• PRAME is aberrantly expressed in childhood and adult acute myeloid leukemia and is absent in normal hematopoietic cells.

  • Mimic TCR CAR T cells effectively target PRAME in vitro and in vivo xenograft models.

Preferentially Expressed Antigen in Melanoma (PRAME), a cancer-testis antigen, provides an ideal target for immunotherapy in acute myeloid leukemia (AML). We have shown expression of PRAME in a significant subset of childhood and adult AML and lack of expression in normal hematopoiesis. Although an intracellular antigen, we developed a novel approach to target PRAME using a chimeric antigen receptor (CAR) construct encoding a targeting domain based on T-cell receptor (TCR) mimic antibodies that target the peptide-HLA complex. We used the antibody sequence from a previously designed TCR mimic (mTCR) antibody, Pr20, that recognizes the PRAME ALY peptide in complex with HLA-A∗02 and verified expression of PRAME in AML cell lines and primary AML blasts. Using the Pr20 antibody sequence, we developed CAR T cells (PRAME mTCRCAR T) to be tested against primary samples from patients with AML and AML cell lines that express the PRAME antigen in the context of HLA-A2 expression. In contrast to appropriate controls, PRAME mTCRCAR T cells demonstrate target-specific and HLA-mediated in vitro activity in OCI-AML2 and THP-1 cell lines, HLA-A2 cell lines expressing the PRAME antigen, and against primary AML patient samples. In vivo cell-derived xenograft models treated with PRAME mTCRCAR T cells demonstrated potent leukemia clearance and improved survival compared with unmodified T-cell controls. Furthermore, the cytolytic activity of PRAME mTCRCAR T cells was enhanced by treating the target cells with interferon gamma, which increases PRAME antigen expression. These results demonstrate the feasibility and efficacy of targeting PRAME with novel PRAME mTCRCAR T cells.

Adoptive transfer of T cells engineered to express chimeric antigen receptors (CARs) has achieved impressive outcomes in the treatment of refractory/relapsed B-cell acute lymphoblastic leukemia, providing a potentially curative options for these patients.1-3 The use of CAR T–cell therapy in acute myeloid leukemia (AML), however, is still in its infancy with limitations because of the innate heterogeneity associated with AML and lack of AML-specific targets for therapeutic development. Current strategies use lineage markers as therapeutic targets (ie, CD33 and CD123), which if effective, could lead to myeloablation.4-8 In an effort to identify AML-specific targets, we profiled the transcriptome of more than 2000 AML cases in children and young adults, compared with normal hematopoiesis. Preferentially Expressed Antigen in Melanoma (PRAME) was identified as one of the highest expressing genes in AML that was not expressed in peripheral blood (PB) CD34+ and bone marrow samples (Figure 1A), providing a promising target for immunotherapeutic development against AML.

Figure 1.

PRAME transcript expression in AML. (A-B) Waterfall plot showing PRAME expression (transcript per million [TPM]) in pediatric AML cohort (pediatric AML transcriptome [TpAML]) compared with normal bone marrow (NBM) and PB CD34+ samples (A) and in adult SWOG-AML cohort (B). (C) Waterfall plot showing PRAME expression (TPM) across AML subtypes in TpAML cohort.

Figure 1.

PRAME transcript expression in AML. (A-B) Waterfall plot showing PRAME expression (transcript per million [TPM]) in pediatric AML cohort (pediatric AML transcriptome [TpAML]) compared with normal bone marrow (NBM) and PB CD34+ samples (A) and in adult SWOG-AML cohort (B). (C) Waterfall plot showing PRAME expression (TPM) across AML subtypes in TpAML cohort.

Close modal

PRAME is a cancer-testis antigen, as its expression is restricted to the testes, ovaries, and endometrium in normal adult tissues.9-11 PRAME is overexpressed in a variety of cancers, including melanoma,12,13 neuroblastoma,14,15 breast,14 ovarian,16 cervical,11 and lung cancers,17-19 and hematologic malignancies.20-22 PRAME binds to the retinoic acid receptor and blocks retinoic acid–mediated proliferation arrest, differentiation, and apoptosis.23 Depending on the specific cancer type, PRAME has been shown to act as both an oncogene or a tumor suppressor gene.11 In breast cancer,24 sarcoma,11 and neuroblastoma,15 PRAME expression is associated with poor prognosis and increased risk for metastasis. In hematologic malignancies, PRAME expression has been shown to inhibit cell differentiation, growth arrest, and apoptosis, whereas in other systems it can promote cell death, reduce tumorigenicity, and increase sensitivity to chemotherapy.11,16,23 

Given its broad expression in cancer, PRAME is a promising target for immunotherapy. Because PRAME is an intracellular protein, it cannot be targeted by conventional CAR T cells that are restricted to cell surface antigens. Chang et al10 developed a T-cell receptor (TCR) mimic (mTCR) antibody called Pr20 that recognizes the peptide-HLA complex formed by the PRAME ALY peptide and HLA-A2. Here, we demonstrate that the Pr20 antigen is expressed on the cell surface in cell lines and patient samples restricted to cells with HLA-A2 expression. Using the Pr20 monoclonal antibody (mAb), we developed CAR T cells targeting the PRAME antigen in AML (referred to as PRAME mTCRCAR T cells). We show that PRAME mTCRCAR T cells demonstrate in vitro and in vivo efficacy against HLA-A2–restricted AML cells expressing the PRAME antigen, providing a novel approach to target PRAME with CAR T cells. Importantly, these results provide compelling data to support the evaluation of PRAME mTCRCAR T cells in clinical trials for refractory/relapsed AML.

Generation of PRAME CAR constructs

The CAR construct containing immunoglobulin G4 hinge and 41-BB/CD3ζ signaling domain is previously described.3 The variable light and heavy sequences from the Pr20 antibody10 is used to construct the single-chain fragment variable domain of the 41BB/CD3 CAR vector.

