Anti-BCMA chimeric antigen receptors with fully human heavy-chain-only antigen recognition domains

Lam, Norris; Trinklein, Nathan D.; Buelow, Benjamin; Patterson, George H.; Ojha, Namrata; Kochenderfer, James N.

doi:10.1038/s41467-019-14119-9

Download PDF

Article
Open access
Published: 15 January 2020

Anti-BCMA chimeric antigen receptors with fully human heavy-chain-only antigen recognition domains

Nature Communications volume 11, Article number: 283 (2020) Cite this article

11k Accesses
68 Citations
19 Altmetric
Metrics details

Subjects

An Author Correction to this article was published on 09 March 2020

This article has been updated

Abstract

Chimeric antigen receptor (CAR)-expressing T cells targeting B-cell maturation antigen (BCMA) have activity against multiple myeloma, but improvements in anti-BCMA CARs are needed. We demonstrated recipient anti-CAR T-cell responses against a murine single-chain variable fragment (scFv) used clinically in anti-BCMA CARs. To bypass potential anti-CAR immunogenicity and to reduce CAR binding domain size, here we designed CARs with antigen-recognition domains consisting of only a fully human heavy-chain variable domain without a light-chain domain. A CAR designated FHVH33-CD8BBZ contains a fully human heavy-chain variable domain (FHVH) plus 4-1BB and CD3ζ domains. T cells expressing FHVH33-CD8BBZ exhibit similar cytokine release, degranulation, and mouse tumor eradication as a CAR that is identical except for substitution of a scFv for FHVH33. Inclusion of 4-1BB is critical for reducing activation-induced cell death and promoting survival of T cells expressing FHVH33-containing CARs. Our results indicate that heavy-chain-only anti-BCMA CARs are suitable for evaluation in a clinical trial.

Chimeric antigen receptor T cells targeting FcRH5 provide robust tumour-specific responses in murine xenograft models of multiple myeloma

Article Open access 20 June 2023

Systematically optimized BCMA/CS1 bispecific CAR-T cells robustly control heterogeneous multiple myeloma

Article Open access 08 May 2020

Antigen-independent activation enhances the efficacy of 4-1BB-costimulated CD22 CAR T cells

Article 22 April 2021

Introduction

Chimeric antigen receptors (CAR) are artificial proteins that include antigen-recognition moieties, T-cell activation domains such as CD3ζ, and costimulatory domains such as 4-1BB and CD28^{1,2,3,4,5,6,7,8}. B-cell maturation antigen (BCMA) is expressed by normal and malignant plasma cells and a small subset of B cells^9,10,11. This restricted expression pattern makes BCMA a good target antigen for immunotherapies.

Multiple myeloma is an almost always incurable malignancy of plasma cells^12,13,14,15. We and others have used CAR T cells targeting BCMA to treat multiple myeloma in clinical trials^16,17,18,19. These clinical trials showed that 81–88% of patients had objective anti-myeloma responses after the treatment with anti-BCMA CAR T cells, but most anti-myeloma responses were not permanent^17,18,19. Clearly, improvements in CAR T-cell therapies for multiple myeloma are needed.

Recipient anti-CAR immune responses that might limit CAR T-cell survival have been demonstrated against CAR T cells in clinical trials^20,21,22. The 11D5-3 single-chain variable fragment (scFv) that has been incorporated into CARs used in several clinical trials^16,17,19 is derived from a murine monoclonal antibody¹¹. One way to potentially reduce immunogenicity of CAR binding domains is to use human instead of murine sequences^23,24. Another potential way to decrease immunogenicity of CARs is to simplify the structure of the CAR’s antigen-binding domain by using heavy-chain-only binding domains²⁵.

Immunoglobulin-like molecules with antigen-binding domains made up of only heavy chains without light chains were first described in camelids and cartilaginous fish^25,26,27,28. Binding domains made up of only a single immunoglobulin heavy-chain variable region domain have been reported to exhibit strong and specific antigen binding^{25,29,30,31,32,33}. Unlike a scFv, single heavy-chain-only binding domains have no need for a potentially immunogenic linker to connect the heavy chain to the light chain. Because human heavy-chain-only antigen-recognition domains have no linker or light chain, two potentially immunogenic junctions that must be included in scFvs are eliminated²⁵. The smaller size of a heavy-chain-only antigen-recognition domain might have steric advantages over the larger scFv in reaching certain small or partially hidden cell surface antigens²⁵. The smaller size of heavy-chain-only antigen-recognition domains is also an advantage for expression from gene therapy vectors, since decreasing the size of transgenes increases the titer of viral gene therapy vectors³⁴. Compared with designs that utilize two scFvs, using two heavy-chain-only antigen-recognition domains simplifies design of bispecific CAR constructs capable of recognizing two antigens^35,36.

The advantages of reduced size and potentially reduced immunogenicity of heavy-chain-only binding domains compared with scFvs led us to design and test the novel heavy-chain-only anti-BCMA CARs. T cells expressing a CAR with a heavy-chain-only antigen-recognition domain exhibit BCMA-specific functions equivalent to those of T cells expressing a CAR with the clinically proven 11D5-3-derived scFv. Including a 4-1BB costimulatory domain is especially important for improving survival of T cells expressing heavy-chain-only CARs. These results indicate that heavy-chain-only binding domains might offer general advantages for CARs targeting antigens other than BCMA.

Results

Immunogenicity of murine scFv-containing CARs

We attempted to elicit T-cell immune responses against the murine 11D5-3 scFv by methods similar to those previously used by others^37,38. We prepared a nonsignaling CAR called 11D5-3-NS. The extracellular and transmembrane sequences of 11D5-3-NS were identical to the corresponding sequences of 11D5-3-CD828Z, a CAR that has been previously tested in a clinical trial¹⁷. We transduced T cells from patients on the previous clinical trial of 11D5-3-CD828Z-expressing T cells¹⁷ with gamma-retroviral vectors encoding 11D5-3-NS (nine cases) or 11D5-3-CD828Z (one case). We used these transduced patient T cells as antigen-presenting cells to stimulate autologous peripheral blood mononuclear cells (PBMC) in culture. Two weekly stimulations of the PBMC with irradiated 11D5-3-expressing T cells were conducted. Next, we cultured the stimulated PBMC overnight with either autologous 11D5-3-expressing T cells or autologous negative-control T cells expressing the nerve growth factor receptor (NGFR) gene. We then performed an interferon gamma (IFNγ) enzyme-linked immunosorbent assay (ELISA) on the culture supernatants. We found specific PBMC reactivity against the 11D5-3-CAR in five out of ten patients evaluated. CAR-specific reactivity by the PBMC was defined as IFNγ release that was three-fold or more higher with 11D5-3-expressing T-cell targets compared with IFNγ release with negative-control NGFR-expressing T-cell targets. Another requirement for CAR-specific reactivity was a minimum of 150 pg/mL of IFNγ release by PBMC in response to 11D5-3-expressing T-cell targets (Fig. 1a).

**Fig. 1: A murine anti-BCMA CAR can be immunogenic.**

To define the exact target of the anti-11D5-3 T-cell responses detected in the ELISA assays, we prepared 15-mer peptides that covered the whole sequence of the 11D5-3 scFv. We pulsed autologous dendritic cells with these peptides and used the peptide-pulsed dendritic cells to assay for 11D5-3-specific T cells among patient PBMC that had previously been stimulated two times in culture with autologous 11D5-3-expressing T cells. We detected peptide-specific immune responses by intracellular cytokine staining (ICCS) flow cytometry for IFNγ (Fig. 1b). We found 11D5-3-peptide-specific T cells in two out of two patients evaluated with this approach; this approach could only be used in two patients because an insufficient number of cells were available from other patients to conduct the assay. Peptides that elicited T-cell responses were from the complementarity determining region 3 plus framework 3 region of the 11D5-3 heavy chain in one patient and from the complementarity determining region 1 plus framework 1 region of the 11D5-3 light chain in the other patient.

Design of CARs with heavy-chain-only binding domains

The potential for decreased immunogenicity of heavy-chain-only binding domains compared with scFvs and other advantages of CARs with heavy-chain-only binding domains led us to develop fully human heavy-chain-only CARs targeting BCMA. We previously reported novel rats transgenic for the complete functional human heavy-chain variable gene repertoire³⁰. The rats are referred to as Unirats³⁰. Unirats produce Uniabs that consist of two covalently linked antibody heavy chains. Unirats were immunized with human BCMA protein in adjuvant³⁰. RNA was isolated from draining lymph nodes, and high-abundance heavy-chain variable region sequences were expressed as proteins^30,39. These heavy-chain variable region domains were tested for BCMA recognition^30,39. Sequences of four heavy-chain variable domains with the ability to specifically bind BCMA were used as antigen-recognition domains in novel CARs⁴⁰. The CAR binding domains consisted of only a single fully human heavy-chain variable region. The four heavy-chain-only sequences incorporated into CARs were designated FHVH (fully human heavy-chain variable domain) 74, 32, 33, and 93⁴⁰. In initial experiments, we constructed gamma-retroviral vectors encoding CARs that each contained one of the four heavy-chain-only domains. The CAR sequences started at the N-terminus with each FHVH domain followed by CD8α hinge and transmembrane domains, a 4-1BB costimulatory domain and a CD3ζ T-cell activation domain. A diagram of FHVH33-CD8BBZ is shown (Fig. 1c). T-cell surface expression of CARs containing FHVH74 was inferior to CARs containing the other three FHVH domains when 4-1BB was included in the CAR (Supplementary Fig. 1). CARs containing FHVH33, 32, and 93 were similar in expression and function, but FHVH33 was slightly superior in expression on T cells, so it was selected for further development. Supplementary Table 1 shows BCMA-specific IFNγ production by CARs with each FHVH domain.

