Autologous HIV-specific T cell therapy targeting conserved epitopes is well-tolerated in six adults with HIV: an open-label, single-arm phase 1 study

Abstract

Novel cellular therapies may enable HIV control or cure. HIV-specific T cells targeting conserved immunogenic protein regions of HIV Gag/Pol and the entirety of HIV Nef, termed HST-NEETs, eliminate HIV infected cells in vitro. Here we enroll seven participants in an open-label, single-arm phase 1 study (NCT03485963) to evaluate the safety (primary endpoint) of two autologous administrations of HST-NEET products without prescribed lymphodepletion. Adults with well-controlled HIV on anti-retroviral therapy are eligible. Six participants completed safety monitoring. No serious product-related toxicities are observed. Secondary endpoints are to assess expansion and persistence of HIV-reactive T cell clones, and changes to the HIV reservoir for each infused participant. HIV-specific T cell and HIV anti-Env antibody responses increase in two participants after infusion two. A trend towards decreasing levels of intact proviruses is observed in 2 participants. Three participants show persistence of HIV-reactive, product-associated T cell clones for ≥40 weeks post infusions. HST-NEETs infusions are well-tolerated. Future trials are needed to evaluate the efficacy of HST-NEETs in this population.

Fig. 1. CONSORT diagram.

Seven eligible participants were enrolled in the present study. Six out of seven HIV-specific T cell products targeting conserved epitopes (HST-NEETs) were expanded to clinical dose levels. Six participants received two administrations of autologous HST-NEET products with infusion one indicated at week zero. Participants were monitored for adverse events throughout the study (primary endpoint) and blood was drawn at follow-up visits to evaluate immunologic and virologic cell responses (secondary endpoints) following infusion one. Some secondary analyses for RESIST-004 HST-NEETs were not conducted following product release studies and infusion administrations due to limited cell numbers. Follow-up visits were conducted for a range of 46–103 weeks following infusion one.

Fig. 2. HST-NEET infusion products were polyfunctional T cells that recognized multiple HIV antigens.

a HST-NEET products (n = 6) exhibited variable antigen-specific IFN-γ responses against HIV Gag, Pol, Nef, and conserved epitope targets in the Gag/Pol Mos1 and Mos2 peptide mixes. RESIST-004 HST-NEET numbers after treatment-related studies were insufficient to test Gag/Pol Mos1 and Mos2 specificity. Positive ELISpot results were defined as IFN-γ spot forming cells ≥2 times the actin negative control. Spot forming cells (SFC) per 10⁵ cells indicative of specific responses compared to actin stimulated cells: RESIST-001 (actin [17 SFC], nef [139 SFC]), RESIST-003 (actin [19 SFC], gag [659 SFC], pol [466 SFC], nef [1057 SFC], mos1 [540 SFC], mos2 [520 SFC]), RESIST-004 (actin [34 SFC], nef [117 SFC]), RESIST-005 (actin [65 SFC], nef [804 SFC]), RESIST-006 (actin [0 SFC], nef [236 SFC]), and RESIST-007 (actin [25 SFC], gag [1063 SFC], nef [854 SFC], mos1 [291 SFC], mos2 [290 SFC]). b Each bar reports the total sum (infusion 1 and infusion 2) of infused T cells specific for HIV Gag, Pol, and Nef epitopes for each HST-NEETs product (n = 5). c Each bar reports the total sum (infusion 1 and infusion 2) of infused T cells specific for conserved epitopes in Gag/Pol Mosaic 1 and Mosaic 2 immunogens for each HST-NEETs product (n = 5). d, e HST-NEET products (n = 6) were predominantly CD8+ effector memory T cells and had variable expression of T cell surface activation and exhaustion markers. f CD3+ CD8+ HST-NEET products (n = 5) exhibited polyfunctional T cell responses upon stimulation with Gag, Pol, Nef and Gag/Pol Mos1 and Mos2 peptides as measured by intracellular production of IFN-γ and TNF-α. HIV-specific T cell activation was dominant in the CD8+ T cell fraction of all HST-NEET products. Source data are provided as a Source Data file.

