Ex vivo expanded tumour-infiltrating lymphocytes from ovarian cancer patients release anti-tumour cytokines in response to autologous primary ovarian cancer cells.

Epithelial ovarian cancer (EOC) is the leading cause of gynaecological cancer-related death in Europe. Although most patients achieve an initial complete response with first-line treatment, recurrence occurs in more than 80% of cases. Thus, there is a clear unmet need for novel second-line treatments. EOC is frequently infiltrated with T lymphocytes, the presence of which has been shown to be associated with improved clinical outcomes. Adoptive T-cell therapy (ACT) using ex vivo-expanded tumour-infiltrating lymphocytes (TILs) has shown remarkable efficacy in other immunogenic tumours, and may represent a promising therapeutic strategy for EOC. In this preclinical study, we investigated the efficacy of using anti-CD3/anti-CD28 magnetic beads and IL-2 to expand TILs from freshly resected ovarian tumours. TILs were expanded for up to 3 weeks, and then subjected to a rapid-expansion protocol (REP) using irradiated feeder cells. Tumours were collected from 45 patients with EOC and TILs were successfully expanded from 89.7% of biopsies. Expanded CD4+ and CD8+ subsets demonstrated features associated with memory phenotypes, and had significantly higher expression of key activation and functional markers than unexpanded TILs. Expanded TILs produced anti-tumour cytokines when co-cultured with autologous tumour cells, inferring tumour cytotoxicity. Our findings demonstrate that it is possible to re-activate and expand tumour-reactive T cells from ovarian tumours. This presents a promising immunotherapy that could be used sequentially or in combination with current therapeutic strategies.

CAR T cells: driving the road from the laboratory to the clinic.

Blockbuster antibody therapies have catapulted immune-based approaches to treat cancer into the consciousness of mainstay clinical research. On the back of this, other emerging immune-based therapies are providing great promise. T-cell therapy is one such area where recent trials using T cells genetically modified to express an antibody-based chimeric antigen receptor (CAR) targeted against the CD19 antigen have demonstrated impressive responses when adoptively transferred to patients with advanced chronic lymphocytic leukemia. The general concept of the CAR T cell was devised some 20 years ago. In this relatively short period of time, the technology to redirect T-cell function has moved at pace facilitating clinical translation; however, many questions remain with respect to developing the approach to improve CAR T-cell therapeutic activity and also to broaden the range of tumors that can be effectively targeted by this approach. This review highlights some of the underlying principles and compromises of CAR T-cell technology using the CD19-targeted CAR as a paradigm and discusses some of the issues that relate to targeting solid tumors with CAR T cells.

The clinical efficacy of first-generation carcinoembryonic antigen (CEACAM5)-specific CAR T cells is limited by poor persistence and transient pre-conditioning-dependent respiratory toxicity.

The primary aim of this clinical trial was to determine the feasibility of delivering first-generation CAR T cell therapy to patients with advanced, CEACAM5+ malignancy. Secondary aims were to assess clinical efficacy, immune effector function and optimal dose of CAR T cells. Three cohorts of patients received increasing doses of CEACAM5+-specific CAR T cells after fludarabine pre-conditioning plus systemic IL2 support post T cell infusion. Patients in cohort 4 received increased intensity pre-conditioning (cyclophosphamide and fludarabine), systemic IL2 support and CAR T cells. No objective clinical responses were observed. CAR T cell engraftment in patients within cohort 4 was significantly higher. However, engraftment was short-lived with a rapid decline of systemic CAR T cells within 14 days. Patients in cohort 4 had transient, acute respiratory toxicity which, in combination with lack of prolonged CAR T cell persistence, resulted in the premature closure of the trial. Elevated levels of systemic IFNγ and IL-6 implied that the CEACAM5-specific T cells had undergone immune activation in vivo but only in patients receiving high-intensity pre-conditioning. Expression of CEACAM5 on lung epithelium may have resulted in this transient toxicity. Raised levels of serum cytokines including IL-6 in these patients implicate cytokine release as one of several potential factors exacerbating the observed respiratory toxicity. Whilst improved CAR designs and T cell production methods could improve the systemic persistence and activity, methods to control CAR T 'on-target, off-tissue' toxicity are required to enable a clinical impact of this approach in solid malignancies.

Exploiting natural killer group 2D receptors for CAR T-cell therapy.