We compared FHVH33-CD8BBZ to 11D5-3-CD8BBZ, a CAR that has the same sequence as FHVH33-CD8BBZ except the murine 11D5-3 scFv was included instead of FHVH33 (Fig. 1d). We also compared FHVH33-CD8BBZ to FHVH33-CD828Z that has the same sequence as FHVH33-CD8BBZ except for replacement of 4-1BB with CD28 (Fig. 1e).

Expression and affinity of 11D5-3-CD8BBZ and FHVH33-CD8BBZ

We used a BCMA-immunoglobulin fragment crystallizable (Fc)-phycoerythrin (PE) reagent to assess CAR expression on the surface of T cells. The percentages of T cells expressing 11D5-3-CD8BBZ versus FHVH33-CD8BBZ were not different 5 days after transduction (Fig. 2a). The median %CAR⁺ T cells was 57.2% (range 33.2–71.4%) for 11D5-3-CD8BBZ and 61.9% (range 49.0–74.3%) for FHVH33-CD8BBZ (n = 6; P = not significant, N.S.). Among CAR⁺ T cells after optimal BCMA-Fc-PE staining, there was no difference in median fluorescence intensity (MFI) between 11D5-3-CD8BBZ and FHVH33-CD8BBZ (Fig. 2b). We compared the relative affinity of 11D5-3-CD8BBZ and FHVH33-CD8BBZ by using methods adapted from previous work of others (Fig. 2c)^41,42. We calculated dissociation constants (K_D) by nonlinear regression. We found no difference in the K_D of 11D5-3-CD8BBZ versus FHVH33-CD8BBZ (Fig. 2d).

**Fig. 2: Murine scFv 11D5-3 versus fully human heavy-chain domain FHVH33.**

Function of 11D5-3-CD8BBZ T cells versus FHVH33-CD8BBZ T cells

We did not find a difference in degranulation between 11D5-3-CD8BBZ and FHVH33-CD8BBZ except a small difference in CD8⁺ T-cell degranulation, when T cells were assayed with one out of two target cell lines tested (Fig. 2e–h). As with degranulation, we did not find a difference in cytotoxicity when 11D5-3-CD8BBZ and FHVH33-CD8BBZ were compared (Fig. 2i).

T cells expressing either 11D5-3-CD8BBZ or FHVH33-CD8BBZ specifically recognized BCMA⁺ target cell lines (Fig. 3a). The amounts of IFNγ and interleukin-(IL)-2 released by T cells expressing either CAR were not different; there was slightly higher tumor necrosis factor (TNF) production by FHVH33-CD8BBZ T cells (Fig. 3b–d). We quantified BCMA expression on target cell lines and assessed the ability of 11D5-3-CD8BBZ and FHVH33-CD8BBZ to recognize these cell lines by ICCS (Supplementary Fig. 2). In agreement with ELISA experiments, we found slightly higher TNF production by FHVH33-CD8BBZ T cells versus 11D5-3-CD8BBZ T cells. We assessed whether 11D5-3-CD8BBZ-expressing and FHVH33-CD8BBZ-expressing T cells could recognize primary human bone marrow multiple myeloma cells. We found that both CARs degranulated in response to the BCMA⁺ primary multiple myeloma cells (Fig. 3e). When T cells expressing 11D5-3-CD8BBZ or FHVH33-CD8BBZ were compared, there was no difference in BCMA-specific activation-induced cell death (AICD) or proliferation (Supplementary Figs. 3, 4). In contrast to the 4-1BB-containing CARs, when we compared CARs with CD28 domains in place of 4-1BB, we found higher levels of AICD with FHVH33-CD828Z versus 11D5-3-CD828Z (Supplementary Fig. 5). We did not observe a difference in aggregation of 11D5-3-CD828Z versus FHVH33-CD828Z CARs on the T-cell surface in experiments similar to previous work (Supplementary Fig. 6)⁴³.

**Fig. 3: T cells expressing either 11D5-3-CD8BBZ or FHVH33-CD8BBZ specifically recognized BCMA and primary multiple myeloma cells.**

Function of CARs with 4-1BB versus CD28 domains

We performed a functional comparison of T cells expressing either FHVH33-CD8BBZ or FHVH33-CD828Z. The fraction of T cells expressing FHVH33-CD8BBZ was slightly higher than the fraction of T cells expressing FHVH33-CD828Z after transduction (Fig. 4a). We assessed CAR expression by the antibody binding capacity approach;⁴⁴ we found similar levels of genomic integrations and CAR expression per genomic integration for T cells transduced with either FHVH33-CD828Z or FHVH33-CD8BBZ (Supplementary Fig. 7). We found that T cells expressing either FHVH33-CD8BBZ or FHVH33-CD828Z released IFNγ in a BCMA-specific manner (Fig. 4b). We found no difference in cytotoxicity (Fig. 4c) or degranulation (Fig. 4d, e) when FHVH33-CD8BBZ and FHVH33-CD828Z were compared. The quantities of IFNγ, TNF, and IL-2 released by CAR T cells expressing either FHVH33-CD8BBZ or FHVH33-CD828Z were not different (Fig. 4f–h). Because BCMA is a secreted protein that has been detected in the serum of multiple myeloma patients at a median level of 17.8 ng/mL¹⁰, we assessed whether solubilized BCMA could either block FHVH CAR recognition of BCMA⁺ target cells or cause nonspecific activation of CAR T cells. We found neither blocking of target cell recognition nor nonspecific CAR T-cell activation (Supplementary Table 2).

**Fig. 4: FHVH33 CARs with either a CD28 or 4-1BB costimulatory domain have similar in vitro function.**

Survival of T cells with FHVH33-CD8BBZ versus FHVH33-CD828Z

We labeled T cells expressing either FHVH33-CD8BBZ or FHVH33-CD828Z with carboxyfluorescein succinimidyl ester (CFSE) and cultured them together with BCMA-K562 cells or negative-control NGFR-K562 cells. During a 4-day in vitro culture, the number of FHVH33-CD8BBZ CAR T cells increased more than the number of FHVH33-CD828Z T cells (Fig. 5a). We expressed BCMA-specific dilution of CFSE as the CAR T-cell CFSE MFI with BCMA-K562 stimulation divided by the CAR T-cell CFSE MFI with NGFR-K562 stimulation. A lower BCMA-K562/NGFR-K562 MFI ratio indicates greater CFSE dilution and more BCMA-specific proliferation. T cells expressing FHVH33-CD828Z proliferated more than T cells expressing FHVH33-CD8BBZ (Fig. 5b).

**Fig. 5: 4-1BB versus CD28 CARs: proliferation and survival.**

We hypothesized that the lesser accumulation of FHVH33-CD828Z T cells despite greater proliferation was due to increased T-cell death in FHVH33-CD828Z T cells relative to FHVH33-CD8BBZ T cells. We stimulated FHVH33-CD8BBZ and FHVH33-CD828Z T cells with BCMA-K562 (BCMA⁺) cells or BCMA-negative target cells overnight, and then we performed annexin V staining to detect apoptosis caused by AICD. We found higher AICD levels among T cells expressing FHVH33-CD828Z compared with T cells expressing FHVH33-CD8BBZ (Fig. 5c). We also found higher AICD levels with FHVH33-CD828Z versus FHVH33-CD8BBZ when RPMI8226 target cells were used in AICD assays (Supplementary Fig. 8). RPMI8226 cells express lower levels of BCMA compared with BCMA-K562 cells. With RPMI8226 target cells, levels of AICD were higher in CD8⁺ T cells than in CD4⁺ T cells regardless of which CAR was expressed; we did not see a difference in AICD between CD4⁺ and CD8⁺ T cells with BCMA-K562 target cells (Fig. 5c). There was no difference in AICD between T cell expressing either 11D5-3-CD828Z or 11D5-3-CD8BBZ when the T cells were cultured overnight with RPMI8226 target cells (Supplementary Fig. 9).

We next compared in vitro cell counts of FHVH33-CD8BBZ versus FHVH33-CD828Z T cells after 7 days of culture with BCMA-K562 cells. FHVH33-CD8BBZ T cells accumulated more than FHVH33-CD828Z T cells (Fig. 5d). We also detected lower levels of T-cell apoptosis as measured by annexin V staining of FHVH33-CD8BBZ T cells versus FHVH33-CD828Z T cells in this culture system (Fig. 5e). The cell surface phenotype of T cells at the start of the 7-day cultures in these experiments is shown in Supplementary Fig. 10. We also compared in vitro cell counts and apoptosis of 11D5-3-CD828Z T cells and 11D5-3-CD8BBZ T cells after 7 days of culture with BCMA-K562. Accumulation of T cells was higher and apoptosis was marginally lower for 11D5-3-CD8BBZ versus 11D5-3-CD828Z (Supplementary Fig. 11). In two series of culture experiments with BCMA-K562-stimulated T cells, we found a higher CD4 to CD8 ratio with FHVH33-CD8BBZ than with FHVH33-CD828Z (Fig. 5f, Supplementary Fig. 12).