Fig. 3. Epitope specificities of HIV-specific CD8+ T cell responses in HST-NEET products.

a Comprehensive pooled peptide epitope mapping of HST-NEETs revealed 15-mer peptides driving HIV-specific responses in each product (n = 5). RESIST-001, -003, -005, and -006 ELISpots were conducted as singlets based on HST-NEET numbers available for characterization studies, and RESIST-007 ELISpots were conducted in duplicate. Plotted SFC per 10⁵ cells is shown above each bar. b–f HST-NEET products were stimulated with 15-mer peptides in (a) for 5 h and intracellular production of IFN-γ and TNF-α in CD8+ T cells was evaluated to confirm epitope-specific responses. Peptides evaluated are named by the first two amino acids in the 15-mer sequence followed by the peptide length. g Minimal (9 amino acid) epitopes associated with CD8+ T cell responses were determined for each product. RESIST-004 HST-NEETs were unavailable for these product characterization studies. Positive ELISpot results were defined as IFN-γ spot forming cells ≥2 times the actin negative control. RESIST-005 ELISpots were conducted as singlets, RESIST-001, -003, -006, and -007 were conducted in duplicate. Plotted SFC per 10⁵ cells is shown above each bar.

Fig. 4. Differences in HIV-specific T cell responses and anti-Env antibody responses were observed in RESIST-003 and RESIST-006.

a HIV-specific T cell responses showed significant changes in RESIST-003 post infusion two in response to Gag/Pol Mos2 (p = 0.007), Nef FPVRPQVPL (p = 0.003), Nef RPQVPLRPM (p = 0.002), Mos1/Pol LVGPTPVNI (p = 0.02), Mos1/Pol IIKDYGKQM (p = 0.05) and Mos2/Pol IIRDYGKQM (p = 0.002) peptides. Significant differences were also observed in RESIST-006 post infusion two in response to Gag/Pol Mos2 (p = 0.05) and Nef RPQVPLRPM (p = 0.05) peptides. ELISpots were performed in singlets to assess RESIST-001, RESIST-003, RESIST-004, and RESIST-007 responses due to limited PBMC numbers, and in duplicates for RESIST-005 and RESIST-006 responses. Statistical significance (P-value < 0.05) of IFN-γ T cell responses was compared between grouped SFC values up to and including infusion two, and after infusion two for each HIV antigen as shown in graph legends. b Anti-Env antibody responses were maintained across participant plasma samples with significant differences in antibody responses between pre and post infusion two in RESIST-003 (p = 0.006) and RESIST-006 (p = 0.02). Statistical significance (P-value <0.05) of changes in antibody responses was compared between ED50 values up to and including infusion two, and after infusion two. Blue shaded areas indicate timepoints up to and including the week of infusion two. Each x-axis timepoint has a corresponding y-axis value on the plotted trendlines. Vertical dashed lines (--) indicate infusion one and infusion two, respectively. Two-sided Wilcoxon rank-sum tests were used to determine significant differences between grouped values for timepoints up to and including infusion two, and after infusion two. Bolded p-values in the legends indicate significant differences (p <0.05) between pre and post infusion two responses. Immunological responses were evaluated for all infused participants (n = 6). Summary statistics and source data are reported as a Source Data file.

Fig. 5. Post-infusion HIV reservoir size dynamics.

a–f Genome-intact proviruses, 5’ defective proviruses and 3’ defective proviruses per million CD4+ T cells were quantified for each participant (n = 6). Four (n = 4) technical replicates were conducted per timepoint for RESIST-001, -003, -005, -006, and -007. Five (n = 5) technical replicates were conducted per timepoint for RESIST-006. Values for each timepoint are plotted as the mean ± standard deviation (SD). Blue shaded areas indicate timepoints up to and including the week of infusion two. Source data are provided as a Source Data file.

Fig. 6. HIV-specific CD8+ T cell clones identified in HST-NEET products expanded in the peripheral blood post infusion two in three participants.

a Two to four T cell clones were present in both the HST-NEET product and peripheral blood TCR repertoires for n = 4 participants. Reported TCRs were sequenced from PBMC samples pulsed with HIV peptides shown in parentheses above respective graphs. b RESIST-006 HST-NEETs and sequenced follow-up PBMC samples did not share any T cell clones. All Gag/Pol Mos1/2 and Nef reactive clones present in RESIST-006 HST-NEETs are shown. c Tracking of T cell clones present in two or more sequenced RESIST-006 follow-up PBMC samples. d T cell clones revealed in three of the highest IFN-γ producing RESIST-003 follow-up PBMCs at weeks 63, 90, and 103. These T cell clones were not present in the RESIST-003 HST-NEETs but had high productive frequencies coinciding with high IFN-γ T cell activation upon stimulation with Gag/Pol Mos1 and Mos2 peptides. Vertical dashed lines (--) indicate infusion one and infusion two, respectively. Source data are provided as a Source Data file.