Targeting the HIV reservoir: chimeric antigen receptor therapy for HIV cure

Abstract

Although antiretroviral therapy (ART) can reduce the viral load in the plasma to undetectable levels in human immunodeficiency virus (HIV)-infected individuals, ART alone cannot completely eliminate HIV due to its integration into the host cell genome to form viral reservoirs. To achieve a functional cure for HIV infection, numerous preclinical and clinical studies are underway to develop innovative immunotherapies to eliminate HIV reservoirs in the absence of ART. Early studies have tested adoptive T-cell therapies in HIV-infected individuals, but their effectiveness was limited. In recent years, with the technological progress and great success of chimeric antigen receptor (CAR) therapy in the treatment of hematological malignancies, CAR therapy has gradually shown its advantages in the field of HIV infection. Many studies have identified a variety of HIV-specific CAR structures and types of cytolytic effector cells. Therefore, CAR therapy may be beneficial for enhancing HIV immunity, achieving HIV control, and eliminating HIV reservoirs, gradually becoming a promising strategy for achieving a functional HIV cure. In this review, we provide an overview of the design of anti-HIV CAR proteins, the cell types of anti-HIV CAR (including CAR T cells, CAR natural killer cells, and CAR-encoding hematopoietic stem/progenitor cells), the clinical application of CAR therapy in HIV infection, and the prospects and challenges in anti-HIV CAR therapy for maintaining viral suppression and eliminating HIV reservoirs.

Introduction

Although antiretroviral therapy (ART) has made tremendous progress in suppressing human immunodeficiency virus (HIV) replication, it cannot eliminate HIV due to viral reservoirs. The main HIV reservoir is a long-lived pool of latently infected cells harboring replication-competent viral DNA,^[1,2] which is the main obstacle for HIV cure. To eradicate the HIV reservoir, several strategies, including "shock and kill", "block and lock", and gene therapy, have been developed,^[3,4] but have failed to reduce the size of the reservoir and to control viral rebound after ART cessation. Thanks to the model of rare patients known as HIV elite controllers in whom HIV replication is spontaneously controlled,^[5,6] therapeutic strategies to get a functional cure (also considered as HIV remission) combining reduction of the size of the reservoir and improving anti-HIV immunity are promising. Immunotherapy has been extensively practiced to achieve HIV remission,^[7] including chimeric antigen receptor (CAR) cell therapy.^[8]

The CAR is a synthetic modular protein aimed at redirecting immune cell reactivity toward a target of interest,^[9] which consists of an extracellular antigen recognition domain and an intracellular signaling domain. The extracellular antigen recognition domain is a single-chain variable fragment (scFv) derived from a monoclonal antibody (mAb), which allows antigen recognition in a major histocompatibility complex (MHC)-unrestricted manner. Intracellular signaling domains include components of activating molecules such as cluster of differentiation 3 (CD3) (e.g. CD3ζ), CD28, 4-1BB (CD137, tumor necrosis factor receptor superfamily member 9), or OX40 (CD134, tumor necrosis factor receptor superfamily member 4) and are designed to increase cell activation.^[10] In the recent past, CAR T cells have demonstrated remarkable success in the treatment of relapsed or refractory B-cell lymphomas,^[11–14] B-cell acute lymphoblastic leukemia^[15–17] and multiple myeloma.^[18–20] T cells modified with a CAR can directly recognize CAR-targeted antigens and can then trigger T-cell activation, proliferation, cytokine secretion, and cytotoxicity to eradicate tumor cells expressing the CAR-specific antigens.^[21] Similar to CAR-expressing T cells, CAR-modified natural killer (NK) cells, CAR-NKT cells, and CAR macrophage-based immunotherapy are also promising strategies in patients with cancer. In addition, stem cell-derived cells expressing CARs have also become an effective approach against fatal malignancies.^[22]