Murine studies of anti-BCMA CAR T cells

Disseminated malignancy burdens of the MM.1 S human multiple myeloma cell line were established in NOD.Cg-Prkdc^scid Il2rg^tm1Wjl/SzJ (NSG) mice. The MM.1 S cells were reduced to below detectable levels by single injections of CAR T cells expressing each of the anti-BCMA CARs assessed: 11D5-3-CD8BBZ, FHVH33-CD828Z, and FHVH33-CD8BBZ; however, most mice receiving FHVH33-CD828Z-expressing T cells developed high-tumor-burden relapses (Fig. 6a, numerical bioluminescence comparisons are in Supplementary Fig. 13). We also evaluated anti-BCMA CAR T-cell therapy in mice bearing solid tumors of the RPMI8226 human myeloma cell line. A single intravenous injection of T cells expressing 11D5-3-CD8BBZ, FHVH-CD828Z, or FHVH33-CD8BBZ completely eliminated tumors at doses of 1 × 10⁶ CAR⁺ T cells/mouse (Fig. 6b) or 2 × 10⁶ CAR⁺ T cells/mouse (Fig. 6c). None of the anti-BCMA CARs exhibited anti-tumor activity at a dose of 0.5 × 10⁶ CAR T cells/mouse (Supplementary Fig. 14).

To evaluate persistence of CAR⁺ T cells, we established disseminated MM.1 S cells in NSG mice and injected T cells expressing either FHVH33-CD828Z or FHVH33-CD8BBZ. Ten days later, splenic CAR T cells were quantified. We found higher absolute numbers of FHVH33-CD8BBZ T cells than FHVH33-CD828Z T cells (Fig. 6d, e). Ratios of the median absolute number of CD4⁺CAR⁺ T cells to median absolute number of CD8⁺CAR⁺ T cells were calculated. The ratio for mice receiving FHVH33-CD8BBZ was 1.8, and the ratio for mice receiving FHVH33-CD828Z was 0.7.

Because MM.1 S cells were used in murine experiments, we also performed in vitro proliferation assays with the MM.1 S cells as BCMA⁺ target cells. MM.1 S expresses substantially lower levels of BCMA than the BCMA-K562 cells used in most of our in vitro proliferation assays (Supplementary Fig. 15). In vitro, we found greater accumulation of FHVH33-CD8BBZ versus FHVH33-CD828Z T cells for CD4⁺ T cells (Supplementary Fig. 16). There was also a trend toward greater accumulation of FHVH33-CD8BBZ versus FHVH33-CD828Z T cells for CD8⁺ T cells that did not reach statistical significance.

Discussion

To address some of the limitations of CARs with scFv binding domains, we developed anti-BCMA CARs with antigen-binding domains consisting of only a single human heavy-chain variable region. There are three main rationales for using heavy-chain-only binding domains. First, the simpler structure of heavy-chain-only domains should be less immunogenic than an scFv that has an artificial linker and junctions between the light chain, linker, and heavy chain components. Our results demonstrate that T-cell responses can be elicited against the clinically used murine 11D5-3 scFv (Fig. 1). These results provide a rationale for developing less immunogenic CARs by using human rather than murine sequences and by using less complex antigen-binding domains, such as heavy-chain-only domains. Second, single heavy-chain-only domains are smaller than scFvs, and limiting the size of genes expressed by gene therapy vectors generally leads to the better gene expression by transduced T cells³⁴. Limiting the size of expressed genes is especially important when more than one protein, such as two CARs need to be expressed^35,45,46. Third, targeting multiple antigens simultaneously with the same CAR-expressing T cell is a major goal of the CAR field because of antigen loss from malignant cells targeted by CARs^{35,45,46,47,48,49}. Heavy-chain-only binding domains ease design of bispecific CARs^35,36. Targeting more than one antigen is important in multiple myeloma because some patients developed progressive multiple myeloma with loss of BCMA expression after anti-BCMA CAR T-cell therapy^16,17.

We believe that our work is notable for the careful comparison that we have made to CARs with the 11D5-3 scFv, the binder of the CAR T-cell product that is the most advanced in clinical trials, bb2121. CARs incorporating 11D5-3 have been proven effective in two clinical trials^16,17,19. When compared to CARs containing the clinically effective 11D5-3 scFv, the FHVH33-CD8BBZ CAR provides the advantages of fully human heavy-chain-only binding domains without any decrease of function of CAR-expressing T cells in vitro or in murine models.

We have initiated a phase I clinical trial of autologous T cells transduced with MSGV1-FHVH33-CD8BBZ. This is a dose-escalation trial that enrolls patients with relapsed or refractory multiple myeloma. The treatment protocol consists of a single infusion of CAR T cells preceded by cyclophosphamide and fludarabine conditioning chemotherapy. The primary objectives of the trial are to assess safety and feasibility of FHVH33-CD8BBZ T cells. Secondary objectives include assessment of the activity of FHVH33-CD8BBZ T cells against multiple myeloma and an assessment of in vivo proliferation and persistence of the CAR T cells.

CARs containing 4-1BB have different functional characteristics than CARs containing CD28, others have previously reported the ability of 4-1BB to increase T-cell persistence and resistance to AICD^2,8,50,51. We have previously conducted extensive preclinical and clinical work with CARs containing CD28 costimulatory domains^{11,16,17,23,52,53}. With traditional scFvs, we found that the CD28-containing CARs proliferated and survived well during in vitro culture^11,23,52. In contrast to our previous work, in vitro survival of T cells expressing the CD28-containing CAR FHVH33-CD828Z was poor relative to the survival of T cells expressing the 4-1BB-containing CAR FHVH33-CD8BBZ. These observations were confirmed when we saw greater accumulation of FHVH33-CD8BBZ T cells compared with FHVH33-CD828Z T cells in two types of in vitro culture experiments (Fig. 5a, d). There was greater accumulation of FHVH33-CD8BBZ T cells despite the finding that FHVH33-CD828Z T cells proliferated more than FHVH33-CD8BBZ T cells when stimulated with BCMA⁺ target cells (Fig. 5b). We found more apoptosis in FHVH33-CD828Z versus FHVH33-CD8BBZ T cells after BCMA stimulation in three different experimental systems (Fig. 5c, e, and Supplemental Fig. 8); therefore, the likely explanation for lesser accumulation of FHVH33-CD828Z T cells relative to FHVH33-CD8BBZ T cells was increased death among FHVH33-CD828Z T cells versus FHVH33-CD8BBZ T cells. These in vitro findings were consistent with the lower numbers of FHVH33-CD828Z T cells versus FHVH33-CD8BBZ T cells persisting in mice 10 days after injection (Fig. 6d, e). Importantly, use of a 4-1BB costimulatory domain protected FHVH33 CARs from AICD; AICD levels were not different for CARs containing FHVH33 versus 11D5-3 when the CARs contained a 4-1BB costimulatory domains (Supplementary Fig. 3); however, when CARs with CD28 costimulatory domains were compared, T cells expressing FHVH33-CD828Z had higher levels of AICD compared with T cells expressing 11D5-3-CD828Z (Supplementary Fig. 5). 11D5-3-CD828Z had an identical sequence to FHVH33-CD828Z except for substitution of the 11D5-3 scFv for FHVH33. Interestingly, when we compared AICD in 11D5-3-CD828Z and 11D5-3-CD8BBZ, there was no statistically significant difference, which suggests that 4-1BB is more important in protecting against AICD with a CAR incorporating a heavy-chain-only binding domain versus a CAR with an scFv (Supplementary Fig. 9).

We consistently found a higher CD4 to CD8 ratio among FHVH33-CD8BBZ T cells versus FHVH33-CD828Z T cells (Fig. 5f, Supplemental Fig. 12). One possible factor contributing to the higher CD4 to CD8 ratio with FHVH33-CD8BBZ was that the mean ratio of CD4⁺ to CD8⁺ T-cell apoptosis was lower for FHVH33-CD8BBZ T cells than FHVH33-CD828Z T cells. (Supplementary Fig. 8). The difference in CD4⁺ to CD8⁺ ratio for FHVH33-CD8BBZ versus FHVH33-CD828Z was not due to differences in proliferation of CD8⁺ and CD4⁺ T cells (Supplementary Fig. 17). There was also a modestly higher CD4 to CD8 ratio of splenic CAR⁺ T cells in mice receiving FHVH33-CD8BBZ T cells versus mice receiving FHVH33-CD828Z T cells. Since the goal of our work is to produce an effective clinical CAR T-cell treatment, we are concerned about both the in vitro growth characteristics of CAR T cells and the in vivo properties of the T cells. Similar to predominance of CD4⁺CAR⁺ T cells in cultures of FHVH33-CD8BBZ T cells, there was a high CD4 to CD8 ratio of CAR⁺ T cells observed on a clinical trial of the bb2121 CAR T-cell product that has been clinically effective at treating multiple myeloma¹⁹. Like FHVH33-CD8BBZ, the bb2121 T-cell product has a CAR with a 4-1BB domain.

