (71% to 58%, P=0.429,] and the change in PB was reflective of the smaller mucosal changes [duodenum Beta=0.39, P<0.01, and rectum Beta=0.23, P<0.01). However, the % of activated CD8 T-cells in PB and in colon did not reach normal levels. In addition, the % of PB activated CD8 T-cells at baseline predicted the extent of CD4 depletion in duodenum (Beta=-0.7, P<0.001) and colon (Beta=-0.7, P<0.001). HIV+ subjects showed similar % of naive CD8+ T-cells in colon and duodenum than controls, but not in PB. The CD8 naive/memory ratio significantly improved in blood but did not reach normal levels. Conclusions: Following ART initiation we observed incomplete recovery of GALT CD4 T-cells. Immune reconstitution of the duodenal mucosa was impaired compared to changes in blood and rectum. The duodenal compartment may represent a portal for systemic immune activation and CD4 depletion during HIV infection and further studies of this compartment are warranted.