Similar to CAR therapy for tumors, CAR therapy for HIV infection was also developed almost 30 years ago. A previous study demonstrated that CD8⁺ cytotoxic T lymphocytes (CTLs) expressing chimeric proteins combining the extracellular domain of CD4 with the transmembrane and intracellular signaling domains of Fc receptors showed the ability to recognize and kill cells expressing HIV envelope (Env) proteins.^[23] Other early studies have also evaluated the ability of CTLs transduced with chimeric T-cell receptor (TCR) containing CD4 or a single-chain antibody linked to the signaling domain of the TCRζ chain to lyse target cells expressing HIV antigen.^[24–26] Similar to CAR T cells, CAR NK cells can also effectively inhibit HIV replication. In an early study of anti-HIV CARs, Tran et al^[27] showed that CD4-TCRζ NK cells can specifically and efficiently lyse CD4⁺ T cells infected with HIV. In addition, it has been shown that stem cell-based gene therapy with a CAR was feasible and effective in HIV infection.^[28] Recently, remarkable progress in CAR therapy and the leveraging of advances made in oncology have also brought new opportunities for the development of CARs to enhance the HIV-specific immune response for long-term suppression of the reactivated latent HIV reservoir without continuing ART.^[29] In this review, we discuss an overview of recent studies on advances in CAR cells to target the HIV reservoir and mainly focus on the strategies for CAR design, cell types for anti-HIV CAR, and CAR therapy in clinical trials aimed to cure HIV.

CAR T Therapy for HIV Cure

ART can block multiple stages of the life cycle of HIV infection to prevent HIV-infected individuals from developing acquired immunodeficiency syndrome (AIDS). However, due to the persistence of the HIV reservoir in latently infected CD4⁺ T cells and in other cell types such as macrophages, lifelong ART is necessary to prevent viral rebound. Current approaches to reducing the HIV reservoir first require reactivating the latent HIV reservoir cells so that they become visible to the human immune system. A critical second step is killing these cells to reduce the size of the reservoir. This second step is mandatory. Indeed, the strategy known as "shock and kill" uses latency-reversing agents to reverse HIV latency and increase the expression of HIV genes. Its efficacy should rely on the clearance of infected cells by cytolytic immune cells, including CTLs, NK cells, and broadly neutralizing antibodies (bNAbs).^[30] Unfortunately, it has been shown that the CTL response in HIV-infected individuals receiving ART cannot eliminate latently HIV-infected cells even after successful viral reactivation.^[31] However, CTLs from elite controllers can control efficiently HIV replication,^[32] evidencing the interest to develop CAR T cells. Therefore, improving the antiviral function of the immune system might be required for the "kill" strategy to efficiently eliminate the HIV reservoir. T cells modified with a CAR that can specifically recognize and kill cells expressing HIV antigen might be a promising strategy for curing HIV. In addition, CTLs mediate the lysis of HIV-infected cells through MHC-I molecules; however, HIV negative regulatory factor (Nef) has been shown to downregulate the surface expression of MHC-I molecules in infected cells to escape this immune response.^[33] CAR T cells can directly recognize the antigen without MHC-I restriction and therefore could overcome this viral escape mechanism.^[34] Moreover, given the functional heterogeneity and the ability to coordinate the antiviral immunity of HIV-specific CD4⁺ T cells, therapeutic interventions that restore or enhance CD4⁺ T-cell function may be crucial for developing effective HIV cure strategies. A recent study demonstrated that CD4⁺ T cells expressing an HIV-specific CAR can directly control HIV replication and augment the virus-specific CD8⁺ T-cell response, highlighting the therapeutic potential of engineered CD4⁺ T cells as a functional HIV cure.^[35] Furthermore, studies have also demonstrated the trafficking and antiviral activity of CAR T cells against HIV reservoirs.^[36,37] Therefore, T cells modified with a CAR might have the advantage of actively trafficking to tissue reservoirs such as lymph nodes and rectal tissue, offering new possibilities to reduce the HIV reservoir.

Autologous CAR T or CAR NK therapy comprises several steps.^[22] T or NK cells are isolated from the blood of the patient or donor and genetically modified to express CARs. Subsequently, CAR T or CAR NK cells are expanded and injected into HIV-infected individuals. In HIV-infected individuals, CAR T cells can recognize HIV binding sites to kill HIV-infected cells and eliminate HIV reservoirs [Figure 1].^[38]

Figure 1.