In summary, CARs with fully human heavy-chain-only binding domains have important advantages over scFv binding domains. The FHVH33 domain had no functional disadvantages in vitro or in murine models when compared to the 11D5-3 scFv, when a 4-1BB moiety is included in CARs.

Methods

Use of human cells and use of mice in experiments

PBMC and bone marrow cells were used from patients enrolled on National Cancer Institute (NCI) clinical trials. Use of patient samples for research was approved by the NCI Institutional Review Board. Informed consent was obtained from all patients. All animal studies were carried out on protocols approved by the NCI Animal Care and Use Committee.

Gamma-retroviral transductions and T-cell culture

To produce replication-incompetent gamma-retroviruses, packaging cells were transfected with plasmids encoding CARs along with a plasmid encoding the RD114 envelope protein as previously described⁵². Gamma-retroviral transduction of T cells with genes encoding CARs or other proteins was performed as previously described 2 days after initiation of T-cell cultures⁵². T cells were cultured as described previously⁵². In brief, PBMC were stimulated with the anti-CD3 monoclonal antibody OKT3 (Ortho) in AIM V complete medium (Invitrogen) and 300 international units (IU)/mL of IL-2 (Teceleukin, Roche). AIM V complete medium consisted of AIM V medium (Thermo), 5% human AB serum (Valley Biomedical), 100 U/mL penicillin, and 100 μg/mL streptomycin. Non-tissue culture-treated six-well plates were coated with 10 μg/mL Retronectin (Takara). Two days after OKT3 stimulation, 2 mL of retroviral supernatant was applied to each well of the retronectin-coated, plates and incubated for 2 h at 37°C. Cells were resuspended in AIM V complete medium, and 2 × 10⁶ cells in 2 mL of medium were added directly to the viral supernatant. IL-2 was added to a final concentration of 300 IU/mL. Transduction was performed at 37 °C for 16–18 hours. Following transduction, cells were suspended in fresh AIM V complete media containing 300 IU/mL IL-2. Cultures were maintained by suspending the desired cell number to a concentration of 0.5 × 10⁶ cells/mL in AIM V complete media with 300 IU/mL IL-2 every 2 days.

Generation of autologous stimulator cells

Autologous stimulator cells expressing a nonfunctioning version of the 11D5-3-CD828Z CAR were used to stimulate PBMC to induce proliferation of 11D5-3-specific T cells. The sequence of the 11D5-3-NS gene was as follows from N-terminus to C-terminus: the CD8α signal sequence, murine anti-BCMA scFv (11D5-3), and CD8α hinge and transmembrane sequence. The 11D5-3-NS sequence was synthesized as a gBlock fragment by Integrated DNA Technologies (IDT) and cloned into the MSGV1 gamma-retroviral vector backbone⁵⁴ by using XhoI and BamHI restriction enzymes (NEB). Gamma-retroviral vector encoding 11D5-3-NS was produced and patient T cells were transduced as described under “Gamma-retroviral transductions and T-cell culture”. Patient T cells were also transduced with a gamma retroviruses made with a previously described plasmid⁵² encoding human full-length low-affinity NGFR. Cells were cryopreserved (Cryostor CS10, StemCell Technologies) for subsequent stimulations.

Generation of a 15-mer peptide library from the 11D5-3 scFv

Individual 15-mer peptides with 11 amino acid overlapping sequences were generated against amino acids 8–279 of the 11D5-3-CD828Z CAR¹¹ (PEPScreen, Sigma Aldrich). The 15-mers cover the light chain, linker, and heavy chain of the 11D5-3 scFv sequence. Individual peptides were dissolved in dimethyl sulfoxide to a concentration of 20 mg/mL. For rapid screening, 13 pools, each consisting of five peptides, were generated.

Stimulation of PBMC with CAR-expressing stimulator cells

We cultured patient PBMC to assess for CAR-reactive T cells by modifying previously described methods^37,38. In nine out of ten patients, patient PBMCs were stimulated with irradiated (3000 rad) autologous 11D5-3-NS-transduced stimulator T cells generated as described under “Generation of autologous stimulator cells”. In one patient, PBMC were stimulated with irradiated (3000 rad) autologous 11D5-3-CD828Z-transduced stimulator T cells because of insufficient starting cells to produce 11D5-3-NS stimulator cells.

CAR-reactive T cells were expanded using two different culture processes. Using the first method³⁷, patient PBMC were thawed and cultured with irradiated autologous 11D5-3-NS stimulator cells at a 2:1 responder:stimulator ratio in T-cell complete media: RPMI1640 + L-glutamine (2 mM), heat-inactivated human AB serum (10%, Valley Biomedical), HEPES (25 mM, Corning), 2-mercaptoethanol (0.025 mM, Thermo Fisher), and Pen/Strep (100 U/mL penicillin, 100 mcg/mL streptomycin, Thermo Fisher). At day 3, IL-2 was added to the cultures (20 IU/mL). At day 7, responder PBMC were restimulated with irradiated 11D5-3-NS stimulator cells and irradiated autologous PBMC as feeder cells at a 2:1:4 responder:stimulator:feeder ratio. At days 9 and 11, IL-2 was added to the cultures (20 IU/mL). Using the second method³⁸, patient PBMCs were thawed and cultured with irradiated autologous 11D5-3-NS stimulator cells at a 1:2 responder:stimulator ratio in T-cell complete media. At day 2, IL-2 was added to the cultures (300 IU/mL). At day 7, responder PBMC were restimulated with irradiated 11D5-3-NS stimulator cells at a 1:8 responder:stimulator ratio. At day 8, IL-2 was added to the cocultures (300 IU/mL). For both methods, day 14 responder PBMC were used for experiments to assess for reactivity to the 11D5-3 scFv.

Assays to detect CAR-specific T-cell responses

For the detection of anti-11D5-3 scFv T-cell responses by IFNγ ELISA, day 14 responder PBMC were cultured alone or with autologous T cells expressing either 11D5-3-NS or NGFR at a 1:1 responder:stimulator ratio in T-cell complete media. Supernatants were assayed for IFNγ by ELISA after a 20-h incubation at 37°C.

For ICCS to assess stimulated PBMC reactivity with 15-mer peptide libraries, autologous monocyte-derived immature dendritic cells (MoDCs) were generated using methods previously described⁵⁵. MoDCs were pulsed with 10 µg/mL of mixed peptides from peptide libraries or individual peptides overnight in T-cell complete media at 37°C. As a negative control, MoDCs were pulsed with 10 µg/mL hepatitis B core protein (HBC_128–140) peptide. Responder PBMC were cultured with peptide-pulsed MoDCs at a 2:1 responder:stimulator ratio in the presence of no-azide/low-endotoxin-grade anti-CD28 and anti-CD49d antibodies (BD Biosciences) both at 1 µg/mL in T-cell complete media for 1 h at 37°C. After 1 h, Brefeldin A (GolgiPlug, BD Biosciences) was added (1:1000 dilution by the manufacturer’s instructions) and the cultures were incubated for an additional 5 h. Cells were stained for cell surface markers for CD3, CD4, and CD8, fixed with Cytofix/Cytoperm, and intracellular staining for IFNγ was conducted as recommended (BD Biosciences). Samples were analyzed by flow cytometry.

Cell lines

RPMI8226 and MM.1 S are BCMA⁺ multiple myeloma cell lines that were obtained from ATCC. L363 is a BCMA⁺ myeloma cell line from DSMZ. A549 is a BCMA-negative lung cancer cell line (ATCC). CCRF-CEM is a BCMA-negative T-cell line (ATCC). HepG2 is a BCMA-negative hepatic carcinoma cell line (ATCC). Colo205 is a BCMA-negative colon carcinoma cell line (ATCC). BCMA-K562 are K562 cells (ATCC) transduced with the gene for full-length BCMA in our laboratory. C17-BCMA-K562 cells are BCMA-K562 cells cloned to limiting dilution and selected for low expression of BCMA. NGFR-K562 are K562 cells transduced with the gene for low-affinity nerve growth factor in our laboratory⁵². The same gamma-retroviral vector and methods were used to transduce BCMA-K562 and NGFR-K562. All cell lines were tested for mycoplasma and found to be negative.

Design and construction of plasmids encoding FHVH CARs

We designed a series of CARs that contained fully human heavy-chain variable-region-only antigen-recognition domains (FHVH). The sequence of each CAR followed this pattern from the 5′-end to the 3′-end: CD8α signal sequence, one of 4 single FHVH domains, and the hinge and transmembrane regions of the human CD8α molecule. the cytoplasmic portion of either the CD28 or 4-1BB molecules, and the cytoplasmic portion of the CD3ζ molecule. The sequences used for CD8α, CD28, 4-1BB, and CD3ζ were obtained from the National Center for Biotechnology Information website (www.ncbi.nlm.nih). Guidance regarding the portions of each molecule to include in the CARs was obtained from prior work⁵².

The four fully human heavy-chain variable-region-only antigen-recognition domains were designated FHVH 74, 32, 33, and 93⁴⁰. CARs were named in a systematic manner. For example, FHVH33-CD8BBZ has the FHVH33 antigen-recognition domain, a hinge and transmembrane region from CD8α, a 4-1BB costimulatory domain, and the CD3ζ T-cell activation domain.

