Long-term persistence and function of hematopoietic stem cell-derived chimeric antigen receptor T cells in a nonhuman primate model of HIV/AIDS

Abstract

Chimeric Antigen Receptor (CAR) T-cells have emerged as a powerful immunotherapy for various forms of cancer and show promise in treating HIV-1 infection. However, significant limitations are persistence and whether peripheral T cell-based products can respond to malignant or infected cells that may reappear months or years after treatment remains unclear. Hematopoietic Stem/Progenitor Cells (HSPCs) are capable of long-term engraftment and have the potential to overcome these limitations. Here, we report the use of a protective CD4 chimeric antigen receptor (C46CD4CAR) to redirect HSPC-derived T-cells against simian/human immunodeficiency virus (SHIV) infection in pigtail macaques. CAR-containing cells persisted for more than 2 years without any measurable toxicity and were capable of multilineage engraftment. Combination antiretroviral therapy (cART) treatment followed by cART withdrawal resulted in lower viral rebound in CAR animals relative to controls, and demonstrated an immune memory-like response. We found CAR-expressing cells in multiple lymphoid tissues, decreased tissue-associated SHIV RNA levels, and substantially higher CD4/CD8 ratios in the gut as compared to controls. These results show that HSPC-derived CAR T-cells are capable of long-term engraftment and immune surveillance. This study demonstrates for the first time the safety and feasibility of HSPC-based CAR therapy in a large animal preclinical model.

Fig 1. C46CD4CAR cells resist HIV infection and respond to cognate antigen.

A) Lentiviral vector schematics. B) Jurkat cells were mock transduced, or transduced with CD4CAR or C46CD4CAR, then infected with HIV and cultured for 3 days. Intracellular expression of HIV-1 Gag was measured by flow cytometry using KC57 antibody. C) Pigtail macaque T cells were activated and transduced with either control or C46CD4CAR vector. 2 days following transduction, cells were stimulated with either uninfected control T1 cells or HIV-infected T1 cells. Intracellular cytokines were measured by flow analysis.

Fig 2. Engraftment of C46CD4CAR modified HSPCs is multilineage.

A) Lentivirus gene marking in transplanted animals following autologous transplantation, measured by Taqman. B) Detection of C46CD4CAR-modified cells by flow cytometry with anti-human CD4 antibody clone 13B8.2, which does not detect Macaca nemestrina CD4. C) Percent of C46CD4CAR cells among total peripheral PBMCs, at approximately 28 days post-transplant. D) Multilineage engraftment of lentivirus-modified cells in peripheral blood was measured from control and C46CD4CAR animals. CAR+ cells were first gated on huCD4+ cells. T cells are defined as CD45+CD3+, CD3-CAR+ (blue gate) cells are gated and NK cells are defined as CD45+CD3-CD2+NKG2A+, B cells are defined as CD45+CD3-CD20+, macrophage (MQ)&monocytes are defined as CD45+CD3-CD14+. One representative control (top panels) and CAR animal (bottom panels) are shown.

Fig 3. Reduced viral rebound in C46CD4CAR animals.

A) CD4-CAR and Δzeta control animals were challenged with SHIV-1157ipd3N4 (“SHIV-C”) via the intravenous route. Approximately 24 weeks later, cART was initiated and administered for 28 weeks. Following cART withdrawal, animals were monitored for approximately 15 weeks prior to necropsy. B) Plasma viral load was monitored longitudinally after SHIV challenge. Dashed lines indicate limit of detection (LOD) of 30 copies/mL. Closed circles and open circles indicate beginning and end of cART, respectively. C) Average viral load before cART and after cART withdrawal, and log reduction of plasma viremia following post-cART viral rebound, relative to comparable time points during primary infection. Average viral load calculations are described in Materials and Methods.

Fig 4. C46CD4CAR cells expand in response to SHIV antigen in vivo.

A) Lentiviral gene marking was measured by Taqman from peripheral blood at the indicated time points from C46CD4CAR (solid lines) and C46CD4CAR Δzeta transplanted animals (dashed lines). (B-E) Viral load (left Y axis, blue) and percentage of CAR cells in PBMCs (right Y axis, red) were measured longitudinally from peripheral blood from Control1 (B), Control 2 (C), CAR 1 (D), and CAR 2 (E) animals.

Fig 5. C46CD4CAR T cells develop into effector cells in response to viral replication.

A) Percentages of unmarked (CAR-) naïve, memory, and effector T cells and C46CD4CAR+ T cells from peripheral blood from one representative animal, collected at the indicated time points. (B-E) Quantitation of naïve, memory, and effector subsets from unmarked T cells and gene modified T cells in peripheral blood from Control 1 (B), Control 2 (C), CAR 1 (D) and CAR 2 (E) animals.

Fig 6. C46CD4CAR cells protect CD4⁺ T-cells and decrease viral load in multiple tissues.

(A-C) GI biopsies were collected from colon and duodenum/jejunum from C46CD4CAR and control animals prior to SHIV infection, and after infection and cART withdrawal. Shown are CD4/8 ratio (A), %CD4⁺ T cells (B) and %CD4⁺ effector memory T-cells (C). D) At necropsy, Taqman was used to measure lentivirus gene marking from lymphoid tissues (spleen and mesenteric, axillary, inguinal, and submandibular lymph nodes), gastrointestinal tract (duodenum, jejunum, ileum, cecum, colon, rectum) and brain tissues (hippocampus, basal ganglia, thalamus, parietal cortex, cerebellum). E) Normalized SHIV RNA copy number from same tissues as in (D). * p<0.01, **p<0.001, ***p<0.0001 by Mann-Whitney test.