Generation of human PRAME mTCRCAR T cells

CAR T cells were generated by transducing healthy donor T cells (Bloodworks Northwest) with lentivirus carrying the CAR vector under the approval of Fred Hutchinson Cancer Research Center (FHCRC) Institutional Review Board (protocol 5608).25 PB mononuclear cells from healthy donors were isolated over Lymphoprep (StemCell Technologies, catalog no. 07851). CD4 or CD8 T cells were isolated by negative magnetic selection using Easy Sep Human CD4+ T-cell Isolation Kit II (StemCell Technologies, catalog no. 17952) and Easy Sep Human CD8+ T-cell Isolation Kit II (StemCell Technologies, catalog no. 17953). Purified T cells cultured in CTL medium (RPMI supplemented with 10% human serum [Bloodworks Northwest], 2% l-glutamine [Gibco, catalog no. 25030–081], 1% penicillin-streptomycin [Gibco, catalog no. 15140–122], 0.5 mol/L β-mercaptoethanol [Gibco, catalog no. 21985–023], and 50 U/mL interleukin 2 [IL-2; aldesleukin, Prometheus]) at 37°C in 5% CO2. T cells were activated with anti-CD3/CD28 beads (3:1, beads:cells; Gibco, 11131D) on Retronectin-coated plates (5 μg/mL, coated overnight at 4°C; Takara, catalog no. T100B) and transduced with CAR lentivirus (multiplicity of infection = 50) 1 day after activation via spinoculation at 800 × g for 90 minutes at 25°C in CTL medium (+50 U/mL IL-2) supplemented with 8 μg/mL protamine sulfate. Transduction used 200 000 cells per well in 24-well plates. Transduced cells were expanded in CTL medium (+50 U/mL IL-2) and separated from the beads on day 5. Because truncated CD19 was coexpressed with the CAR by a T2A ribosomal skip element, it was used to select for transduced cells, which were sorted for CD19 expression (using antihuman CD19 phycoerythrin [PE; BioLegend, catalog no. 982402]) on FACSAria II, 8 to 10 days after activation. Sorted cells were further expanded in CTL (+50 U/mL IL-2) medium before in vitro and in vivo cytotoxicity assays.

Cell lines

OCI-AML2, THP-1, K562, MV4;11, and RS4;11 cells were obtained from the American Type Culture Collection and maintained per manufacturer’s instructions. MV4;11 and RS4;11 cells were transduced with an HLA-A2 expression construct to generate MV4;11 HLA-A2+ and RS4;11 HLA-A2+ cell lines.

Primary samples

Frozen aliquots of mononuclear CD45+ cells isolated from AML diagnostic bone marrow samples were obtained from the Children’s Oncology Group.25 Freshly thawed aliquots were assessed for PRAME by flow cytometry. Some samples were sent to Hematologics, Inc. for immunophenotype analysis of PRAME expression. All specimens used in this study were obtained after written consent from patients and donors. The research was performed after approval by the FHCRC Institutional Review Board (protocol #5608). The study was conducted in accordance with the Declaration of Helsinki.

In vitro studies

Target cells (OCI-AML2, THP-1, and K562) were split 1 to 2 days before the cytotoxicity assay. Target cells were labeled with 2.5 μmol/L carboxyfluorescein succinimidyl ester (CFSE; Invitrogen, catalog no. C34554) per manufacturer’s directions, washed with 1× phosphate-buffered saline (PBS), and resuspended in CTL medium (without IL-2). For T-cell proliferation assay, effector cells (unmodified or PRAME mTCRCAR T cells) were labeled with 2.5 mol/L Violet Cell Proliferation Dye (Invitrogen, catalog no. C34557), washed with 1× PBS, serially diluted in CTL medium (without IL-2), and combined with target cells at various effector-to-target (E:T) ratios in a 96-well U-bottom plate.25 Cytotoxicity was assessed by flow cytometry after staining cells with live or dead fixable viability dyes (FVD; Invitrogen, catalog no. L34964). The percentage of dead cells among target cells was assessed by gating on FVD+ among CFSE+ target cells. For primary patient samples, the percentage of live cells were assessed by gating on FVD among CFSE+ target cells. Percent-specific lysis was calculated by subtracting the average of the 3 replicate wells containing target cells only from each well containing target and effector cells at each E:T ratio. After 24 hours of coculture, media supernatant was assessed for IL-2, interferon gamma (IFN-γ), and tumor necrosis factor α (TNF-α) production by Luminex microbead technology (provided by FHCRC Immune Monitoring Core).

In vivo studies

For cell line–derived xenograft models, we transduced OCI-AML2, THP-1, and K562 cells with green fluorescent protein/luciferase construct (plasmid #104834, Addgene) and sorted for green fluorescent protein positive cells. Luciferase-expressing cells were injected IV into NSG mice through the tail vein at 1 × 106 cells per mouse.25 Mice were treated with 5 × 106 cells (1:1, CD4:CD8) per mouse of either PRAME mTCRCAR or unmodified T cells via tail vein IV injection 7 days after the injection of OCI-AML2, THP-1, and K562 cells. Leukemia burden was measured by bioluminescence imaging weekly. Leukemia burden and T-cell expansion were monitored by flow cytometric analysis of mouse PB drawn by retro-orbital bleeds for the indicated time points starting from the first week of T-cell injection. Mice were monitored and euthanized when they exhibited symptomatic leukemia (tachypnea, hunchback, persistent weight loss, fatigue, or hind-limb paralysis). Tissues (blood, bone marrow, liver, spleen, and tumors) were harvested at necropsy and analyzed for the presence of T and leukemia cells. This study was performed after approval by the FHCRC institutional animal care and use committee (protocol #51068).

Flow cytometry of xenograft cells

PE-conjugated Pr20 antibody was used to confirm PRAME expression on target cell lines and primary samples from patients with AML. Cell lines and primary samples were washed in 2% fetal bovine serum (FBS) in PBS, blocked with 20 ug/mL Fc receptor block (BD Pharmingen, catalog no. 564219) in PBS, then stained with PE-conjugated antihuman Pr20 antibody and allophycocyanin (APC)–conjugated HLA-A2 antibody for 20 minutes on ice. Labeled cells were washed with PBS and resuspended in 2% FBS/PBS before flow cytometric analysis.

Tissues were harvested at necropsy and passed through a 70 μm cell strainer to dissociate tissues into single cells before antibody staining. Cells from mouse PB were processed with red blood cell lysis buffer, washed in 2% FBS in PBS, blocked with 20 ug/mL Fc receptor block (BD Pharmingen, catalog no. 564219) in PBS, then stained with a cocktail of fluorescently labeled mAbs that included a combination of APC/cyanine 7–conjugated anti-mouse CD45.1 (BioLegend, catalog no. 110716), BUV805-conjugated antihuman CD45 (BD Biosciences, catalog no. 612891), APC-conjugated antihuman CD19 (BD Biosciences, catalog no. 555415), PE-cyanine 7–conjugated antihuman CD3 (BD Biosciences, catalog no. 563423), BV605-conjugated antihuman CD4 (BioLegend, catalog no. 317438), BV711-conjugated antihuman CD8 (BD Biosciences, catalog no. 563677), peridinin chlorophyll protein complex/cyanine 5.5–conjugated antihuman CD33 (BioLegend, catalog no. 303414), and PE-conjugated antihuman Pr20 for 20 minutes on ice.25 Labeled cells were washed with PBS and resuspended in 2% FBS/PBS before flow cytometric analysis. Symphony with FACSDiva software (BD Biosciences) was used to assess cell surface expressions, and FlowJo software was used for the analysis. Dead cells were excluded using 4′,6-diamidino-2-phenylindole staining.