All of these CARs were constructed and the CAR sequences were ligated into the MSGV1 gamma-retroviral vector backbone⁵⁴ by standard methods. BCMA-specific variable heavy-chain sequences were synthesized as gBlocks by IDT. Each synthesized fragment consisted of a GTC trinucleotide, a NcoI site, the CD8α signal sequence, the FHVH sequence, part of the CD8α hinge and transmembrane domain, a BlpI site, and a TATCGT hexanucleotide. The GTC and TATCGT nucleotides were added to ensure complete end cleavage with NcoI and BlpI. Fragments were digested with BlpI and NcoI-HF (New England Biolabs) for 2 h at 37 °C. Digested fragments were then purified using the QIAquick PCR Purification kit (Qiagen). Fragments were ligated into the BlpI/NcoI-HF digested and gel-purified MSGV1 vector backbones that also included other components of the CAR not included in the gBlock fragments. The CAR components included in the MSGV1 vector backbones were: the remainder of the CD8α domain that was not included in the gBlock fragment, either CD28 or 4-1BB, and the CD3ζ domain. The ligation of each gBlock CAR fragment and the MSGV1 vector backbone fragment was carried out by using the Rapid DNA Ligation Kit (Roche Applied Science).

We also utilized the previously reported 11D5-3-CD828Z anti-BCMA CAR¹¹, and 11D5-3-CD8BBZ, a CAR identical to 11D5-3-CD828Z except for replacement of CD28 by 4-1BB. The SP6 scFv recognizes the hapten 2, 4, 6-trinitrobenzenesulfonic acid⁵⁶. We constructed a gamma-retroviral plasmid encoding a CAR with the SP6 scFv designated MSGV1-SP6-CD828Z, and we used it as a negative control.

CAR detection on T cells and flow cytometry

T cells that were transduced with one of the CAR vectors and untransduced T cells were washed and stained with a BCMA-Fc protein labeled with PE (BCMA-Fc-PE) to detect cell-surface CAR molecules. T cells (5 × 10⁵) were suspended in 50 μL of staining buffer (0.4% bovine serum albumin, 0.1% sodium azide in phosphate-buffered saline; PBS), and a titrated amount of the BCMA-Fc-PE reagent (Creative Biomart) was added. For T-cell phenotyping experiments, cells were stained with the following antibodies: CD3 allophycocyanin (APC)-Cy7 (Clone UCHT1, BD Biosciences), CD4 FITC/BV510 (Clone RPA-T4, BD Biosciences; Biolegend), CD8 PE-Cy7/eFluor450 (Clone RPA-T8, BD Biosciences; Thermo Scientific), CD45RA FITC (Clone HI100, BD Thermo Scientific), CCR7 APC (Clone 150503, BD Biosciences), and CD57 FITC (Clone HCD57, Biolegend). For intracellular staining, cells were stained with IFNγ APC (Clone B27, BD Biosciences).

Flow cytometry was performed by standard methods. Dead cells were excluded by using 7-amino-actinomycin D (7-AAD, BD Biosciences). Flow cytometry analysis for all experiments was performed by using FlowJo (Tree Star, Inc.). The general flow cytometry gating strategies used are shown in Supplementary Fig. 18.

Nonlinear regression analysis for CAR affinity

CAR affinity for BCMA protein was analyzed using the methods previously described^41,42. A total of 5 × 10⁵ CAR T cells were stained for 30 minu at 37°C with BCMA-Fc-PE protein (Creative Biomart) diluted in four-fold serial dilutions; the cells were also stained with anti-CD3. For the first dilution, BCMA-Fc-PE protein was added to a final concentration of 13.175 nM in a 57.5 μL test volume. Flow cytometry was performed and the MFI of BCMA-Fc-PE-bound CAR T cells was determined. K_D values were calculated using the one-site binding (hyperbola) nonlinear regression model (GraphPad Prism 7) and the MFI of BCMA-Fc-PE-bound CAR T cells was expressed as a percentage of maximum binding sites (B_max).

CD107a assay

For each T-cell culture that was tested, two tubes were prepared. One tube contained C17-BCMA-K562 cells or RPMI8226 cells, and the other tube contained NGFR-K562 cells. Both tubes contained CAR-transduced T cells, 1 ml of AIM-V complete medium, a titrated concentration of an anti-CD107a antibody (Clone H4A3, Thermo Scientific), and 1 μL of Golgi Stop (monensin, BD). All tubes were incubated at 37 °C for 4 h and then stained for CD3, CD4, and CD8. Samples were analyzed by flow cytometry. Normalization was carried out by dividing the percentage of CD4⁺ or CD8⁺ T cells that were CD107a⁺ by the percentage of CD4⁺ or CD8⁺ T cells that were CAR⁺ by BCMA-Fc-PE staining.

Cytotoxicity assay

Cytotoxicity assays were conducted as previously described^52,57. Cytotoxicity was measured by comparing survival of RPMI8226 BCMA⁺ target cells relative to the survival of negative-control CCRF-CEM cells. Both of these cell types were combined in the same tubes with CAR-transduced T cells. CCRF-CEM negative-control cells were labeled with the fluorescent dye 5-(and-6)-(((4-chloromethyl)benzoyl)amino) tetramethylrhodamine (CMTMR; Thermo Scientific), and RPMI8226 BCMA⁺ target cells were labeled with labeled with carboxyfluorescein diacetate succinimidyl ester (CFSE; Thermo Scientific). In some experiments, CAR T-cell cultures were depleted of natural killer cells by flow cytometry (FACSAria, BD) prior to use in cytotoxicity assays. Cultures were set up in sterile 5 mL test tubes (BD) in duplicate at multiple T cell to target cell ratios. The target cells contained in the tubes were 50,000 CFSE-labeled RPMI8226 target cells along with 50,000 CMTMR-labeled CCRF-CEM negative-control cells. The cultures were incubated for 4 h at 37 °C. Immediately after the incubation, 7-AAD (BD Biosciences) was added, and flow cytometry acquisition was performed. For each T cell plus target-cell culture, the percent survival of RPMI8226 target cells was determined by dividing the percent live RPMI8226 cells by the percent live CCRF-CEM negative-control cells. The corrected percent survival of RPMI8226 target cells was calculated by dividing the percent survival of RPMI8226 target cells in each T cell plus target cell culture by the ratio of the percent live RPMI8226 target cells to percent live CCRF-CEM negative-control cells in tubes containing only RPMI8226 target cells and CCRF-CEM cells without effector T cells. This correction was necessary to account for variation in the starting cell numbers and for spontaneous target cell death. Cytotoxicity was calculated as follows: the percent cytotoxicity of RPMI8226 target cells = 100-corrected percent survival of RPMI8226 target cells.

Annexin V staining

CAR-transduced T cells were incubated overnight in 24-well plates with either BCMA-K562, RPMI8226, or NGFR-K562 target cells with 1.5 × 10⁶ T cells and 1 × 10⁶ target cells in each well. After overnight incubation, cells were stained with BCMA-Fc-PE, CD3, CD4, and CD8. The cells were washed twice with PBS, resuspended in Annexin V Binding buffer (BD Biosciences), and incubated with APC-conjugated Annexin V (BD Biosciences) and 7-AAD (BD Biosciences) for 15 mins at room temperature. The cells were immediately analyzed by flow cytometry.

Cytokine ELISA

BCMA⁺ or BCMA-negative target cells were combined with CAR-transduced T cells in duplicate wells of a 96-well round bottom plate in AIM-V complete medium at a 1:1 effector:target ratio. The plates were incubated at 37 °C for 18–20 h. Following the incubation, ELISAs for IFNγ were performed by using standard methods as previously described²³. IL-2 and TNF ELISAs (R&D Systems) were performed as recommended by the manufacturer. When two or more CARs were compared, cytokine release was normalized for CAR expression by dividing the cytokine levels by the fraction of T cells expressing a given CAR.

ICCS with BCMA⁺ cell lines

The %CAR⁺ cells within tested CAR T-cell cultures was normalized using autologous untransduced T cells (UT), such that the %CAR⁺ was the same for all tested CAR T-cell cultures. T cells were cocultured with BCMA⁺ cell lines at a 2:1 effector:target ratio in AIM V complete media without IL-2 for 6 h at 37°C in the presence of Brefeldin A (GolgiPlug, BD Biosciences, 1:1000 dilution by the manufacturer’s instructions). Cells were stained for CD3, fixed with Cytofix/Cytoperm, and intracellular staining for IFNγ, IL-2, and TNF was performed.

Solubilized BCMA protein ELISA

Anti-BCMA CAR T cells were cultured either alone or with BCMA-K562 or RPMI8226 at a 1:1 effector:target ratio in AIM V complete media. Functional grade BCMA protein (Novus Biologicals, NBP2-34903) was added to the cultures to final concentrations of 25, 50, or 150 ng/mL. After 16 h at 37°C, IFNγ was assayed by ELISA of the culture supernatants as described²³.

Proliferation assays

Cultures were set up in 24-well plates. Target cells included in cultures were either 0.5 × 10⁶ irradiated BCMA-K562 cells or 0.5 × 10⁶ irradiated NGFR-K562 cells. In some experiments, MM.1 S replaced BCMA-K562 as the BCMA⁺ target. The cultures also included 0.75 × 10⁶ T cells from cultures that had been transduced with an anti-BCMA CAR. The T cells were labeled with CFSE (Thermo Scientific) as previously described⁵⁸. The medium used in the cocultures was AIM V complete media. IL-2 was not added to the medium. Four days after initiation, the live cells in each culture were counted with trypan blue for dead cell exclusion, and flow cytometry was performed after staining with BCMA-Fc-PE, anti-CD3, anti-CD4, and anti-CD8. BCMA-specific proliferation presented as CFSE MFI of T cells stimulated with BCMA-K562 divided by the CFSE MFI of T cells stimulated with NGFR-K562.