CAR-T or CAR-NK cell therapy for HIV infection. T or NK cells are isolated from the blood of the patient or donor and genetically modified to express CARs. CAR-T or CAR-NK cells are expanded and injected into HIV-infected individuals. In HIV-infected individuals, CAR-T cells recognize HIV binding sites to kill HIV-infected cells and eliminate HIV reservoirs. CAR: Chimeric antigen receptor; HIV: Human immunodeficiency virus; NK cell: Natural killer cell.

Design of anti-HIV CAR constructs

As mentioned above, the first generation of anti-HIV CAR in early trials contained a single intracellular signaling domain derived from the CD3ζ chain of the TCR, fused to the extracellular region of CD4 (CD4ζ-CAR), or a scFv derived from a mAb (scFv-CAR) [Figure 2A–C]. However, studies have shown that first-generation CAR T cells have limited in vivo expansion and cytotoxicity and are more prone to apoptosis.^[39] In addition, the extracellular protein domains of the anti-HIV CARs were shown to affect not only the receptor stability and substrate binding affinity but also the cell surface expression and the CAR-mediated killing capacity.^[40] Therefore, second-generation CARs with additional costimulatory molecule domains, such as CD28 or 4-1BB, were generated, which might improve the proliferation, persistence, cytotoxicity, and sustained response of the effector cells [Figure 2C]. Previous data suggested that CD4 CAR containing the 4-1BB zeta signaling domain could control HIV replication more effectively in vitro than the original CD4 CAR, possibly due to its ability to rapidly prevent HIV spread, durably prevent viral rebound, and promote T-cell survival in the absence of antigen.^[41] Third-generation CARs were generated by incorporating multiple costimulatory molecules into secondary-generation CARs, such as 4-1BB or OX40 [Figure 2C]. Liu et al^[42] developed a novel third-generation anti-HIV CAR and found that the newly designed CAR T cells showed superior potency compared to the previously described CD4-CAR. The latest generation of CAR T cells, known as T cells redirected for universal cytokine-mediated killing, added a third stimulatory signal that might lead to the production of cytokines, such as interleukin (IL)-7, IL-12, IL-15, or IL-18 [Figure 2C]. This structure can improve the expansion and persistence of CAR T cells and enhance the activation of T cells and innate immune cells.^[39]

Figure 2.

Schematic representation of CAR therapy to cure HIV. (A) CARs consist of an extracellular antigen-recognition domain, a transmembrane domain and the intracellular signaling domain. Standard second-generation CAR contains a scFv derived from a mAb linked via a transmembrane domain to a costimulatory domain (CD28 or 4-1BB) and a signaling domain (CD3ζ chain). (B) CD4ζ-CAR, the extracellular domain of CD4ζ-CAR is a CD4 molecule. (C) scFv-based CAR, the first-generation CAR contains a single intracellular signaling domain derived from the CD3ζ chain. The second- and third-generation CARs add one or more costimulatory signaling domains within the signaling domain (e.g., CD28, 4-1BB, OX-40). The fourth-generation CAR adds a third stimulatory signal that might lead to the production of cytokines. (D) Dual-CAR, dual CAR simultaneously expresses 4-1BB/CD3-ζ and CD28/CD3-ζ endodomains. (E) Bi-specific CAR, bi-specific CARs contain two linked antigen-binding moieties consisting of two anti-HIV scFv or generate by fusing a CD4 segment to either a bNAb-based scFv or the CRD of a human C-type lectin. (F) Bi-DuoCAR and (G) Tri-DuoCAR, multispecific anti-HIV duoCARs contain a two-molecule CAR architecture to target multiple sites on the HIV Env. (H) Convertible CAR-T, convertible CAR-T consists of a mutant NKG2D which can recognize the mutant ligand domain fused to HIV bNAbs called MicAbody. (I) Universal CAR-NK cell, NK cell expressed an anti-DNP CAR can be redirected by DNP-labeled antibodies to target HIV gp160. bNAb: Broadly neutralizing antibody; CAR: Chimeric antigen receptor; CD3: Cluster of differentiation 3; CRD: Carbohydrate recognition domain; DNP: 2,4-dinitrophenyl; HIV: Human immunodeficiency virus; mAb: Monoclonal antibody; NK cell: Natural killer cell; NKG2D: Natural killer group 2D receptor; scFv: Single-chain variable fragment.