Statistical analysis

Unpaired, two-tailed Student t test was used to determine statistical significance for all in vitro studies. Log-rank (Mantel-Cox) test was used to compare the Kaplan-Meier survival curves between experimental groups. P values of <.05 were statistically significant.

PRAME transcript is expressed in AML

Analysis of TpAML identified PRAME to be highly expressed in a subset of leukemias without expression in normal CD34+ PB or bone marrow samples (Figure 1A). Defining PRAME-positive leukemia at TPM >5, we found 451 out of 1493 (30%) pediatric leukemias are positive for PRAME expression. Among PRAME-positive leukemias, the median expression was 24.74 TPM (range: 5.02-230.53). In adult AML (SWOG-AML), PRAME was also expressed, albeit at a lower prevalence (Figure 1B). To determine whether PRAME expression correlates with specific AML-associated molecular alterations, we evaluated PRAME expression by fusion and mutation groups (Figure 1C). PRAME transcript was detected in all AML subtypes, with expression in 31% of KMT2A-r AML, 23% in Inv(16) and high enrichment in t(8;21) (83% PRAME+). Furthermore, patients with AML with t(8;21) had significantly higher PRAME expression compared with all other groups (median expression, range: 41.87, 5.47-230.53 for t(8;21); 24.1, 5.16-113.74 for inv16; 19.55, 5.14-228.58 for KMT2A-r; and 17.82, 5.02-139.19 for other). We evaluated PRAME expression as compared with age and found that younger (aged <3 years, P < .001) and older (aged >18 years, P = .024) patients had lower PRAME expression, but there was no statistical difference among the other age groups nor was there a correlation with the outcome (supplemental Figure 1). These results corroborate previous reports demonstrating PRAME expression in AML and further define the unique expression of PRAME, especially in t(8;21) AML.

PRAME/HLA-A2 complex is expressed on the cell surface of AML blasts

We developed a TCR mimic antibody that recognizes a specific PRAME peptide (PRAME ALY peptide) when presented in complex with HLA-A2 using the Pr20 antibody sequence previously described.10 We verified the specificity of the Pr20 mAb against AML cell lines. As expected, Pr20 bound to HLA-A2 positive THP-1 and OCI-AML2 that express PRAME (Figure 2A). Pr20 did not recognize HLA-A2–transduced MV4;11 cells, which lack PRAME expression. Pr20 also did not bind to K562 and RS4;11 cell lines, which express PRAME but not HLA-A2. However, RS411 cells transduced with an HLA-A2 expression construct were recognized by the Pr20 mAb demonstrating that Pr20 specificity is dependent on PRAME expression in the context of HLA-A2. These results confirm the specificity of Pr20 against HLA-A2 positive leukemias that express PRAME.

Figure 2.

Pr20 antigen expression in cell lines and primary AML blasts. (A) Flow cytometric analysis of Pr20 binding (red) in cell lines expressing both PRAME and HLA-A2 (THP-1, OCI-AML2, and RS4;11 HLA-A2+) or expressing only PRAME (K562 and parental RS4;11) or HLA-A2 (MV4;11 HLA-A2+) compared with isotype control (gray; top). Expression of HLA-A2 (blue) compared with isotype control (gray) for the indicated cell lines. Data representative of 3 independent experiments (bottom). (B) Flow cytometric analysis of Pr20 antigen expression in 3 patient index cases.

Figure 2.

Pr20 antigen expression in cell lines and primary AML blasts. (A) Flow cytometric analysis of Pr20 binding (red) in cell lines expressing both PRAME and HLA-A2 (THP-1, OCI-AML2, and RS4;11 HLA-A2+) or expressing only PRAME (K562 and parental RS4;11) or HLA-A2 (MV4;11 HLA-A2+) compared with isotype control (gray; top). Expression of HLA-A2 (blue) compared with isotype control (gray) for the indicated cell lines. Data representative of 3 independent experiments (bottom). (B) Flow cytometric analysis of Pr20 antigen expression in 3 patient index cases.

Close modal

We next evaluated binding of Pr20 in AML cells from patients. In 3 index cases, we detected near uniform expression of PRAME on AML blasts (Figure 2B) with additional patient samples (N = 224) showing cell surface PRAME antigen detected by the Pr20 antibody. Among the Pr20-positive cases (defined by expression above autofluorescence in >20% of AML blasts), PRAME antigen was moderately to highly expressed with median mean fluorescence intensity of 30.6 (range: 4.4-1174.9).

PRAME mTCRCAR induces potent cytotoxicity and cytokine production in vitro

Having verified cell surface expression of PRAME in AML cell lines and patient samples, we investigated whether PRAME-positive cells can be targeted with PRAME-specific CAR T cells. We developed a Pr20-specific CAR by reformatting the sequences from the Pr20 mAb into a single-chain variable fragment (scFv; “Materials and methods”) and incorporated the scFv into the standard CAR with 41-BB costimulatory and CD3zeta signaling domains (Figure 3A). We tested the cytotoxicity of the CAR T cells directed at the Pr20 antigen (PRAME mTCRCAR T cells) against OCI-AML2, THP-1, and K562 cells. Coincubation of CD8 PRAME mTCRCAR T cells with OCI-AML2 and THP-1 cells for 24 hours at the indicated E:T ratios resulted in robust killing, whereas control, unmodified CD8 T cells did not result in cytotoxicity of the target cells (Figure 3B and supplemental Figure 2A). Consistent with specificity, K562 cells were not susceptible to CD8 PRAME mTCRCAR T cell–mediated killing. We then tested the cytotoxicity of the CAR T cells against primary blasts from patients with AML. Coincubation of CD8 PRAME mTCRCAR T cells with PRAME+/HLA-A2+ primary AML blasts resulted in specific killing, most notably at the higher E:T ratios (Figure 3C). PRAME+/HLA-A2 primary AML blasts coincubated with CD8 PRAME mTCRCAR T cells did not demonstrate CAR T cell–mediated lysis, confirming specificity in primary samples from patients (Figure 3C). To further demonstrate the reactivity of PRAME mTCRCAR T cells, we measured cytokine production after 24 hours of coincubation with OCI-AML2 and THP-1 cells with CD4 and CD8 PRAME mTCRCAR T cells at 1:1, E:T ratio. Both CD4 and CD8 CAR T cells produced significant levels of proinflammatory cytokines (IL-2, IFN-γ, and TNF-α) when coincubated with OCI-AML2 and THP-1 cells (Figure 3D; supplemental Figure 2B) with no cytokine production noted when coincubated with K562 cells (results not shown). Together, these results demonstrate the antigen-dependent reactivity of PRAME mTCRCAR T cells against AML cell lines and primary samples from patients expressing PRAME.

Figure 3.