Antibody and BCMA-Fc-PE binding capacity measurements

Methods for the calculation of antibody or BCMA-Fc-PE binding capacity were based on the methods published in ref. ⁴⁴. PE anti-human BCMA (FAB193P, R&D Systems) and BCMA-Fc-PE (Creative Biomart) were first titered on BCMA-K562 and FHVH33-CD8BBZ CAR T cells, respectively by using previously published methods⁵⁹. PE antibody or BCMA-Fc-PE binding capacity was enumerated using BD Quantibrite™ PE beads using the manufacturer’s suggested protocol (Catalog: 340495, BD Biosciences).

Vector copy number quantitative PCR

CD3⁺CAR⁺ FHVH33 T cells were sorted by flow cytometry using a BD FACSAria or BD Influx sorter. Genomic DNA from cells was extracted using the DNeasy Blood and Tissue Kit using the manufacturer’s suggested protocol (Qiagen). A transgene-specific quantitative PCR (qPCR) assay designed against the MSGV1 γ-retroviral vector was used.

RVV Forward: 5′-TTAGGTCACTGGAAAGATGTCG-3′

RVV Probe: 5′-6-FAM-AGACGTTGG-ZEN-GTTACCTTC-3′-IBFQ

RVV Reverse: 5′-GATGAGGTCTCGGTTAAAGGTG-3′

For a copy number reference, an RNaseP qPCR assay was used (catalog: 4403326, Thermo Fisher). For absolute quantification of vector copy number/cell, a calibrator plasmid was generated by cloning the human RNaseP gene sequence (NR_002312.1) into the MSGV1-FHVH33-CD8BBZ plasmid using BlpI and BamHI sites (New England Biolabs). For qPCR, the Quanta Perfecta Master Mix was used (catalog: 95076-012, Quanta) and reactions were cycled in a Roche LightCycler 96.

Seven-day in vitro BCMA-specific culture assay

CAR T cells were cultured with irradiated BCMA-K562 target cells (18,000 rad) at a 2:1 effector:target ratio in AIM V complete media without IL-2 for 3 days. After 3 days, the total cells in culture were counted and AIM V complete media was added so that the final concentration of cells was 1 × 10⁶/mL. Initial and final CAR T cell numbers were calculated at the start and end of the 7-day culture period by counting cells with trypan blue for dead cell exclusion and performing flow cytometry with staining for BCMA-Fc-PE and anti-CD3. Cells at the end of the culture were also stained with Annexin V (BD Biosciences) to evaluate the percentage of apoptotic CAR⁺ T cells.

Solid myeloma tumor in vivo experiments

NSG mice at 6–8 weeks of age from NCI-Frederick or the Jackson Laboratories were injected with 8 × 10⁶ RPMI8226 cells in PBS intradermally. Tumors were allowed to grow for 10 days until measurable tumors were present. CAR T cells that had been started in culture 7 days earlier were injected intravenously at doses ranging from 0.5 to 2 × 10⁶ CD3⁺CAR⁺ cells/mouse. Mice received one injection of CAR T cells. Tumors were measured using a caliper every 3 days, and the volume of the tumors were calculated using the formula (length × width × height)/2. Mice were sacrificed once tumors reached 15 mm in the longest length.

Disseminated myeloma tumor experiments

MM.1 S cells were transduced in our lab with a retrovirus encoding an enhanced firefly luciferase gene⁶⁰. MM.1S-luciferase cells were stained with a monoclonal antibody against Thy1.1 (BD Biosciences) and sorted by flow cytometry to purity (FACSAria, BD Biosciences). NSG mice from NCI-Frederick or the Jackson Laboratories were injected with 8 × 10⁶ MM.1S-luciferase cells intravenously. After 10 days, CAR T cells at a dose of 1–2 × 10⁶ CD3⁺CAR⁺ cells/mouse were infused intravenously. Mice received 1 injection of CAR T cells. For imaging, mice were injected with 100 µL of luciferin solution (15 mg/mL in PBS, GoldBio) and anesthetized with 3% isoflurane. After 10 min, bioluminescence imaging (BLI) was captured using a Xenogen IVIS Imaging System. Ventral images were taken using a 1-min exposure on a 24 cm field of view with a binning factor of 4. BLI was quantified over the body of the mouse without the tail in units of radiance (p/sec/cm²/sr) using the Living Image software (Xenogen). Images were scaled to 10⁷–10⁸ radiance units. Mice were sacrificed upon the onset of severe hind leg paralysis and wasting in accordance with the NCI Animal Care and Use Committee guidelines.

In vivo T-cell persistence experiments

NSG mice (6–8 weeks old) from NCI-Frederick or the Jackson Laboratories were injected with 8 × 10⁶ MM.1S-luciferase cells in PBS intravenously. After 13 days, 1 × 10⁶ CD3⁺CAR⁺ T cells/mouse were infused intravenously. Mice were sacrificed at day 10 after T-cell infusion, and spleens were dissected from mice. Whole spleens were dissociated with a syringe plunger. Red blood cells were lysed with lysis buffer (Quality Biological) and 0.5 × 10⁶ splenocytes were used for flow cytometry detection of CAR T cells by staining with BCMA-Fc-PE, anti-CD3, anti-CD4, and anti-CD8. To calculate the number of CAR T cells/spleen, the total number of splenocytes per mouse was multiplied by the product of the %CD3⁺, %CD4⁺ or %CD8⁺, and the %CAR⁺.

Dronpa imaging

BCMA CAR-Dronpa fusion proteins were generated by cloning the full-length Dronpa coding region to the 3′-ends of the 11D5-3-CD828Z and FHVH33-CD828Z coding regions in the MSGV1 γ-retroviral backbone.

The microscope used for imaging experiments is a home-built system on a Nikon TE2000 base described previously and reiterated here with necessary alterations⁶¹. The objective lens used for imaging was a Nikon 100 × /1.4 NA Oil Plan Apo. A 405 nm laser (LaserBoxx, Oxxius) was used to photoswitch proteins to the “on” state and a 488 nm laser (Sapphire, Coherent Inc.) was used to image Dronpa. Laser lines were combined using appropriate dichroic mirrors. The 405 nm laser current was controlled using the ESIo AOTF controller (ESImaging) and was shuttered using a diaphragm shutter with controller (part# SH025T, Thorlabs, Inc.) triggered using the ESIo AOTF controller (ESImaging). The 488 nm laser line was controlled using an AOTF (Gooch & Housego PLC). All lasers are passed through a linear polarizer (part# WP25M-VIS, Thorlabs) and directed toward objective using a 488 nm dichroic mirror (part# DiO3-R488, Semrock). The emission is passed through the dichroic and reflected toward camera using 45-degree mirror. Emission is passed through appropriate emission filters and passed through a Dual-View splitter (Photometrics). A DV2 POL cube (part# DV2-POL-CUBE-KIT, Photometrics) was inserted into the Dual-View imager which splits emission in orthogonal polarizations allowing a simultaneous recording of both images using a PCO Edge 4.2 LT (PCO AG) camera. The microscope was controlled using MicroManager. The estimated imaging power density was ~0.03 W/cm² for 488 nm excitation.

Statistics

The statistical tests used in each experiment are given in the figure legends of each figure. In general, we used two-tailed paired Student’s t-tests for analyzing in vitro experiments. Mouse survival curves were generated by the method of Kaplan and Meier. P values were not corrected for multiple comparisons. Mouse tumor measurements were blinded to the investigator performing the tumor measurements. In vitro assays were not blinded. GraphPad Prism 7 was used for statistical analysis and for making graphs. All ELISA and cytotoxicity assays were set up in duplicate wells. In all in vitro experiments, “n” refers to the number of independent experiments with cells from different donors. In murine experiments, “n” refers to an individual mouse. Values for “n” are given in the legends of each figure. Replicates of mouse adoptive T-cell transfer experiments were performed with T cells from different human donors. In all cases, P < 0.05 was considered statistically significant.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.