Chimeric antigen receptors (CARs) are genetically engineered proteins that combine an extracellular antigen-specific recognition domain with one or several intracellular T-cell signaling domains. When expressed in T cells, these CARs specifically trigger T-cell activation upon antigen recognition. While the clinical proof of principle of CAR T-cell therapy has been established in hematological cancers, CAR T cells are only at the early stages of being explored to tackle solid cancers. This special report discusses the concept of exploiting natural killer cell receptors as an approach that could broaden the specificity of CAR T cells and potentially enhance the efficacy of this therapy against solid tumors. New data demonstrating feasibility of this approach in humans and supporting the ongoing clinical trial are also presented.

Differential role of Th1 and Th2 cytokines in autotoxicity driven by CD19-specific second-generation chimeric antigen receptor T cells in a mouse model.

T cells engrafted with chimeric AgRs (CAR) are showing exciting potential for targeting B cell malignancies in early-phase clinical trials. To determine whether the second-generation CAR was essential for optimal antitumor activity, two CD28-based CAR constructs targeting CD19 were tested for their ability to redirect mouse T cell function against established B cell lymphoma in a BALB/c syngeneic model system. T cells armed with either CAR eliminated A20 B cell lymphoma in vivo; however, one construct induced a T cell dose-dependent acute toxicity associated with a raised serum Th1 type cytokine profile on transfer into preconditioned mice. Moreover, a chronic toxicity manifested as granuloma-like formation in spleen, liver, and lymph nodes was observed in animals receiving T cells bearing either CD28 CAR, albeit with different kinetics dependent upon the specific receptor used. This phenotype was associated with an expansion of CD4+ CAR+ T cells and CD11b+ Gr-1(+) myeloid cells and increased serum Th2-type cytokines, including IL-10 and IL-13. Mouse T cells engrafted with a first-generation CAR failed to develop such autotoxicity, whereas toxicity was not apparent when T cells bearing the same receptors were transferred into C57BL/6 or C3H animals. In summary, the adoptive transfer of second-generation CD19-specific CAR T cells can result in a cell dose-dependent acute toxicity, whereas the prolonged secretion of high levels of Th2 cytokines from these CAR T cells in vivo drives a granulomatous reaction resulting in chronic toxicity. Strategies that prevent a prolonged Th2-cytokine biased CAR T cell response are clearly warranted.

Definition and application of good manufacturing process-compliant production of CEA-specific chimeric antigen receptor expressing T-cells for phase I/II clinical trial.

Adoptive cell therapy employing gene-modified T-cells expressing chimeric antigen receptors (CARs) has shown promising preclinical activity in a range of model systems and is now being tested in the clinical setting. The manufacture of CAR T-cells requires compliance with national and European regulations for the production of medicinal products. We established such a compliant process to produce T-cells armed with a first-generation CAR specific for carcinoembryonic antigen (CEA). CAR T-cells were successfully generated for 14 patients with advanced CEA(+) malignancy. Of note, in the majority of patients, the defined procedure generated predominantly CD4(+) CAR T-cells with the general T-cell population bearing an effector-memory phenotype and high in vitro effector function. Thus, improving the process to generate less-differentiated T-cells would be more desirable in the future for effective adoptive gene-modified T-cell therapy. However, these results confirm that CAR T-cells can be generated in a manner compliant with regulations governing medicinal products in the European Union.

CYAD-01, an autologous NKG2D-based CAR T-cell therapy, in relapsed or refractory acute myeloid leukaemia and myelodysplastic syndromes or multiple myeloma (THINK): haematological cohorts of the dose escalation segment of a phase 1 trial.

