Background: We reported the loss of detectable HIV-1 reservoirs in 2 individuals following reduced-intensity conditioning allogeneic hematopoietic stem cell transplantation (HSCT) from CCR5 wild-type donors. HIV-1 remained undetectable from blood and rectal tissue on ART despite the testing of large numbers of PBMCs or CD4+ T cells in the setting of full donor chimerism. As a result, analytical treatment interruption (ATI) was performed, and we now provide an in-depth analysis of HIV-1 viral/immune dynamics during ATI and after viral rebound. Methodology: ATI was performed in consenting patients with weekly to biweekly monitoring of viral load (VL) and proviral DNA by clinical assays. Largevolume blood draws, leukapheresis, and cerebral spinal fluid sampling was performed before, during and/or after ATI to perform: 1) quantification of HIV-1 DNA in PBMCs and in sorted CD4+ T lymphocyte subsets (central memory, effector memory, stem cell memory, terminally differentiated, and naive cells), 2) plasma RNA quantification by single copy assay (SCA), 3) characterization of HIV-specific T cell immunity and viral rebound and decay following ART re-initiation, and 4) single-genome analysis of full-length HIV-1 env sequences from DNA prior to HSCT and post-rebound plasma RNA. Results: During ATI, patient A had no detectable plasma RNA or cell-associated HIV-1 DNA for 3 months and patient B had no detectable virus for 8 months after ART cessation prior to eventual rebound. Patient B also had negative PBMC DNA (<0.07 copies/10^6 PBMCs) and viral RNA by SCA (<0.4 copies/mL) by ultrasensitive assays 5 and 18 weeks into ATI. Once detected, plasma RNA increased rapidly, reaching 2-4x10^6 copies/ml within 2-3 weeks; rebound plasma viremia was mono/oligophyletic. Both patients developed symptoms of acute retroviral syndrome, including aseptic meningitis and new NNRTI resistance in one patient that developed early after ART-initiation. Symptoms rapidly resolved with initiation of active ART and subsequent viral suppression in both patients. Patient A had detectable but low-level HIV-1 DNA in pre-ATI terminally differentiated lymphocytes isolated from 200x10^6 PBMCs from post-hoc testing after rebound, but from no other subsets; patient B had no detectable HIV-1 DNA. Conclusions: Our results suggest that allogeneic HSCT may lead to loss of detectable HIV-1 from blood and gut tissue, but viral rebound can occur despite up to a 3-4 log10 reduction in reservoir size. Long-lived tissue reservoirs, including host macrophages that may be replaced more slowly than T-lymphocytes following HSCT may have contributed to viral rebound. More sensitive assays for detecting HIV persistence are needed; ATI remains the most reliable measure of viral persistence following therapeutic intervention.
CROI Conference