Data availability

The authors will provide any primary data for results presented in this publication upon request. Upon reasonable request, we will provide materials not commercially available that were used in this work. GenBank accession codes of novel CARs reported in this manuscript were as follows:

BankIt2257033 11D5-3-CD8BBZ MN366105

BankIt2257033 FHVH33-CD828Z MN366106

BankIt2257033 FHVH33-CD8BBZ MN366107

Change history

09 March 2020
An amendment to this paper has been published and can be accessed via a link at the top of the paper.

References

Mikkilineni, L. & Kochenderfer, J. N. Chimeric antigen receptor T-cell therapies for multiple myeloma. Blood 130, 2594–2602 (2017).
Article CAS PubMed PubMed Central Google Scholar
Milone, M. C. et al. Chimeric receptors containing CD137 signal transduction domains mediate enhanced survival of T cells and increased antileukemic efficacy in vivo. Mol. Ther. 17, 1453–1464 (2009).
Article CAS PubMed PubMed Central Google Scholar
Sadelain, M., Brentjens, R. & Rivière, I. The basic principles of chimeric antigen receptor design. Cancer Discov. 3, 388–398 (2013).
Article CAS PubMed PubMed Central Google Scholar
Jensen, M. C. & Riddell, S. R. Designing chimeric antigen receptors to effectively and safely target tumors. Curr. Opin. Immunol. 33, 9–15 (2015).
Article CAS PubMed PubMed Central Google Scholar
Brudno, J. N. & Kochenderfer, J. N. Chimeric antigen receptor T-cell therapies for lymphoma. Nat. Rev. Clin. Oncol. 15, 31–46 (2018).
Article CAS PubMed Google Scholar
Dotti, G., Gottschalk, S., Savoldo, B. & Brenner, M. K. Design and development of therapies using chimeric antigen receptor-expressing T cells. Immunol. Rev. 257, 107–126 (2014).
Article CAS PubMed Google Scholar
Boyiadzis, M., et al. The society for immunotherapy of cancer consensus statement on immunotherapy for the treatment of hematologic malignancies: multiple myeloma, lymphoma, and acute leukemia. J. Immunother. Cancer 4, 90 (2016).
Salter, A. I., et al. Phosphoproteomic analysis of chimeric antigen receptor signaling reveals kinetic and quantitative differences that affect cell function. Sci. Signal. 11, eaat6753 (2018).
Article PubMed PubMed Central CAS Google Scholar
Laabi, Y. et al. The BCMA gene, preferentially expressed during B lymphoid maturation, is bidirectionally transcribed. Nucleic Acids Res. 22, 1147–1154 (1994).
Article CAS PubMed PubMed Central Google Scholar
Sanchez, E. et al. Serum B-cell maturation antigen is elevated in multiple myeloma and correlates with disease status and survival. Br. J. Haematol. 158, 727–738 (2012).
Article CAS PubMed Google Scholar
Carpenter, R. O. et al. B-cell maturation antigen is a promising target for adoptive T-cell therapy of multiple myeloma. Clin. Cancer Res. 19, 2048–2060 (2013).
Article CAS PubMed PubMed Central Google Scholar
Chim, C. S., et al. Management of relapsed and refractory multiple myeloma: novel agents, antibodies, immunotherapies and beyond. Leukemia 32, 252–262 (2017).
Article PubMed PubMed Central CAS Google Scholar
Goldschmidt, H., Ashcroft, J., Szabo, Z. & Garderet, L. Navigating the treatment landscape in multiple myeloma: which combinations to use and when? Ann. Hematol. 98, 1–18 (2019).
Kumar, S. K., Buadi, F. K. & Vincent Rajkumar, S. Pros and cons of frontline autologous transplant in multiple myeloma: The debate over timing. Blood 133, 652–659 (2019).
Article CAS PubMed Google Scholar
Kriegsmann, K. et al. Cell-based immunotherapy approaches for multiple myeloma. Br. J. Cancer 120, 38–44 (2019).
Article PubMed Google Scholar
Ali, S. A. et al. T cells expressing an anti-B-cell maturation antigen chimeric antigen receptor cause remissions of multiple myeloma. Blood 128, 1688–1700 (2016).
Article CAS PubMed PubMed Central Google Scholar
Brudno, J. N. et al. T cells genetically modified to express an anti–B-cell maturation antigen chimeric antigen receptor cause remissions of poor-prognosis relapsed multiple myeloma. J. Clin. Oncol. 36, 2267–2280 (2018).
Article CAS PubMed PubMed Central Google Scholar
Zhao, W. H., et al. A phase 1, open-label study of LCAR-B38M, a chimeric antigen receptor T cell therapy directed against B cell maturation antigen, in patients with relapsed or refractory multiple myeloma. J. Hematol. Oncol. 11, 141 (2018).
Raje, N. et al. Anti-BCMA CAR T-cell therapy bb2121 in relapsed or refractory multiple myeloma. N. Engl. J. Med. 380, 1726–1737 (2019).
Article PubMed PubMed Central CAS Google Scholar
Turtle, C. J., et al. Immunotherapy of non-Hodgkin’s lymphoma with a defined ratio of CD8+ and CD4+ CD19-specific chimeric antigen receptor-modified T cells. Sci. Transl. Med. 8, 355ra116 (2016).
Article PubMed PubMed Central CAS Google Scholar
Lamers, C. H. J. et al. Immune responses to transgene and retroviral vector in patients treated with ex vivo-engineered T cells. Blood 117, 72–82 (2011).
Article CAS PubMed Google Scholar
Maus, M. V. et al. T cells expressing chimeric antigen receptors can cause anaphylaxis in humans. Cancer Immunol. Res. 1, 26–31 (2013).
Article CAS PubMed PubMed Central Google Scholar
Alabanza, L. et al. Function of novel anti-CD19 chimeric antigen receptors with human variable regions is affected by hinge and transmembrane domains. Mol. Ther. 25, 2452–2465 (2017).
Article CAS PubMed PubMed Central Google Scholar
Sommermeyer, D. et al. Fully human CD19-specific chimeric antigen receptors for T-cell therapy. Leukemia 31, 2191–2199 (2017).
Article CAS PubMed PubMed Central Google Scholar
Holliger, P. & Hudson, P. J. Engineered antibody fragments and the rise of single domains. Nat. Biotechnol. 23, 1126–1136 (2005).
Article CAS PubMed Google Scholar
Stanfield, R. L., Dooley, H., Flajnik, M. F. & Wilson, I. A. Crystal structure of a shark single-domain antibody V region in complex with lysozyme. Science 305, 1770–1773 (2004).
Article ADS CAS PubMed Google Scholar
Iri-Sofla, F. J., Rahbarizadeh, F., Ahmadvand, D. & Rasaee, M. J. Nanobody-based chimeric receptor gene integration in Jurkat cells mediated by PhiC31 integrase. Exp. Cell Res. 317, 2630–2641 (2011).
Article CAS PubMed Google Scholar
Hamers-Casterman, C. et al. Naturally occurring antibodies devoid of light chains. Nature 363, 446–448 (1993).
Article ADS CAS PubMed Google Scholar
Osborn, M. J. et al. High-affinity IgG antibodies develop naturally in Ig-knockout rats carrying germline human IgH/Igκ/Igλ loci bearing the rat CH region. J. Immunol. 190, 1481–1490 (2013).
Article CAS PubMed PubMed Central Google Scholar
Clarke, S. C. et al. Multispecific antibody development platform based on human heavy chain antibodies. Front. Immunol. 9, 3037 (2018).
Article CAS PubMed Google Scholar
Harris, K. E., et al. Sequence-based discovery demonstrates that fixed light chain human transgenic rats produce a diverse repertoire of antigen-specific antibodies. Front. Immunol. 9, 889 (2018).
Li, N., Fu, H., Hewitt, S. M., Dimitrov, D. S. & Ho, M. Therapeutically targeting glypican-2 via single-domain antibody-based chimeric antigen receptors and immunotoxins in neuroblastoma. Proc. Natl Acad. Sci. USA 114, E6623–E6631 (2017).
Article CAS PubMed PubMed Central Google Scholar
Schneider, D., et al. A unique human immunoglobulin heavy chain variable domain-only CD33 CAR for the treatment of acute myeloid leukemia. Front. Oncol. 8, 539 (2018).
Kumar, M., Keller, B., Makalou, N. & Sutton, R. E. Systematic determination of the packaging limit of lentiviral vectors. Hum. Gene Ther. 12, 1893–1905 (2001).
Article CAS PubMed Google Scholar
De Munter, S., et al. Nanobody based dual specific CARs. Int. J. Mol. Sci. 19, 403 (2018).
Conrath, K. E., Lauwereys, M., Wyns, L. & Muyldermans, S. Camel single-domain antibodies as modular building units in bispecific and bivalent antibody constructs. J. Biol. Chem. 276, 7346–7350 (2001).
Article CAS Google Scholar
Berger, C., Flowers, M. E., Warren, E. H. & Riddell, S. R. Analysis of transgene-specific immune responses that limit the in vivo persistence of adoptively transferred HSV-TK-modified donor T cells after allogeneic hematopoietic cell transplantation. Blood 107, 2294–2302 (2006).
Article CAS PubMed PubMed Central Google Scholar
Riley, J. P., Rosenberg, S. A. & Parkhurst, M. R. Identification of a new shared HLA-A2.1 restricted epitope from the melanoma antigen tyrosinase. J. Immunother. 24, 212–220 (2001).
Article CAS PubMed Google Scholar
Trinklein, N. D., Aldred, S. F., Harris, K. & Van Schooten, W. Anti-BCMA heavy chain-only antibodies. World Intellectual Property Organization Patent Application PCT/US2018/038549 (2018).
Kochenderfer, J. N., et al. Anti-B-cell maturation antigen chimeric antigen receptors with human domains. World Intellectual Property Organization Patent Application PCT/US2018/039917 (2018).
Schmitt, T. M. et al. Generation of higher affinity T cell receptors by antigen-driven differentiation of progenitor t cells in vitro. Nat. Biotechnol. 35, 1188–1195 (2017).
Article CAS PubMed PubMed Central Google Scholar
Savage, P. A., Boniface, J. J. & Davis, M. M. A kinetic basis for T cell receptor repertoire selection during an immune response. Immunity 10, 485–492 (1999).
Article CAS PubMed Google Scholar
Long, A. H. et al. 4-1BB costimulation ameliorates T cell exhaustion induced by tonic signaling of chimeric antigen receptors. Nat. Med. 21, 581–590 (2015).
Article CAS PubMed PubMed Central Google Scholar
Salem, D. A. et al. Quantification of B-cell maturation antigen, a target for novel chimeric antigen receptor T-cell therapy in Myeloma. Leuk. Res. 71, 106–111 (2018).
Article ADS CAS PubMed PubMed Central Google Scholar
Martyniszyn, A., Krahl, A. C., André, M. C., Hombach, A. A. & Abken, H. CD20-CD19 bispecific CAR T cells for the treatment of B-cell malignancies. Hum. Gene Ther. 28, 1147–1157 (2017).
Article CAS PubMed Google Scholar
Ruella, M. et al. Dual CD19 and CD123 targeting prevents antigen-loss relapses after CD19-directed immunotherapies. J. Clin. Invest. 126, 3814–3826 (2016).
Article PubMed PubMed Central Google Scholar
Schneider, D., et al. A tandem CD19/CD20 CAR lentiviral vector drives on-target and off-target antigen modulation in leukemia cell lines. J. Immunother. Cancer 5, 42 (2017).
Zah, E., Lin, M. Y., Anne, S. B., Jensen, M. C. & Chen, Y. Y. T cells expressing CD19/CD20 bispecific chimeric antigen receptors prevent antigen escape by malignant B cells. Cancer Immunol. Res. 4, 498–508 (2016).
Article CAS PubMed PubMed Central Google Scholar
Sotillo, E. et al. Convergence of acquired mutations and alternative splicing of CD19 enables resistance to CART-19 immunotherapy. Cancer Discov. 5, 1282–1295 (2015).
Article CAS PubMed PubMed Central Google Scholar
Cherkassky, L. et al. Human CAR T cells with cell-intrinsic PD-1 checkpoint blockade resist tumor-mediated inhibition. J. Clin. Invest. 126, 3130–3144 (2016).
Article PubMed PubMed Central Google Scholar
Hernandez-Chacon, J. A. et al. Costimulation through the CD137/4-1BB pathway protects human melanoma tumor-infiltrating lymphocytes from activation-induced cell death and enhances antitumor effector function. J. Immunother. 34, 236–250 (2011).
Article CAS PubMed PubMed Central Google Scholar
Kochenderfer, J. N. et al. Construction and preclinical evaluation of an anti-CD19 chimeric antigen receptor. J. Immunother. 32, 689–702 (2009).
Article CAS PubMed PubMed Central Google Scholar
Kochenderfer, J. N. et al. Lymphoma remissions caused by anti-CD19 chimeric antigen receptor T cells are associated with high serum nterleukin-15 levels. J. Clin. Oncol. 35, 1803–1813 (2017).
Article CAS PubMed PubMed Central Google Scholar
Hughes, M. S. et al. Transfer of a TCR gene derived from a patient with a marked antitumor response conveys highly active T-cell effector functions. Hum. Gene Ther. 16, 457–472 (2005).
Article CAS PubMed Google Scholar
Tran, E. et al. Cancer immunotherapy based on mutation-specific CD4+ T cells in a patient with epithelial cancer. Science 344, 641–645 (2014).
Article ADS CAS PubMed PubMed Central Google Scholar
Eshhar, Z., Waks, T., Gross, G. & Schindler, D. G. Specific activation and targeting of cytotoxic lymphocytes through chimeric single chains consisting of antibody-binding domains and the gamma or zeta subunits of the immunoglobulin and T-cell receptors. Proc. Natl Acad. Sci. USA 90, 720–724 (1993).
Article ADS CAS PubMed PubMed Central Google Scholar
Hermans, I. F. et al. The VITAL assay: a versatile fluorometric technique for assessing CTL- and NKT-mediated cytotoxicity against multiple targets in vitro and in vivo. J. Immunol. Methods 285, 25–40 (2004).
Article CAS PubMed Google Scholar
Mannering, S. I. et al. A sensitive method for detecting proliferation of rare autoantigen- specific human T cells. J. Immunol. Methods 283, 173–183 (2003).
Article CAS PubMed Google Scholar
Telford, W. G., Babin, S. A., Khorev, S. V. & Rowe, S. H. Green fiber lasers: an alternative to traditional DPSS green lasers for flow cytometry. Cytom. Part A 75, 1031–1039 (2009).
Article Google Scholar
Rabinovich, B. A. et al. Visualizing fewer than 10 mouse T cells with an enhanced firefly luciferase in immunocompetent mouse models of cancer. Proc. Natl Acad. Sci. USA 105, 14342–14346 (2008).
Article ADS CAS PubMed PubMed Central Google Scholar
Rainey, K. H. & Patterson, G. H. Photoswitching FRET to monitor protein–protein interactions. Proc. Natl Acad. Sci. USA 116, 864–873 (2019).
Article CAS PubMed Google Scholar