BACKGROUND: CYAD-01 is an autologous chimeric antigen receptor (CAR) T-cell product based on the natural killer (NK) group 2D (NKG2D) receptor, which binds eight ligands that are overexpressed in a wide range of haematological malignancies but are largely absent on non-neoplastic cells. Initial clinical evaluation of a single infusion of CYAD-01 at a low dose in patients with relapsed or refractory acute myeloid leukaemia, myelodysplastic syndromes, and multiple myeloma supported the feasibility of the approach and prompted further evaluation of CYAD-01. The aim of the present study was to determine the safety and recommended phase 2 dosing of CYAD-01 administered without preconditioning or bridging chemotherapy. METHODS: The multicentre THINK study was an open-label, dose-escalation, phase 1 study for patients with relapsed or refractory acute myeloid leukaemia, myelodysplastic syndromes, or multiple myeloma, after at least one previous line of therapy. Patients were recruited from five hospitals in the USA and Belgium. The dose-escalation segment evaluated three dose levels: 3 × 108 (dose level one), 1 × 109 (dose level two), and 3 × 109 (dose level three) cells per infusion with a 3 + 3 Fibonacci study design using a schedule of three infusions at 2-week intervals followed by potential consolidation treatment consisting of three additional infusions. The occurrence of dose-limiting toxicities post-CYAD-01 infusion was assessed as the primary endpoint in the total treated patient population. The trial was registered with ClinicalTrials.gov, NCT03018405, and EudraCT, 2016-003312-12, and has been completed. FINDINGS: Between Feb 6, 2017, and Oct 9, 2018, 25 patients were registered in the haematological dose-escalation segment. Seven patients had manufacturing failure for insufficient yield and two had screening failure. 16 patients were treated with CYAD-01 (three with multiple myeloma and three with acute myeloid leukaemia at dose level one; three with acute myeloid leukaemia at dose level two; and six with acute myeloid leukaemia and one with myelodysplastic syndromes at dose level three). Median follow-up was 118 days (IQR 46-180). Seven patients (44%) had grade 3 or 4 treatment-related adverse events. In total, five patients (31%) had grade 3 or 4 cytokine release syndrome across all dose levels. One dose-limiting toxicity of cytokine release syndrome was reported at dose level three. No treatment-related deaths occurred, and the maximum tolerated dose was not reached. Three (25%) of 12 evaluable patients with relapsed or refractory acute myeloid leukaemia or myelodysplastic syndromes had an objective response. Among responders, two patients with acute myeloid leukaemia proceeded to allogeneic haematopoietic stem-cell transplantation (HSCT) after CYAD-01 treatment, with durable ongoing remissions (5 and 61 months). INTERPRETATION: Treatment with a multiple CYAD-01 infusion schedule without preconditioning is well tolerated and shows anti-leukaemic activity, although without durability outside of patients bridged to allogeneic HSCT. These phase 1 data support the proof-of-concept of targeting NKG2D ligands by CAR T-cell therapy. Further clinical studies with NKG2D-based CAR T-cells are warranted, potentially via combinatorial antigen targeted approaches, to improve anti-tumour activity. FUNDING: Celyad Oncology.

Targeting breast cancer vaccines to dendritic cells: improved immunological responses with less protein?

The central goal of cancer immunotherapy is to control tumors through the mobilization of the patient's immune system. Vaccines targeting the Her2/neu proto-oncogene have been tested with some early encouraging responses in breast cancer. However, a more effective set of vaccines targeting specific immune cell subtypes may provide a more potent means to stimulate anti-tumor immunity. Dendritic cell-specific antibodies fused with the Her2/neu protein proved effective at generating immune responses in preclinical models. Importantly, only low amounts of protein vaccine were required to generate this response, which has potentially significant implications for the future clinical development of Her2/neu-targeted vaccines and other vaccine targets.

Targeted immunotherapy of cancer with CAR T cells: achievements and challenges.

The adoptive transfer of chimeric antigen receptor (CAR)-expressing T cells is a relatively new but promising approach in the field of cancer immunotherapy. This therapeutic strategy is based on the genetic reprogramming of T cells with an artificial immune receptor that redirects them against targets on malignant cells and enables their destruction by exerting T cell effector functions. There has been an explosion of interest in the use of CAR T cells as an immunotherapy for cancer. In the pre-clinical setting, there has been a considerable focus upon optimizing the structural and signaling potency of the CAR while advances in bio-processing technology now mean that the clinical testing of these gene-modified T cells has become a reality. This review will summarize the concept of CAR-based immunotherapy and recent clinical trial activity and will further discuss some of the likely future challenges facing CAR-modified T cell therapies.

Regulation of autologous immunity to the mouse 5T4 oncofoetal antigen: implications for immunotherapy.