Improved CAR design to target the HIV reservoir

In addition to the third-generation anti-HIV CAR that added two costimulatory domains to a single CD3ζ, a dual CAR T cell that simultaneously expressed two CD4-based CARs fused to 4-1BB/CD3-ζ and CD28/CD3-ζ endodomains has been developed [Figure 2D].^[43] Compared to CD28-, 4-1BB- and third-generation CARs, dual-CAR T cells exhibit improved proliferation and effector functions in vivo. Moreover, dual-CAR T cells coexpressing the C34-C-X-C motif chemokine receptor 4 (CXCR4) fusion inhibitor can improve the survival and effector function of CAR T cells during early HIV infection, reduce HIV viremia and accelerate HIV suppression in a bone marrow, liver, thymus humanized mouse model.^[43] Notably, HIV-resistant dual-CAR T cells might also reduce the HIV burden in a variety of cell types and tissues, including long-lived memory CD4⁺ T cells, which suggests that CAR T-cell therapy targets the HIV reservoir.^[43]

To increase the sensitivity of CAR T cells against HIV-infected cells and minimize the chance for viral escape, a bispecific CAR was designed to target two distinct highly conserved nonoverlapping Env determinants [Figure 2E]. Liu et al^[44] described a novel bispecific CAR containing a CD4 segment linked to a scFv of the 17b human mAb recognizing a highly conserved CD4-induced epitope on gp120, including CD4-35-17b CAR (a longer polypeptide linker between the CD4 and 17b moieties) and CD4-10-17b CAR (a shorter polypeptide linker between the CD4 and 17b moieties). They found that the CD4-10-17b bispecific CAR displayed enhanced suppressive potency for diverse HIV isolates compared to the CD4 CAR. Ghanem et al^[45] also demonstrated that bispecific CD4-lectin CARs might enhance the potency against HIV isolates by targeting two distinct highly conserved Env determinants. In addition, to further improve the bispecific CAR, the tri-specific CAR has been developed by adding a third targeting moiety against a distinct conserved Env determinant, i.e. a polypeptide sequence derived from the N-terminus of the HIV coreceptor C-C chemokine receptor 5 (CCR5), which provided high anti-HIV potency, offering promise for the tri-specific construct in therapeutic strategies for durable ART-free HIV remission.^[46]

To achieve durable suppression of HIV viremia without daily ART, HIV bNAbs binding with different epitopes of HIV Env have been developed and become major weapons against HIV [Figure 2A,C].^[47,48] bNAb-based CARs also show potent cytolysis of HIV-infected cells.^[49] Seven novel CARs based on diverse bNAb types have been explored against HIV, including 10E8, 3BNC117, PG9, PGT126, PGT128, VRC01, and X5. All of these CARs showed conformationally relevant expression on the cell surface and had the ability to recognize and kill HIV-infected cells.^[50] Other studies also demonstrated that bNAb-based scFvs fused to second- and third-generation CAR signaling domains have the ability to target HIV-infected cells and might effectively eradicate HIV reactivated from latent CD4⁺ T cells from HIV-infected individuals receiving suppressive ART.^[42,51] In addition, Liu et al^[52] also found that bNAb-derived CAR T cells could reduce the HIV reservoir in HIV-infected individuals, which might help in the effort to define a strategy to cure HIV.

However, a single bNAb cannot neutralize all HIV isolates, and bNAb-based CARs might require further engineering to increase their therapeutic effectiveness.^[53] Multispecific anti-HIV duoCARs based on a two-molecule CAR architecture have been developed, which can target multiple sites on the HIV Env [Figure 2F,G].^[53] It has been shown that multispecific anti-HIV duoCARs could potently eliminate peripheral blood mononuclear cells infected with bNAb-resistant HIV strains, achieve long-term HIV control and prevent the loss of CD4⁺ T cells in a humanized Nod-SCID-IL2Rgamma mouse model.^[53] Moreover, eliminating HIV-infected cells in anatomical sites is important for reducing the HIV reservoir. Another study showed that anti-HIV duoCAR T cells injected into humanized mice could migrate to the site of active HIV infection in the mouse spleen and potently suppress HIV replication.^[54] In addition, monocytes and macrophages might be infected with HIV and contribute to the latent HIV reservoir. This study also showed that anti-HIV duoCAR T cells could effectively sense and kill HIV-infected monocytes and macrophages.^[54] Further studies on anti-HIV duoCAR T cells would facilitate further understanding of the ability of redirected T cells to eliminate the HIV reservoir.