Download references

Acknowledgements

We thank Stanley Riddell and Carolina Berger for advice on culture methods to detect anti-CAR T-cell responses. We thank the animal care staff of the NIH Clinical Research Facility animal facility. We also thank the staff of the NCI, ETIB flow cytometry core facility. We thank Jeremy Rose and his staff for processing patient blood samples. This work was supported in part by intramural funding of the Center for Cancer Research, NCI, NIH. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government.

Author information

Authors and Affiliations

National Institutes of Health, National Cancer Institute, Center for Cancer Research, Surgery Branch, NIH Building 10 Room 3-3888, Bethesda, MD, 20892, USA
Norris Lam & James N. Kochenderfer
TeneoBio, Inc. 7999 Gateway Blvd, Newark, CA, 94560, USA
Nathan D. Trinklein & Benjamin Buelow
National Institutes of Health, National Institute of Biomedical Imaging and Bioengineering, Section on Biophotonics. NIH Building 13 Room 3E33 13 South Drive, Bethesda, MD, 20892, USA
George H. Patterson & Namrata Ojha

Authors

Norris Lam
View author publications
You can also search for this author in PubMed Google Scholar
Nathan D. Trinklein
View author publications
You can also search for this author in PubMed Google Scholar
Benjamin Buelow
View author publications
You can also search for this author in PubMed Google Scholar
George H. Patterson
View author publications
You can also search for this author in PubMed Google Scholar
Namrata Ojha
View author publications
You can also search for this author in PubMed Google Scholar
James N. Kochenderfer
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

N.L. conducted experiments, designed experiments, analyzed data, and wrote the manuscript. N.D.T. selected heavy chain binders and reviewed the manuscript, B.B. selected heavy-chain binders, aided in CAR design, and reviewed the manuscript. G.H.P. and N.O. designed and conducted biomedical imaging experiments, wrote part of the manuscript, and edited the manuscript. J.N.K. led the project, designed the CARs, analyzed data, and wrote the manuscript.

Corresponding author

Correspondence to James N. Kochenderfer.

Ethics declarations

Competing interests

N.L. is an inventor on a patent application for the FHVH CARs. (PCT/US2018/039917), N.D.T. is an employee of TeneoBio, and is an inventor on patent applications for the heavy-chain-only antibodies (PCT/US2018/038549) and for the FHVH CARs (PCT/US2018/039917). B.B. is an inventor on a patent application for the FHVH CARs (PCT/US2018/039917) and an employee of TeneoBio. J.N.K. is an inventor on a patent application for FHVH CARs (PCT/US2018/039917) and is principle investigator of an anti-BCMA CAR research agreement between the NCI and Celgene.

Additional information

Peer review information Nature Communications thanks the anonymous reviewers for their contributions to the peer review of this work.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementarty Materials

Reporting Summary

Source data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Lam, N., Trinklein, N.D., Buelow, B. et al. Anti-BCMA chimeric antigen receptors with fully human heavy-chain-only antigen recognition domains. Nat Commun 11, 283 (2020). https://doi.org/10.1038/s41467-019-14119-9

Download citation

Received: 01 May 2019
Accepted: 13 December 2019
Published: 15 January 2020
DOI: https://doi.org/10.1038/s41467-019-14119-9

This article is cited by

Recent Advances and Challenges in Cancer Treatment with Car T Cell Therapy: A Novel Anti-cancer Strategy
- Hassan Mellatyar
- Sina Sattari
- Abolfazl Akbarzadeh
BioNanoScience (2024)
BCMA-targeting chimeric antigen receptor T cell therapy for relapsed and/or refractory multiple myeloma
- Jiamin Fang
- Fuling Zhou
Annals of Hematology (2024)
Harnessing the potential of CAR-T cell therapy: progress, challenges, and future directions in hematological and solid tumor treatments
- Gunjan Dagar
- Ashna Gupta
- Ajaz A. Bhat
Journal of Translational Medicine (2023)
Long-term outcomes following CAR T cell therapy: what we know so far
- Kathryn M. Cappell
- James N. Kochenderfer
Nature Reviews Clinical Oncology (2023)
“Off-the-Shelf” Allogeneic CAR Cell Therapy—Neglected HvG Effect
- Yuxin An
- Xin Jin
- Mingfeng Zhao
Current Treatment Options in Oncology (2023)

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.