PRAME mTCRCAR T cells demonstrate in vitro efficacy against Pr20-positive AML cells. (A) Diagram of PRAME mTCRCAR construct. (B) Cytolytic activity of CD8 T cells unmodified or transduced with PRAME mTCRCAR construct after 24 hours coculture with THP-1, OCI-AML2, and K562 cells. Data presented are mean specific lysis ± SD from 3 technical replicates at indicated E:T ratios. (C) Cytolytic activity of CD8 T cells unmodified or transduced with PRAME mTCRCAR construct after 24 hours coculture with primary AML blasts. Data presented are mean specific lysis ± SD from 3 technical replicates at indicated E:T ratios. (D) Concentration of secreted IL-2, IFN-γ, and TNF-α in the supernatant following 24 hour of coculture with CD4 or CD8 T cells at 1:1 E:T ratio as measured by enzyme-linked immunosorbent assay. Where concentrations of cytokines are too low to discern, the number above the x-axis indicates the average concentration. Statistical significance was determined by unpaired Student t test, assuming unequal variances. ∗P < .05, ∗∗P < .005, ∗∗∗P < .0005. Data are representative of 2 donors. CD, costimulatory domain; SD, stimulatory domain; SP, GM-CSFR signal peptide; tCD19, truncated CD19; TM, transmembrane domain.

Figure 3.

PRAME mTCRCAR T cells demonstrate in vitro efficacy against Pr20-positive AML cells. (A) Diagram of PRAME mTCRCAR construct. (B) Cytolytic activity of CD8 T cells unmodified or transduced with PRAME mTCRCAR construct after 24 hours coculture with THP-1, OCI-AML2, and K562 cells. Data presented are mean specific lysis ± SD from 3 technical replicates at indicated E:T ratios. (C) Cytolytic activity of CD8 T cells unmodified or transduced with PRAME mTCRCAR construct after 24 hours coculture with primary AML blasts. Data presented are mean specific lysis ± SD from 3 technical replicates at indicated E:T ratios. (D) Concentration of secreted IL-2, IFN-γ, and TNF-α in the supernatant following 24 hour of coculture with CD4 or CD8 T cells at 1:1 E:T ratio as measured by enzyme-linked immunosorbent assay. Where concentrations of cytokines are too low to discern, the number above the x-axis indicates the average concentration. Statistical significance was determined by unpaired Student t test, assuming unequal variances. ∗P < .05, ∗∗P < .005, ∗∗∗P < .0005. Data are representative of 2 donors. CD, costimulatory domain; SD, stimulatory domain; SP, GM-CSFR signal peptide; tCD19, truncated CD19; TM, transmembrane domain.

Close modal

PRAME mTCRCAR T cells demonstrate potent leukemia clearance and improved survival in vivo

To evaluate the in vivo efficacy of the PRAME mTCRCAR T cells, we generated human leukemia xenograft models by injecting NSG mice with OCI-AML2, THP-1, and K562 cells transduced with a luciferase expression construct at 1 × 106 cells per mouse. One week after leukemia injection, we treated the leukemia-bearing mice with unmodified or PRAME mTCRCAR T cells at 5 × 106 T cells per mouse with 1:1 ratio of CD4 and CD8 T cells. We monitored leukemia burden by bioluminescence (IVIS) imaging. Treatment with PRAME mTCRCAR T cells led to leukemia clearance in OCI-AML2–bearing mice, which remained disease free after CAR T–cell injection for the entire duration of the study (Figure 4A-B, left). In contrast, OCI-AML2–bearing mice treated with unmodified T cells exhibited disease progression that led to symptomatic leukemia. The PRAME mTCRCAR T cells significantly reduced the growth of THP-1 cells but did not completely eradicate the leukemia in vivo (Figure 4A-B, middle). As expected, treatment with PRAME mTCRCAR T cells did not affect leukemia growth in K562 xenografts (Figure 4A-B, right). Consistent with the robust antileukemia activity of PRAME mTCRCAR T cells in OCI-AML2–bearing mice, we detected significant expansion of the CAR T cells in the PB compared with unmodified T cells at day 6 after T-cell injection (Figure 4C, left). There was no difference in T-cell engraftment between unmodified and CAR T cells in THP-1 and K562-bearing mice (Figure 4C, middle and right, respectively). We confirmed that the CAR T cells that engrafted in the OCI-AML2 xenografts expressed the transduction marker, truncated CD19 (supplemental Figure 3). Importantly, the antileukemia activity of PRAME mTCRCAR T cells led to a significant increase in survival for OCI-AML2 (P = .003) and THP-1 (P = .005) but not K562 xenografts (Figure 4D). Together, these results demonstrate the in vivo efficacy of PRAME mTCRCAR T cells against PRAME+/HLA-A2+ but not PRAME+/HLA-A2 leukemias.

Figure 4.

PRAME mTCRCAR T cells eliminate Pr20-positive AML cells in vivo. (A) Bioluminescent imaging of OCI-AML2–, THP-1-, and K562-bearing mice treated with unmodified or PRAME mTCRCAR T cells, 5 × 106 T cells per mouse. Shown are representative images at the indicated time points. (B) Disease burden measured by radiance from OCI-AML2, THP-1, and K562 xenograft mice treated with unmodified or PRAME mTCRCAR T cells. Shown is the average radiance and standard error of the mean. N = 5 mice per group for OCI-AML2 and THP-1 models and N = 3 for K562. (C) T-cell expansion of unmodified and PRAME mTCRCAR T cells 6 days after T-cell infusion from OCI-AML2, THP-1, and K562 xenograft mice. (D) The Kaplan-Meier survival curves of OCI-AML2, THP-1, and K562 xenograft mice treated with unmodified or PRAME mTCRCAR T cells. Statistical differences in survival were evaluated using log-rank Mantel-Cox tests. n.s., not significant.

Figure 4.

PRAME mTCRCAR T cells eliminate Pr20-positive AML cells in vivo. (A) Bioluminescent imaging of OCI-AML2–, THP-1-, and K562-bearing mice treated with unmodified or PRAME mTCRCAR T cells, 5 × 106 T cells per mouse. Shown are representative images at the indicated time points. (B) Disease burden measured by radiance from OCI-AML2, THP-1, and K562 xenograft mice treated with unmodified or PRAME mTCRCAR T cells. Shown is the average radiance and standard error of the mean. N = 5 mice per group for OCI-AML2 and THP-1 models and N = 3 for K562. (C) T-cell expansion of unmodified and PRAME mTCRCAR T cells 6 days after T-cell infusion from OCI-AML2, THP-1, and K562 xenograft mice. (D) The Kaplan-Meier survival curves of OCI-AML2, THP-1, and K562 xenograft mice treated with unmodified or PRAME mTCRCAR T cells. Statistical differences in survival were evaluated using log-rank Mantel-Cox tests. n.s., not significant.