Effective vaccination against tumour-associated antigens (TAA) such as the 5T4 oncofoetal glycoprotein may be limited by the nature of the T cell repertoire and the influence of immunomodulatory factors in particular T regulatory cells (Treg). Here, we identified mouse 5T4-specific T cell epitopes using a 5T4 knock out (5T4KO) mouse and evaluated corresponding wild-type (WT) responses as a model to refine and improve immunogenicity. We have shown that 5T4KO mice vaccinated by replication defective adenovirus encoding mouse 5T4 (Adm5T4) generate potent 5T4-specific IFN-γ CD8 and CD4 T cell responses which mediate significant protection against 5T4 positive tumour challenge. 5T4KO CD8 but not CD4 primed T cells also produced IL-17. By contrast, Adm5T4-immunized WT mice showed no tumour protection consistent with only low avidity CD8 IFN-γ, no IL-17 T cell responses and no detectable CD4 T cell effectors producing IFN-γ or IL-17. Treatment with anti-folate receptor 4 (FR4) antibody significantly reduced the frequency of Tregs in WT mice and enhanced 5T4-specific IFN-γ but reduced IL-10 T cell responses but did not reveal IL-17-producing effectors. This altered balance of effectors by treatment with FR4 antibody after Adm5T4 vaccination provided modest protection against autologous B16m5T4 melanoma challenge. The efficacy of 5T4 and some other TAA vaccines may be limited by the combination of TAA-specific T regs, the deletion and/or alternative differentiation of CD4 T cells as well as the absence of distinct subsets of CD8 T cells.

The optimal antigen response of chimeric antigen receptors harboring the CD3zeta transmembrane domain is dependent upon incorporation of the receptor into the endogenous TCR/CD3 complex.

Chimeric Ag receptors (CARs) expressed in T cells permit the redirected lysis of tumor cells in an MHC-unrestricted manner. In the Jurkat T cell model system, expression of a carcinoembryonic Ag-specific CD3zeta CAR (MFEzeta) resulted in an increased sensitivity of the transduced Jurkat cell to generate cytokines when stimulated through the endogenous TCR complex. This effect was driven through two key characteristics of the MFEzeta CAR: 1) receptor dimerization and 2) the interaction of the CAR with the endogenous TCR complex. Mutations of the CAR transmembrane domain that abrogated these interactions resulted in a reduced functional capacity of the MFEzeta CAR to respond to carcinoembryonic Ag protein Ag. Taken together, these results indicate that CARs containing the CD3zeta transmembrane domain can form a complex with the endogenous TCR that may be beneficial for optimal T cell activation. This observation has potential implications for the future design of CARs for cancer therapy.

Natural expression of the CD19 antigen impacts the long-term engraftment but not antitumor activity of CD19-specific engineered T cells.

T cells gene-modified to express chimeric Ag receptors (CARs) have shown potent antitumor activity in vivo and are in clinical trials at locations worldwide. However, CAR activity has been investigated in mouse models in which Ag expression is restricted to the tumor. To explore the impact of normal tissue expression of the target Ag, we developed a mouse CD19-specific CAR to investigate antitumor efficacy against a syngeneic B cell lymphoma cell line within a background of normal CD19(+) host B cells. Mouse T cells engrafted with the amCD19CD3zeta CAR specifically lysed A20 lymphoma targets and B cells in vitro. These T cells also eradicated a 12-d established disseminated A20 lymphoma in mice preconditioned with 6 Gy total body irradiation. In the short-term (7 d after adoptive transfer), amCD19z T cells underwent Ag-dependent proliferation in vivo with a concomitant depletion in host B cell levels. However, the levels of amCD19z CAR(+) T cells decreased significantly at later time points, at which point host B cells returned, eventually reaching normal levels. In contrast, CAR(+) T cells lacking a signaling domain or specificity for mCD19 persisted over extended periods in blood and spleen. Importantly, no overt clinical signs of autotoxicity were observed in tumor-free or tumor-bearing mice treated with amCD19z T cells over an extended period of time. These observations highlight the importance of studying the activity of CAR(+) T cells in autologous models that have the normal range of tissue expression of Ag.

Clinical Grade Manufacture of CYAD-101, a NKG2D-based, First in Class, Non-Gene-edited Allogeneic CAR T-Cell Therapy.

Allogeneic chimeric antigen receptor (CAR) T holds the promise of taking this therapeutic approach to broader patient populations while avoiding the intensive manufacturing demands of autologous cell products. One limitation to delivering an allogeneic CAR T is T-cell receptor (TCR) driven toxicity. In this work, the expression of a peptide to interfere with TCR signaling was assessed for the generation of allogeneic CAR T cells. The expression of a truncated CD3ζ peptide was shown to incorporate into the TCR complex and to result in blunted TCR responses. When coexpressed with a natural killer group 2D (NKG2D) CAR, the allogeneic T cells (called CYAD-101) failed to induce graft-versus-host disease in mouse models while maintaining antitumor activity driven by the CAR in vitro and in vivo. Two clinical grade discrete batches of CYAD-101 cells were produced of single donor apheresis resulting in 48 billion CAR T cells sufficient for the entire dose-escalation phase of the proposed clinical trial. The 2 batches showed high consistency producing a predominantly CD4+ T-cell population that displayed an effector/central memory phenotype with no evidence of exhaustion markers expression. These clinical grade CYAD-101 cells secreted cytokines and chemokines in response to ligands expressing target cells in vitro, demonstrating effector function through the CAR. Moreover, CYAD-101 cells failed to respond to TCR stimulation, indicating a lack of allogeneic potential. This bank of clinical grade, non-gene-edited, allogeneic CYAD-101 cells are used in the alloSHRINK clinical trial (NCT03692429).