A universal CAR-T-cell platform was developed to expand the breadth of the killing of HIV-infected cells and minimize the emergence of viral resistance, which was termed convertible CAR™-T (cCAR-T) cells [Figure 2H].^[55] This platform contains an engineered extracellular domain of the Natural Killer Group 2D receptor (NKG2D) variant fused to the intracellular 4-1BB and CD3ζ cosignaling domains to generate a mutated NKG2D CAR. The mutated α1–α2 domains of major histocompatibility complex class I chain-related (MIC) ligands, including MICA, MICB, and UL16 binding proteins, can fuse to HIV bNAbs to generate a molecule called MicAbody.^[55] The convertible CAR-T cells, based on the binding of a mutant NKG2D receptor to an orthogonal MIC ligand fused to HIV bNAbs (MicAbody), can effectively kill HIV-infected cells and reactivated reservoir cells in the blood of HIV-infected individuals on ART.^[55] Therefore, an adaptable CAR T-cell platform might play an important role in the effective killing of HIV-infected cells and the reduction of HIV reservoirs.

CAR NK Cells for HIV Cure

In addition to CAR T cells, CAR NK cell-based therapies have emerged as a promising approach to eradicate HIV-infected cells, which have the ability of antigen recognition.^[56] Extensive studies have shown that in addition to T cells, NK cells also play an important role in the fight against HIV infection.^[57,58] During HIV infection, the virus can release NKG2D ligand into the serum, which indirectly contributes to the exhaustion of NK cells.^[59] Of note, the results reported by Mavilio et al^[60] showed that in HIV-infected individuals with viremia, CD56(-) NK cells were considerably amplified and secreted fewer cytokines associated with the initiation of antiviral immune responses compared to CD56(+) NK cells, and the cytotoxic function of this subset was also severely compromised, which is probably linked to the persistent replication of the virus. In addition, the loss of control of viral replication in HIV controllers was correlated with a decrease in NK-cell responses.^[61] Compared with CAR T cells, CAR NK cells have several advantages, including not requiring human leukocyte antigen (HLA) compliance, therefore reducing the risk of graft-versus-host disease; reduced risk of cytokine release syndrome; and conferring tumor-restricted effects.^[62] Considering the advantages of CAR NK cell therapy and the role of NK cells in HIV infection, there has been interest in developing CAR NK cell therapy in the HIV cure strategy. The most commonly studied NK cell line used to evaluate CARs is the NK-92 cell line.^[63] Other NK cell lines have also been established, including NKG, HANK-1, NK-YS, YT, YTS cells, and NKL cells.^[64] The majority of studies to express CARs in NK-92 cells have focused on the use of first-generation CAR constructs that contain CD3ζ for signal transduction and various extracellular domains involved in targeting specific infected or malignant cells, such as CD19, CD20, Erb2, GD2, and CD138.^[63]

Previously, CD4ζ-expressing induced pluripotent stem cell-derived NK cells were proven to inhibit HIV replication in CD4⁺ T cells more effectively than unmodified cells did in vitro. However, in a subsequent humanized mouse model, the investigators did not find a significant effect of CD4ζ expression on HIV replication.^[65] To improve NK-cell function, replacing T-cell intracellular domains with NK-cell signaling domains may be an effective approach; for example, NKG2D-DAP10-TCRζ CAR-modified NK cells increase the secretion of a variety of cytokines.^[66] In addition, a universal CAR-NK cell has been developed that can recognize various epitopes of gp160 from different HIV subtypes [Figure 2I].^[67] Instead of directly targeting gp160, this CAR recognizes the small molecule ligand 2,4-dinitrophenyl (DNP), which can then be redirected to target different epitopes of HIV gp160 variants using DNP-conjugated antibodies as adaptor molecules. This study showed that these anti-DNP CAR-NK cells are effective in killing HIV-infected cell lines expressing gp160 of subtypes B and C. Moreover, this study also found that adaptor molecules targeting membrane-distal epitopes were more likely to activate anti-DNP CAR-NK cells against gp160⁺ target cells than those targeting membrane-proximal epitopes. Therefore, the CAR-NK cell platform combined with numerous antibodies could potentially be used to overcome HIV diversity and enhance the eradication of HIV reservoirs, which is essential for curing HIV.^[67]