Close modal

IFN-γ treatment increased PRAME/HLA-A2 antigen expression and cytolytic activity of PRAME mTCRCAR T cells

IFN-γ is known to enhance presentation of tumor-associated antigens by upregulating the immunoproteosome and catalyzing nondestructive cleavage, thereby enhancing major histocompatibility complex peptide surface expression leading to increased T-cell recognition and cytotoxicity.10,26-30 To determine whether IFN-γ enhances PRAME antigen expression, we treated OCI-AML2, THP-1, and K562 cells with IFN-γ for 72 hours and assessed PRAME and HLA-A2 expression by flow cytometry. As anticipated, treatment with IFN-γ resulted in increased levels of both HLA-A2 and PRAME on OCI-AML2 and THP-1 cells (Figure 5A, left and middle, respectively). IFN-γ treatment did not affect PRAME and HLA-A2 expression in K562 cells (Figure 5A, right). Given that IFN-γ enhanced PRAME expression, we next investigated whether OCI-AML2 and THP-1 cells pretreated with IFN-γ would be more susceptible to cytotoxicity of PRAME mTCRCAR T cells. We incubated OCI-AML2 and THP-1 cells with IFN-γ for 72 hours, washed them, and then coincubated them with unmodified or PRAME mTCRCAR T cells at various E:T ratios. We assayed T-cell killing after 16 hours of coincubation. Pretreatment of target cells with IFN-γ resulted in enhanced cytolytic activity of PRAME mTCRCAR T cells (Figure 5B, left and middle). This enhanced activity is dependent on PRAME antigen expression as no significant activity was detected in unmodified T cells coincubated with THP-1 and OCI-AML2 that were either untreated or pretreated with IFN-γ. Consistent with target specificity, K562 cells either untreated or pretreated with IFN-γ were not sensitive to the cytolytic activity of PRAME mTCRCAR T cells (Figure 5B, right).

Figure 5.

IFN-γ treatment enhances Pr20 antigen and HLA-A2 expression and increase the cytolytic activity of PRAME mTCRCAR T cells. (A-B) Flow cytometric analysis of Pr20 antigen (A) and HLA-A2 (B) expression in OCI-AML2, THP-1, and K562 treated with dimethyl sulfoxide (blue) or IFN-gamma (10 ng/mL, red) for 3 days. Data representative of 3 independent experiments. (C) Cytolytic activity of unmodified or PRAME mTCRCAR CD8 T cells cocultured for 16 hours with THP-1, OCI-AML2, and K562 cells after pretreatment with IFN-γ (10 ng/mL) or dimethyl sulfoxide control. Data presented are mean specific lysis ± SD from 3 technical replicates at indicated E:T ratios. Solid blue and red lines represent IFN-γ treatment vs dashed lines representing untreated controls.

Figure 5.

IFN-γ treatment enhances Pr20 antigen and HLA-A2 expression and increase the cytolytic activity of PRAME mTCRCAR T cells. (A-B) Flow cytometric analysis of Pr20 antigen (A) and HLA-A2 (B) expression in OCI-AML2, THP-1, and K562 treated with dimethyl sulfoxide (blue) or IFN-gamma (10 ng/mL, red) for 3 days. Data representative of 3 independent experiments. (C) Cytolytic activity of unmodified or PRAME mTCRCAR CD8 T cells cocultured for 16 hours with THP-1, OCI-AML2, and K562 cells after pretreatment with IFN-γ (10 ng/mL) or dimethyl sulfoxide control. Data presented are mean specific lysis ± SD from 3 technical replicates at indicated E:T ratios. Solid blue and red lines represent IFN-γ treatment vs dashed lines representing untreated controls.

Close modal

Identification of targets whose expression is limited to AML leukemic cells and silent in normal hematopoiesis provides an opportunity for effective therapies with limited to no hematopoietic toxicity. Through comprehensive analysis of the AML transcriptome from more than 2000 patients, we identified PRAME as one such AML-restricted target with no expression in normal hematopoiesis. In this study, we prioritized the intracellular protein PRAME as a target for modified CAR T development based on a TCR mimic antibody (Pr20) previously developed by Chang et al, which has been reported to have high binding affinity (4-5 nM KD).10 We used the Pr20 mAb that recognizes the PRAME ALY peptide in complex with HLA-A2, confirmed the specificity of the Pr20 antibody against PRAME+/HLA-A2+ leukemias, and further demonstrated that Pr20 recognizes PRAME/HLA-A2 in primary AML blasts. Using this antibody sequence, we developed PRAME mTCRCAR T cells against AML cells expressing PRAME. We show that PRAME mTCRCAR T cells exhibit preclinical efficacy in eliminating AML cells in vitro and in vivo. This work highlights the therapeutic potential of targeting PRAME in AML and provides a novel approach to target intracellular antigens with CAR T cells.

PRAME represents a promising target for immunotherapy, as its expression is limited to the reproductive tissues9-11 and is broadly expressed in many cancers, including AML.12-22 In our study, we show that PRAME is expressed in a substantial number of pediatric and adult AML. HLA-A2 is the most common HLA-I subtype, found in 47% of pediatric patients with AML, thus targeting HLA-A2+ patients will benefit a substantial portion of patients. PRAME and HLA-A2 are regulatable pharmacologically;10,31-33 therefore, future work might involve upregulating the epitope with a combination therapy. Interestingly, PRAME expression is enriched in t(8;21) AML, in which nearly all of t(8;21) leukemias express PRAME suggesting that the fusion protein AML1/ETO may directly or indirectly promote PRAME expression in this type of leukemia. Importantly, we did not detect PRAME transcript expression in normal PB CD34+ and bone marrow samples, suggesting that targeting PRAME would not affect normal hematopoiesis. There are numerous clinical trials using PRAME-specific TCR therapy with limited toxicity supporting the notion that PRAME is a specific tumor cell antigen with limited systemic expression.34-38 

Given the successes of CAR T cells in treating B-cell malignancies, we developed PRAME mTCRCAR T cells targeting the Pr20 antigen formed by the PRAME ALY peptide-HLA-A2 complex. Although PRAME mTCRCAR T cells completely eradicated OCI-AML2 leukemia in vivo, they were less effective against THP-1 leukemia. We hypothesize this is because of lower antigen density in THP-1 cells (mean fluorescence intensity, fold change to isotype control: 2.75-fold [THP-1] vs fivefold [OCI-AML2]), although antigen density was not directly studied. In contrast to OCI-AML2 cells, with T-cell expansion noted on day 6, we hypothesize the lower antigen density on THP-1 cells led to a delay in T-cell expansion, but ongoing exposure of PRAME-expressing cells resulted in sufficient T-cell activation, leading to reduction in leukemia burden and prolonged survival. Strategies that enhance PRAME antigen expression would result in increased efficacy of PRAME mTCRCAR T cells. IFN-γ treatment increases expression of the PRAME ALY-HLA-A2 complex by inducing nondestructive cleavage sites by the immunoproteosome.10 Here, we demonstrate that treatment with IFN-γ resulted in increased expression of PRAME/HLA-A2, leading to improved cytolytic activity of PRAME mTCRCAR T cells against OCI-AML2 and THP-1 cells. IFN-γ has been shown to be well tolerated in prior clinical trials as monotherapy and in combination with chemotherapy.39-41 We propose that IFN-γ may provide a useful strategy to increase efficacy of PRAME mTCRCAR T cells and should be evaluated in future PRAME mTCRCAR T–cell therapy clinical trials for AML. In addition, incorporating different CAR elements such as scFv binding and signaling components (ie, CD28z) can increase antileukemic sensitivity and efficacy,42 especially against AML cells with low antigen density, and should be evaluated to optimize the efficacy of PRAME mTCRCAR T cells. Additional assays to determine minimum antigen density required for CAR T efficacy can be considered before the implementation of PRAME mTCRCAR T cell in clinical use.