Gene delivery of a mutant TGFß3 reduces markers of scar tissue formation after cutaneous wounding.

The transforming growth factor-ß (TGFß) family plays a critical regulatory role in repair and coordination of remodeling after cutaneous wounding. TGFß1-mediated chemotaxis promotes the recruitment of fibroblasts to the wound site and their resultant myofibroblastic transdifferentiation that is responsible for elastic fiber deposition and wound closure. TGFß3 has been implicated in an antagonistic role regulating overt wound closure and promoting ordered dermal remodeling. We generated a mutant form of TGFß3 (mutTGFß3) by ablating its binding site for the latency-associated TGFß binding protein (LTBP-1) in order to improve bioavailability and activity. The mutated cytokine is secreted as the stable latency-associated peptide (LAP)-associated form and is activated by normal intracellular and extracellular mechanisms including integrin-mediated activation but is not sequestered. We show localized intradermal transduction using a lentiviral vector expressing the mutTGFß3 in a mouse skin wounding model reduced re-epithelialization density and fibroblast/myofibroblast transdifferentiation within the wound area, both indicative of reduced scar tissue formation.

Cytokine stimulation and the choice of promoter are critical factors for the efficient transduction of mouse T cells with HIV-1 vectors.

BACKGROUND: HIV-1 fails to successfully infect mouse T cells as a result of several blocks in the viral replication cycle. We investigated whether this also impacted on the use of HIV-1 derived lentiviral vectors for stable gene transfer into mouse T cells. METHODS: Freshly isolated primary mouse T cells were immediately mixed with lentiviral vectors encoding an enhanced green fluorescent protein marker gene and transduction frequency was determined after 5 days of culture. RESULTS: Optimal transduction required both mouse T cell activation and cytokine support. Furthermore, transduction was also dependent upon the promoter chosen, with the rank order of potency being PGK > EF1 > SFFV > CMV. HIV-1 lentiviral vectors also efficiently transduced cytokine-stimulated T cells (in the absence of antibody driven T cell activation), albeit with a lower level of transgene expression compared to fully-activated T cells. CONCLUSIONS: The present study demonstrates that primary mouse T cells can be efficiently transduced with HIV-1 lentiviral vectors, opening up prospects for their use in mouse models of gene-modified adoptive cellular therapy.

Development of a flow cytometric co-immunoprecipitation technique for the study of multiple protein-protein interactions and its application to T-cell receptor analysis.

Co-immunoprecipitation is the classical approach for investigating protein-protein interactions. Analysis is generally conducted using the Western blot approach. We set out to investigate whether flow cytometry was a feasible alternative to Western blotting. Using the TCR-CD3 complex as a model for intermolecular interactions in the MA5.8 cell line, FLAG-tagged CD3zeta-scFv fusion proteins could be captured on anti-FLAG coupled beads and associated TCRbeta molecules could be detected by flow cytometry. This association was abrogated by mutations to the CD3zeta transmembrane domain. Using multicolor flow cytometry, TCRbeta, CD3epsilon, and the scFv region of the CD3zeta fusion molecule could all be detected from a single sample. This multicolor analysis was then applied to demonstrate the importance of correct lysis conditions for extraction of the TCR complex. In summary, this flow cytometric immunoprecipitation technique is a feasible alternative to classical co-immunoprecipitation analysis technique and offers many potential advantages including rapid analysis with increased target sensitivity, reduced technical demands, amenable to multiple protein analysis from a single sample, and provides a framework that may facilitate the development of high throughput analytical assays investigating protein-protein interactions.

The integrated stress response promotes B7H6 expression.