CAR-Encoding HSPCs for HIV Cure

Allogeneic hematopoietic stem cell transplantation is the only treatment known to achieve a complete HIV cure.^[68] In addition to hematopoietic stem/progenitor cell (HSPC)-based gene therapy, HSPC-based CAR therapy has emerged as a powerful immunotherapy for HIV infection, which has demonstrated successful long-term engraftment and production of anti-HIV CAR cells.^[69] The modification of HSPCs with an HIV-specific CD4ζ CAR can allow them to differentiate into HIV-specific T cells and other cells, which might migrate to multiple anatomic sites and suppress HIV replication in vivo.^[70] Another study also showed that HSPC-derived cells could differentiate into functional T or NK cells to suppress HIV replication in vivo in humanized mice.^[28] Zhen et al^[71] reported the development of second-generation CD4-based CARs against HIV using a hematopoietic stem cell (HSC)-based approach. They indicated that this superior CAR showed better HSC differentiation and HIV suppression and fewer deleterious functions than CD4 CAR. In addition, by using CD4 CAR-specific immunohistochemistry (IHC)-based assays, Barber-Axthelm et al^[72] quantified the trafficking and persistence of HSC-derived CAR-expressing cells within HIV reservoirs in nonhuman primates infected with simian/human immunodeficiency virus. They showed that the HSC-derived CAR cells could traffic to tissue-associated HIV reservoirs, display multilineage engraftment, and persist for nearly two years in the lymphoid germinal centers, the brain, and the gastrointestinal tract. These results strongly suggest that HSPC-based CAR therapy may be feasible and effective in eliminating HIV reservoirs and promoting a functional cure for HIV infection.

Clinical Trials of CAR T cell Therapy in HIV Infection

The history of the clinical use of CAR T cells for HIV treatment dates back as far as 1994 when Walker et al^[73] used CD4/CD3ζ-modified CD4⁺ and CD8⁺ T cells to treat identical twins with discordant HIV infection. Despite the lack of satisfactory efficacy, the study demonstrates that CAR T cell therapy does not pose significant safety concerns. In the same view, Mitsuyasu et al^[36] investigated the efficacy of CD4ζ gene-modified CD4⁺ T cells and CD8⁺ T cells in targeting HIV and reported no statistically significant changes in mean plasma HIV or blood provirus DNA values in 24 HIV-positive subjects, but the high persistence of such modified T cells provided the basis for subsequent studies. In addition, in another phase II clinical trial targeting HIV reservoirs, no significant changes in viral reservoirs were observed after CAR T cell therapy in ART-treated HIV-infected individuals with no viremia.^[74]

In the wake of the poor efficacy of previous CD4ζ-based CAR T cells, researchers have begun to improve CAR constructs in search of more satisfactory therapeutic outcomes. After a series of updated iterations, a team recently developed dual CAR T cells expressing both 4-1BB/CD3-ζ and CD28/CD3-ζ endodomains.^[43] Despite the failure to control viremia after ART cessation, this double CAR T cell has a strong proliferative capacity and protects a large amount of memory and CCR5 CD4⁺ T cells from HIV-induced depletion in BLT humanized mice. Furthermore, given the ability of bNAbs to broadly neutralize HIV strains, there has been a wave of interest in constructing bNAb-derived CAR-T-cell therapies.^[75] More recently, a phase I clinical trial reported the efficacy of infusion of bNAb-derived CAR T cells on viral rebound after discontinuation of ART and indicated that the treatment was safe and without serious adverse events.^[52] The investigators sorted CD8⁺ T cells from HIV-infected patients and transduced the cells with the VRC01-28BBζ-shPTL transgene to express a gp120-specific bNAb-derived CAR (containing CD28 and 4-1BB), followed by insertion of a combination of sh-PD-1, sh-Lag-3 and sh-Tim-3 in the vector to increase the durability of the CAR T cells in vivo. Of the 14 subjects, all eventually experienced viral rebound and developed drug-resistant strains, but it is worth noting that there was a prominent reduction in cell-associated viral RNA and intact provirus after administration of CAR T-cell therapy, which may provide some evidence demonstrating the potential of this approach for HIV functional cure.