The PRAME mTCRCAR T cell is one of a handful of mTCRCAR T cells that have been reported thus far. The first mTCRCAR T cell developed was against MAGE-A1 presented on HLA-A1 in 2001,43 and since then, a number of mTCRCAR T cells were reported, including CAR T cells targeting NY-ESO-1:HLA-A∗02,44-46 WT1:HLA-A∗02,47 WT1:HL-A∗24,48 GP100:HLA-A∗02,49 AFP:HLA-A∗02,50 and recently NDC80:HLA-A∗02.51 Although CAR T cells equipped with TCR mimic antibodies provide a strategy to target intracellular antigens, many considerations will need to be addressed before this new class of CAR T cells can be delivered to patients. One such barrier is the low density of peptide-to-major histocompatibility complex expressed on target cancer cells as discussed earlier. Some guidelines to fine-tune TCR mimic antibodies to increase their affinity have been described that may help improve the sensitivity of mTCRCAR T cells.52,53 Second, concerns of on-target/off-tumor and off-target toxicities may limit this approach, as such toxicities were observed in affinity-matured TCR T-cell therapy against MAGE-A3. To prevent these toxicities in patients, a systematic screening system for cross-reactivity of PRAME mTCRCAR T cells should be evaluated.

In this study, we demonstrate the therapeutic potential of targeting PRAME with mTCRCAR T cells in AML. We show potent, target-specific reactivity of PRAME mTCRCAR T cells against PRAME+/HLA-A2+ AML cell lines but not PRAME+/HLA-A2 K562 cell lines, both in vitro and in vivo. In addition, we demonstrate the target-specific cytolytic activity of PRAME mTCRCAR T cells against PRAME+/HLA-A2+ primary samples from patients with AML. Samples from patients grow poorly under cell culture conditions. Therefore, increased background cell death was noted in the patient samples treated with both unmodified CAR T cells and PRAME mTCRCAR T cells. To overcome this caveat and evaluate the efficacy of PRAME mTCRCAR T cells in samples from patients, in vivo treatment of a patient-derived xenograft model should be explored. The results presented provide a novel approach to target PRAME with mTCRCAR T cells and compelling data to further evaluate PRAME mTCRCAR T cells in AML clinical trials. As part of discovery efforts in the current COG Phase 3 AML1831 trial, all diagnostic specimens from patients enrolled are prospectively tested for PRAME expression at Hematologics, Inc. This effort will establish prevalence of patients who would be eligible for a future clinical trial with our PRAME mTCRCAR T cells. This work also supports further development of mTCRCAR T cells against other intracellular AML-restricted antigens in context of HLA-A2 and other HLA molecules.

This study was supported by the Leukemia and Lymphoma Society and St. Baldrick’s Foundation.

Contribution: D.C.K. conducted experiments, formal data analysis, and wrote the manuscript; A.M.L. conducted experiments, formal data analysis, and reviewed and edited the manuscript; S.C., L. Pardo, C.N.M., L. Perkins, and T.T.T. conducted the experiments; A.R.L. provided statistical analysis; D.A.S. was involved in conceptualization; S.M., M.R.L., L.E.B., K.R.L. and D.A.S. provided general guidance; and Q.L. and S.M. conceptualized and designed the experiments, conducted formal data analysis, and provided supervision for reviewing and editing the manuscript.

Conflict-of-interest disclosure: M.R.L. and L.E.B. are employees of and have equity ownership in Hematologics Inc. L. Pardo is an employee of Hematologics Inc. D.A.S. is on a board of, or has equity in, or income from Lantheus, Sellas, Iovance Biotherapeutics, Pfizer, Actinium Pharmaceuticals, OncoPep, Repertoire, Sapience, CoImmune, and Eureka Therapeutics.

Correspondence: Danielle C. Kirkey, Fred Hutchinson Cancer Center, 1100 Fairview Ave N., Seattle, WA 98109; e-mail: dkirkey@fredhutch.org.