The B7 family member, B7H6, is a ligand for the natural killer cell receptor NKp30. B7H6 is hardly expressed on normal tissues, but undergoes upregulation on different types of tumors, implicating it as an attractive target for cancer immunotherapy. The molecular mechanisms that control B7H6 expression are poorly understood. We report that in contrast to other NK cell ligands, endoplasmic reticulum (ER) stress upregulates B7H6 mRNA levels and surface expression. B7H6 induction by ER stress requires protein kinase R-like ER kinase (PERK), one of the three canonical sensors of the unfolded protein response. PERK phosphorylates eIF2α, which regulates protein synthesis and gene expression. Because eIF2α is phosphorylated by several kinases following different stress conditions, the program downstream to eIF2α phosphorylation is called the integrated stress response (ISR). Several drugs were reported to promote the ISR. Nelfinavir and lopinavir, two clinically approved HIV protease inhibitors, promote eIF2α phosphorylation by different mechanisms. We show that nelfinavir and lopinavir sustainably instigate B7H6 expression at their pharmacologically relevant concentrations. As such, ER stress and ISR conditions sensitize melanoma targets to CAR-T cells directed against B7H6. Our study highlights a novel mechanism to induce B7H6 expression and suggests a pharmacological approach to improve B7H6-directed immunotherapy. KEY MESSAGES: B7H6 is induced by ER stress in a PERK-dependent mechanism. Induction of B7H6 is obtained pharmacologically by HIV protease inhibitors. Exposure of tumor cells to the HIV protease inhibitor nelfinavir improves the recognition by B7H6-directed CAR-T.

Chimeric Antigen Receptor-T Cells for Targeting Solid Tumors: Current Challenges and Existing Strategies.

Chimeric antigen receptor-T cells (CAR-Ts) are an exciting new cancer treatment modality exemplified by the recent regulatory approval of two CD19-targeted CAR-T therapies for certain B cell malignancies. However, this success in the hematological setting has yet to translate to a significant level of objective clinical responses in the solid tumor setting. The reason for this lack of translation undoubtedly lies in the substantial challenges raised by solid tumors to all therapies, including CAR-T, that differ from B cell malignancies. For instance, intravenously infused CAR-Ts are likely to make rapid contact with cancerous B cells since both tend to reside in the same vascular compartments within the body. By contrast, solid cancers tend to form discrete tumor masses with an immune-suppressive tumor microenvironment composed of tumor cells and non-tumor stromal cells served by abnormal vasculature that restricts lymphocyte infiltration and suppresses immune function, expansion, and persistence. Moreover, the paucity of uniquely and homogeneously expressed tumor antigens and inherent plasticity of cancer cells provide major challenges to the specificity, potency, and overall effectiveness of CAR-T therapies. This review focuses on the major preclinical and clinical strategies currently being pursued to tackle these challenges in order to drive the success of CAR-T therapy against solid tumors.

Phase I Trial of Autologous CAR T Cells Targeting NKG2D Ligands in Patients with AML/MDS and Multiple Myeloma.

NKG2D ligands are widely expressed in solid and hematologic malignancies but absent or poorly expressed on healthy tissues. We conducted a phase I dose-escalation study to evaluate the safety and feasibility of a single infusion of NKG2D-chimeric antigen receptor (CAR) T cells, without lymphodepleting conditioning in subjects with acute myeloid leukemia/myelodysplastic syndrome or relapsed/refractory multiple myeloma. Autologous T cells were transfected with a γ-retroviral vector encoding a CAR fusing human NKG2D with the CD3ζ signaling domain. Four dose levels (1 × 106-3 × 107 total viable T cells) were evaluated. Twelve subjects were infused [7 acute myeloid leukemia (AML) and 5 multiple myeloma]. NKG2D-CAR products demonstrated a median 75% vector-driven NKG2D expression on CD3+ T cells. No dose-limiting toxicities, cytokine release syndrome, or CAR T cell-related neurotoxicity was observed. No significant autoimmune reactions were noted, and none of the ≥ grade 3 adverse events were attributable to NKG2D-CAR T cells. At the single injection of low cell doses used in this trial, no objective tumor responses were observed. However, hematologic parameters transiently improved in one subject with AML at the highest dose, and cases of disease stability without further therapy or on subsequent treatments were noted. At 24 hours, the cytokine RANTES increased a median of 1.9-fold among all subjects and 5.8-fold among six AML patients. Consistent with preclinical studies, NKG2D-CAR T cell-expansion and persistence were limited. Manufactured NKG2D-CAR T cells exhibited functional activity against autologous tumor cells in vitro, but modifications to enhance CAR T-cell expansion and target density may be needed to boost clinical activity.