In a further drive to pursue a durable remission for HIV, a team of investigators recently developed novel HIV-specific CAR T cells that target HIV in T follicular helper cells reservoirs by expressing the follicular homing receptor (CXCR5).^[76] The investigators selected DRAGA (HLA-A2. HLA-DR4. RAG1 KO. IL-2Rγc KO. NOD) mice, which produce class-switching specific antibodies and can develop a human immune system by injecting human HSCs.^[77] Regretfully, no differences in HIV viral load and CD4⁺ T-cell counts in DRAGA mice infected by humanized HIV in the intervention group were found in the study. In addition, an open phase I/IIa IND-enabling study is evaluating the safety and efficacy of HIV DuoCAR T-cell therapy, and data regarding its effectiveness against reservoirs are anticipated.^[76] Currently, there are many promising clinical trials underway utilizing CAR T cell therapy to treat HIV [Table 1].

Table 1.

Ongoing and completed clinical trials of CAR therapy in the treatment of HIV infection.

NCT/ChiCTR number	Participants (n)	CAR construct	Other treatments or dispositions	Primary outcomes	Estimated/actual study completion date (status)
NCT03617198	12	Autologous CD4 CAR+CCR5 ZFN T cells	ATI: One group was performed 24 h after the intervention and the second group after eight weeks.	Number of subjects with treatment-related adverse events	December 2027 (active, not recruiting)
NCT03980691	4	CAR-T or TCR-T cell therapy	Combined with chidamide	Incidence of treatment-associated adverse events	May 31, 2020 (completed)
NCT03240328	40	HIV-1 specific CAR-T cells	Stop ART when conditions are met	Incidence of treatment-associated adverse events of CAR-T cell therapy	December 31, 2030 (recruiting)
NCT03666871	30	Autologous CD4+ T cells with or without ex vivo modification of the CCR5 gene by zinc finger nucleases	Pre-treated with cyclophosphamide at a dose of 1 g/m2 before infusion	The difference in the magnitude of change in IUPM from before infusion to 24 months after infusion and the proportion of participants who experience a grade ≥3 adverse event	January 31, 2024 (active, not recruiting)
NCT04648046	18	Autologous CD4+ and CD8+ T cells transduced with a lentiviral vector encoding bispecific anti-gp120 CAR molecules (LVgp120duoCAR-T)	ATI + non-ablative conditioning with cyclophosphamide	Number of participants reporting a new Grade 3 or greater adverse event one year after infusion, Number of participants achieving post-treatment control within 36 weeks of product administration	December 31, 2025 (recruiting)
NCT04863066	8	CTL-based autologous CAR-T cells expressing scFV	–	Evaluation of the safety of CAR-T-cell treatment	October 1, 2022 (unknown status)
NCT05784415	30	CD19-directed CAR-T cell	–	Rate of toxicities related to CAR19 therapy	August 31, 2025 (active, not recruiting)
ChiCTR2100047544	9	Anti-gp120-CAR-T cells	ART (TDF+3TC+EFV)	Adverse effects rate	December 31, 2022 (recruiting)

3TC: Lamivudine; ART: Antiretroviral therapy; ATI: Analytic treatment interruption; CAR: Chimeric antigen receptor; CCR5: C-C chemokine receptor 5; CTL: Cytotoxic T lymphocyte; EFV: Efavirenz; HIV: Human immunodeficiency virus; IUPM: Infectious units per million; scFv: Single-chain variable fragment; TCR: T-cell receptor; TDF: Tenofovir disoproxil fumarate; ZFN: Zinc finger nucleases; –: None.