1.
Gardner
RA
,
Finney
O
,
Annesley
C
, et al
.
Intent-to-treat leukemia remission by CD19 CART cells of defined formulation and dose in children and young adults
.
Blood
.
2017
. ;
129
(
25
):
3322
-
3331
.
2.
Maude
SL
,
Frey
N
,
Shaw
PA
, et al
.
Chimeric antigen receptor T cells for sustained remissions in leukemia
. [published correction appears in N Engl J Med. 2016 Mar 10;374(10):998].
N Engl J Med
.
2014
. ;
371
(
16
):
1507
-
1517
.
3.
Turtle
CJ
,
Hanafi
LA
,
Berger
C
, et al
.
CD19 CAR-T cells of defined CD4+:CD8+ composition in adult B cell ALL patients
.
J Clin Invest
.
2016
. ;
126
(
6
):
2123
-
2138
.
4.
Gill
S
,
Tasian
SK
,
Ruella
M
, et al
.
Preclinical targeting of human acute myeloid leukemia and myeloablation using chimeric antigen receptor-modified T cells
. [published correction appears in Blood. 2016 Nov 24;128(21):2585].
Blood
.
2014
. ;
123
(
15
):
2343
-
2354
.
5.
Leong
SR
,
Sukumaran
S
,
Hristopoulos
M
, et al
.
An anti-CD3/anti-CLL-1 bispecific antibody for the treatment of acute myeloid leukemia
.
Blood
.
2017
. ;
129
(
5
):
609
-
618
.
6.
Mardiros
A
,
Dos Santos
C
,
McDonald
T
, et al
.
T cells expressing CD123-specific chimeric antigen receptors exhibit specific cytolytic effector functions and antitumor effects against human acute myeloid leukemia
.
Blood
.
2013
. ;
122
(
18
):
3138
-
3148
.
7.
Pizzitola
I
,
Anjos-Afonso
F
,
Rouault-Pierre
K
, et al
.
Chimeric antigen receptors against CD33/CD123 antigens efficiently target primary acute myeloid leukemia cells in vivo
.
Leukemia
.
2014
. ;
28
(
8
):
1596
-
1605
.
8.
Tashiro
H
,
Sauer
T
,
Shum
T
, et al
.
Treatment of acute myeloid leukemia with T cells expressing chimeric antigen receptors directed to C-type lectin-like molecule 1
.
Mol Ther
.
2017
. ;
25
(
9
):
2202
-
2213
.
9.
Al-Khadairi
G
,
Decock
J
.
Cancer testis antigens and immunotherapy: where do we stand in the targeting of PRAME?
.
Cancers (Basel)
.
2019
. ;
11
(
7
):
984
. Published 2019 Jul 15.
10.
Chang
AY
,
Dao
T
,
Gejman
RS
, et al
.
A therapeutic T cell receptor mimic antibody targets tumor-associated PRAME peptide/HLA-I antigens
.
J Clin Invest
.
2017
. ;
127
(
9
):
3557
.
11.
Xu
Y
,
Zou
R
,
Wang
J
,
Wang
ZW
,
Zhu
X
.
The role of the cancer testis antigen PRAME in tumorigenesis and immunotherapy in human cancer
.
Cell Prolif
.
2020
. ;
53
(
3
):
e12770
.
12.
Wadelin
F
,
Fulton
J
,
McEwan
PA
,
Spriggs
KA
,
Emsley
J
,
Heery
DM
.
Leucine-rich repeat protein PRAME: expression, potential functions and clinical implications for leukaemia
.
Mol Cancer
.
2010
. ;
9
:
226
.
13.
Ikeda
H
,
Lethé
B
,
Lehmann
F
, et al
.
Characterization of an antigen that is recognized on a melanoma showing partial HLA loss by CTL expressing an NK inhibitory receptor
.
Immunity
.
1997
. ;
6
(
2
):
199
-
208
.
14.
Epping
MT
,
Bernards
R
.
A causal role for the human tumor antigen preferentially expressed antigen of melanoma in cancer
.
Cancer Res
.
2006
. ;
6
(
22
):
10639
-
10642
.
15.
Oberthuer
A
,
Hero
B
,
Spitz
R
,
Berthold
F
,
Fischer
M
.
The tumor-associated antigen PRAME is universally expressed in high-stage neuroblastoma and associated with poor outcome
.
Clin Cancer Res
.
2004
. ;
10
(
13
):
4307
-
4313
.
16.
Tajeddine
N
,
Gala
JL
,
Louis
M
,
Van Schoor
M
,
Tombal
B
,
Gailly
P
.
Tumor-associated antigen preferentially expressed antigen of melanoma (PRAME) induces caspase-independent cell death in vitro and reduces tumorigenicity in vivo
.
Cancer Res
.
2005
. ;
65
(
16
):
7348
-
7355
.
17.
Thongprasert
S
,
Yang
PC
,
Lee
JS
, et al
.
The prevalence of expression of MAGE-A3 and PRAME tumor antigens in East and South East Asian non-small cell lung cancer patients
.
Lung Cancer
.
2016
. ;
101
:
137
-
144
.
18.
Bankovic
J
,
Stojsic
J
,
Jovanovic
D
, et al
.
Identification of genes associated with non-small-cell lung cancer promotion and progression
.
Lung Cancer
.
2010
. ;
67
(
2
):
151
-
159
.
19.
Pan
SH
,
Su
KY
,
Spiessens
B
, et al
.
Gene expression of MAGE-A3 and PRAME tumor antigens and EGFR mutational status in Taiwanese non-small cell lung cancer patients
.
Asia Pac J Clin Oncol
.
2017
. ;
13
(
5
):
e212
-
e223
.
20.
Greiner
J
,
Ringhoffer
M
,
Taniguchi
M
, et al
.
mRNA expression of leukemia-associated antigens in patients with acute myeloid leukemia for the development of specific immunotherapies
.
Int J Cancer
.
2004
. ;
108
(
5
):
704
-
711
.
21.
Ding
K
,
Wang
XM
,
Fu
R
,
Ruan
EB
,
Liu
H
,
Shao
ZH
.
PRAME gene expression in acute leukemia and its clinical significance
.
Cancer Biol Med
.
2012
. ;
9
(
1
):
73
-
76
.
22.
Radich
JP
,
Dai
H
,
Mao
M
, et al
.
Gene expression changes associated with progression and response in chronic myeloid leukemia
.
Proc Natl Acad Sci U S A
.
2006
. ;
103
(
8
):
2794
-
2799
.
23.
Epping
MT
,
Wang
L
,
Edel
MJ
,
Carlée
L
,
Hernandez
M
,
Bernards
R
.
The human tumor antigen PRAME is a dominant repressor of retinoic acid receptor signaling
.
Cell
.
2005
. ;
122
(
6
):
835
-
847
.
24.
Doolan
P
,
Clynes
M
,
Kennedy
S
,
Mehta
JP
,
Crown
J
,
O'Driscoll
L
.
Prevalence and prognostic and predictive relevance of PRAME in breast cancer
.
Breast Cancer Res Treat
.
2008
. ;
109
(
2
):
359
-
365
.
25.
Le
Q
,
Castro
S
,
Tang
T
, et al
.
Therapeutic targeting of mesothelin with chimeric antigen receptor T cells in acute myeloid leukemia
.
Clin Cancer Res
.
2021
. ;
27
(
20
):
5718
-
5730
.
26.
Patel
SJ
,
Sanjana
NE
,
Kishton
RJ
, et al
.
Identification of essential genes for cancer immunotherapy
.
Nature
.
2017
. ;
548
(
7669
):
537
-
542
.
27.
Ayers
M
,
Lunceford
J
,
Nebozhyn
M
, et al
.
IFN-γ-related mRNA profile predicts clinical response to PD-1 blockade
.
J Clin Invest
.
2017
. ;
127
(
8
):
2930
-
2940
.
28.
Gao
J
,
Shi
LZ
,
Zhao
H
, et al
.
Loss of IFN-γ pathway genes in tumor cells as a mechanism of resistance to anti-CTLA-4 therapy