Prospects and Challenges in CAR Therapy for HIV Infection

CAR therapy remains a widely promising HIV therapy with considerable opportunity for improvement despite its underperformance in current preclinical and clinical trials. Several issues are still hindering the advancement of CAR therapy that need to be overcome.

First, for CAR T cell therapy, to know how the modified cells are delivered to the sites of the HIV reservoir, how they can continue to proliferate in the tissular microenvironment are a prerequisite for the therapy to work. A recent study described the modification of HSCs to carry a truncated CD4-based CAR (D1D2CAR) containing 4-1BB that eventually differentiated into CAR T cells and showed better persistence and greater suppression of viral replication during ART.^[71] Previously, not only can CD4 CAR mediate HIV infection, but IL-16 binds to the D4 domain in CD4 and this domain is also in close proximity to the domains involved in CD4–CD4 dimerization, which may cause cross-reactivity and nonspecific signaling in CD4 CAR.^[78] Excitingly, since D1D2CAR (D4 domain removed) can circumvent the above problems, there is a low risk of off-target effects. In another animal experiment, Barber-Axthelm et al^[72] observed that HSPC-derived CD4 CAR cells could traffic to multiple tissue-associated viral reservoirs in macaque models of HIV infection for long-term multilineage engraftment, which may yield new insight into the future elimination of hard-to-reach HIV reservoirs in multiple tissues.

Second, the susceptibility of CD4-based constructs to HIV infection is also a major hurdle in CAR T cell therapy for HIV.^[79] To overcome this obstacle, one group proposed using zinc-finger nucleases to modify both CCR5 and CXCR4 coreceptors in primary human CD4⁺ T cells and demonstrated that in a humanized mouse model, these coreceptor-negative cells can be normally engrafted and trafficked and protected from infection by HIV strains using CCR5 and CXCR4.^[80] Alternatively, the construction of CARs using bNAbs specific for HIV envelope glycoproteins may provide another approach.^[51] Some researchers have knocked the CAR of bNAb into the CCR5 gene by clustered regularly interspaced short palindromic repeats technology, the results of which show that CAR T cells have specific activation and killing functions on HIV-infected cells and indirectly protect CAR T cells from HIV infection.

Last, two concerns remain. The development of CAR cells at a large scale has to be improved. The fast development in oncology is a good example to follow. The safety of CAR cell therapies also remains a continuing concern. In contrast to the high tumor-specific antigen density in leukemia, CAR cells are less likely to generate cytokine release syndrome when treating HIV-infected patients with no or low viremia and are therefore safer; although this is corroborated by the safety results of most previous clinical trials, vigilance is still required regarding the occurrence of serious adverse events.^[81] Moreover, while CAR cells target latent HIV-infected cells, there is parallel caution against the off-target effect of CAR cells attacking normal cells.

Conclusion

Collectively, the eradication of HIV reservoirs to achieve a complete HIV cure will continue to be a concerted effort by a wide range of researchers for some time to come. The reservoirs of HIV are distributed throughout the body, and a recent study found that HIV can exist as a persistent reservoir in microglia in the brain,^[82] so eliminating HIV reservoirs using CAR T or CAR NK cells requires that these modified cells be able to access and persist in many sites that are difficult to reach with conventional drugs or latent reversal agents. In addition, improving the construction of CARs or combining them with other drugs (e.g., bNAbs) to overcome HIV escape mutations or even developing other new CAR cell therapies to achieve long-term remission without ART will remain a goal to be pursued.

Funding

This project is financially supported by grants from the National Key R&D Program of China (Nos. 2021YFC2301900 and 2021YFC2301905), the National Natural Science Foundation of China (Nos. NSFC, 81974303 and 82072271), Beijing Natural Science Foundation (Nos. L222068 and Z220018), the High-Level Public Health Specialized Talents Project of Beijing Municipal Health Commission (Nos. 2022-2-018 and 2022-1-007), the Climbing the peak (Dengfeng) Talent Training Program of Beijing Hospitals Authority (Nos. DFL20191701), the Beijing Health Technologies Promotion Program (BHTPP202002) and Beijing Key Laboratory for HIV/AIDS Research (No. BZ0089). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Conflicts of interest

None.