.
Cell
.
2016
. ;
167
(
2
):
397
-
404.e9
.
29.
Shankaran
V
,
Ikeda
H
,
Bruce
AT
, et al
.
IFN gamma and lymphocytes prevent primary tumour development and shape tumour immunogenicity
.
Nature
.
2001
. ;
410
(
6832
):
1107
-
1111
.
30.
Zhang
S
,
Kohli
K
,
Black
RG
, et al
.
Systemic interferon-γ increases MHC class I expression and T-cell infiltration in cold tumors: results of a phase 0 clinical trial
.
Cancer Immunol Res
.
2019
. ;
7
(
8
):
1237
-
1243
.
31.
Bourne
CM
,
Mun
SS
,
Dao
T
, et al
.
Unmasking the suppressed immunopeptidome of EZH2-mutated diffuse large B-cell lymphomas through combination drug treatment
.
Blood Adv
.
2022
. ;
6
(
14
):
4107
-
4121
.
32.
Oh
CY
,
Klatt
MG
,
Bourne
C
, et al
.
ALK and RET inhibitors promote HLA class I antigen presentation and unmask new antigens within the tumor immunopeptidome
.
Cancer Immunol Res
.
2019
. ;
7
(
12
):
1984
-
1997
.
33.
Brea
EJ
,
Oh
CY
,
Manchado
E
, et al
.
Kinase regulation of human MHC class I molecule expression on cancer cells
.
Cancer Immunol Res
.
2016
. ;
4
(
11
):
936
-
947
.
34.
Greiner
J
,
Ringhoffer
M
,
Taniguchi
M
, et al
.
mRNA expression of leukemia-associated antigens in patients with acute myeloid leukemia for the development of specific immunotherapies
.
Int J Cancer
.
2004
. ;
108
(
5
):
704
-
711
.
35.
Quintarelli
C
,
Dotti
G
,
De Angelis
B
, et al
.
Cytotoxic T lymphocytes directed to the preferentially expressed antigen of melanoma (PRAME) target chronic myeloid leukemia
.
Blood
.
2008
. ;
112
(
5
):
1876
-
1885
.
36.
Griffioen
M
,
Kessler
JH
,
Borghi
M
, et al
.
Detection and functional analysis of CD8+ T cells specific for PRAME: a target for T-cell therapy
.
Clin Cancer Res
.
2006
. ;
12
(
10
):
3130
-
3136
.
37.
Kessler
JH
,
Beekman
NJ
,
Bres-Vloemans
SA
, et al
.
Efficient identification of novel HLA-A(∗)0201-presented cytotoxic T lymphocyte epitopes in the widely expressed tumor antigen PRAME by proteasome-mediated digestion analysis
.
J Exp Med
.
2001
. ;
193
(
1
):
73
-
88
.
38.
Weber
JS
,
Vogelzang
NJ
,
Ernstoff
MS
, et al
.
A phase 1 study of a vaccine targeting preferentially expressed antigen in melanoma and prostate-specific membrane antigen in patients with advanced solid tumors
.
J Immunother
.
2011
. ;
34
(
7
):
556
-
567
.
39.
Alberts
DS
,
Marth
C
,
Alvarez
RD
, et al
.
Randomized phase 3 trial of interferon gamma-1b plus standard carboplatin/paclitaxel versus carboplatin/paclitaxel alone for first-line treatment of advanced ovarian and primary peritoneal carcinomas: results from a prospectively designed analysis of progression-free survival
.
Gynecol Oncol
.
2008
. ;
109
(
2
):
174
-
181
.
40.
Giannopoulos
A
,
Constantinides
C
,
Fokaeas
E
, et al
.
The immunomodulating effect of interferon-gamma intravesical instillations in preventing bladder cancer recurrence
.
Clin Cancer Res
.
2003
. ;
9
(
15
):
5550
-
5558
.
41.
Windbichler
GH
,
Hausmaninger
H
,
Stummvoll
W
, et al
.
Interferon-gamma in the first-line therapy of ovarian cancer: a randomized phase III trial
.
Br J Cancer
.
2000
. ;
82
(
6
):
1138
-
1144
.
42.
Srivastava
S
,
Riddell
SR
.
Engineering CAR-T cells: design concepts
.
Trends Immunol
.
2015
. ;
36
(
8
):
494
-
502
.
43.
Willemsen
RA
,
Debets
R
,
Hart
E
,
Hoogenboom
HR
,
Bolhuis
RL
,
Chames
P
.
A phage display selected fab fragment with MHC class I-restricted specificity for MAGE-A1 allows for retargeting of primary human T lymphocytes
.
Gene Ther
.
2001
. ;
8
(
21
):
1601
-
1608
.
44.
Wang
RF
,
Wang
HY
.
Immune targets and neoantigens for cancer immunotherapy and precision medicine
.
Cell Res
.
2017
. ;
27
(
1
):
11
-
37
.
45.
Robbins
PF
,
Morgan
RA
,
Feldman
SA
, et al
.
Tumor regression in patients with metastatic synovial cell sarcoma and melanoma using genetically engineered lymphocytes reactive with NY-ESO-1
.
J Clin Oncol
.
2011
. ;
29
(
7
):
917
-
924
.
46.
Rapoport
AP
,
Stadtmauer
EA
,
Binder-Scholl
GK
, et al
.
NY-ESO-1-specific TCR-engineered T cells mediate sustained antigen-specific antitumor effects in myeloma
.
Nat Med
.
2015
. ;
21
(
8
):
914
-
921
.
47.
Rafiq
S
,
Purdon
TJ
,
Daniyan
AF
, et al
.
Optimized T-cell receptor-mimic chimeric antigen receptor T cells directed toward the intracellular Wilms tumor 1 antigen
.
Leukemia
.
2017
. ;
31
(
8
):
1788
-
1797
.
48.
Akahori
Y
,
Wang
L
,
Yoneyama
M
, et al
.
Antitumor activity of CAR-T cells targeting the intracellular oncoprotein WT1 can be enhanced by vaccination
.
Blood
.
2018
. ;
132
(
11
):
1134
-
1145
.
49.
Zhang
G
,
Wang
L
,
Cui
H
, et al
.
Anti-melanoma activity of T cells redirected with a TCR-like chimeric antigen receptor
.
Sci Rep
.
2014
. ;
4
:
3571
. Published 2014 Jan 6.
50.
Liu
H
,
Xu
Y
,
Xiang
J
, et al
.
Targeting alpha-fetoprotein (AFP)-MHC complex with CAR T-cell therapy for liver cancer
.
Clin Cancer Res
.
2017
. ;
23
(
2
):
478
-
488
.
51.
Klatt
MG
,
Zhiyuan
Y
,
Jianying
L
, et al
.
A TCR mimic antibody-directed CAR T cell specific for intracellular NDC80 is broadly cancer reactive and displays high activity against hematological malignancies
.
Blood
.
2020
. ;
136
(
1
):
20
-
21
.
52.
Akatsuka
Y
.
TCR-like CAR-T cells targeting MHC-bound minor histocompatibility antigens
.
Front Immunol
.
2020
. ;
11
:
257
. Published 2020 Feb 28.
53.
Zhu
L
,
Yang
X
,
Zhong
D
, et al
.
Single-domain antibody-based TCR-like CAR-T: a potential cancer therapy
.
J Immunol Res
.
2020
. ;
2020
:
2454907
. Published 2020 Sep 7.

Author notes

D.C.K. and A.M.L. contributed equally to this study.

Q.L. and S.M. contributed equally to this study.

For original data, please contact the corresponding author, Danielle C. Kirkey (dkirkey@fredhutch.org).

The full-text version of this article contains a data supplement.

Supplemental data