Chimeric antigen receptor T cells have hit the prime time, with 2 FDA approvals already for this class of cellbased therapy. Undoubtedly this will be a game changer for patients with B-cell malignancies who have a small number of treatment options; however, questions regarding the realworld application of CAR T-cell therapies remain.
Martin Pule, PhD
Martin Pule, PhD
Chimeric antigen receptor (CAR) T cells have hit the prime time, with 2 FDA approvals already for this class of cellbased therapy. Undoubtedly this will be a game changer for patients with B-cell malignancies who have a small number of treatment options; however, questions regarding the realworld application of CAR T-cell therapies remain. Several challenges may affect the adoption of CAR T-cell therapy, which range from the logistical complexities of therapeutic development to the potentially fatal toxicities of treatment, as Stephan A. Grupp, MD, PhD, indicated yesterday in his keynote address. A number of innovations are already entering the scene, including next-generation “armored” CARs and the application of state-of-the-art genome-editing technologies. These innovations and more will be discussed during tomorrow’s plenary session entitled “Genetically Modified Cell Therapy,” starting with an explanation of gene-edited CAR T cells by Martin Pule, PhD. It will be several years before the true impact of CAR T cells becomes apparent, but it is evident that long-term success requires unlocking the potential of CAR T-cell therapy beyond B-cell malignancies, particularly in solid tumors, which account for the vast majority of cancer deaths.
Harnessing Antitumor Immune Cells
CAR T-cell therapy falls under the banner of adoptive cellular therapy (ACT), in which the effector cells of the immune system, predominantly the T cells, are transplanted into a patient. The graft-versus-tumor effect observed in patients undergoing allogeneic hematopoietic stem cell transplantation served as the stimulus for pursuing ACT. T cells present in the transplanted material were seemingly capable of mounting an effective antitumor immune response, but this was limited by the development of graft-versus host-disease (GvHD), wherein the transplanted T cells also attacked healthy tissue, causing severe toxicity.
Genetically modifying host T cells to endow them with tumor-killing specificity originated in the 1980s. Seminal studies demonstrated that replacing the region of the T-cell receptor (TCR) responsible for antigen recognition with the single- chain variable fragment (scFv) of an antibody could couple antibody specificity for a tumorassociated antigen (TAA) with the T-cell activating machinery of the TCR.1This permitted direct activation of T cells by tumor cells that express the target antigen, in a manner that bypasses the need for antigen presentation by the major histocompatibility complex.2
CAR T cells are generated by collecting host T cells through leukapheresis, where the CAR is introduced to the T cells through a viral vector. Clinical trials of the resulting CAR T cells, which can then be infused back into the patient, were pioneered by investigators at Memorial Sloan Kettering Cancer Center and the University of Pennsylvania.
The prototypical CAR was composed of an extracellular portion, consisting of the antibody scFv, joined to a hinge or spacer and then a transmembrane domain that spans into the cell and links to an intracellular signaling region.3The latter mediates T-cell activation and was most often derived from the CD3ζ chain of the TCR, which harbors 3 immunoreceptor tyrosine- based activation motifs (ITAMs). When the CAR engages its target antigen, the ITAMs become phosphorylated and serve as adaptors for a panel of signaling proteins that coordinate T-cell activation and proliferation.
Subsequent generations have been developed to improve on the efficacy of the first generation. The realization that full activation of T cells required 2 signals led to the incorporation of costimulatory domains in second- and third-generation CARs, with enhanced T-cell activity and persistence and greater antitumor efficacy.3-5
The CAR T-Cell Race to the Market
In this rapidly evolving field of immuno-oncolytic therapeutic development, the most clinically advanced CAR T cells are currently those engineered against CD19, a cell surface receptor constitutively expressed in hematologic malignancies. Several pharmaceutical companies raced to be the first to secure regulatory approval, but ultimately the firstto- market advantage went to Novartis’ CTL019 (tisagenlecleucel; Kymriah).
In August, tisagenlecleucel was approved for the treatment of pediatric and young-adult patients (up to 25 years) with relapsed/refractory B-cell precursor acute lymphoblastic leukemia (ALL). The efficacy and safety of tisagenlecleucel was demonstrated in the phase II ELIANA trial that led to its approval for this indication. A total of 68 pediatric and young-adult patients with CD19-positive B-cell precursor ALL were infused with a single dose of CAR T cells. After a follow-up of 3 months, 63 patients treated with tisagenlecleucel were evaluable for response; 83% achieved complete remission (CR) or incomplete CR within 3 months of infusion with minimal residual disease (MRD)negative status achieved in the bone marrow.6
Hot on Novartis’ heels was Kite Pharma, whose CD19-targeted CAR T-cell therapy, axicabtagene ciloleucel (KTE-C19; axi-cel; Yescarta), received approval as treatment for adults with relapsed or refractory non-Hodgkin lymphoma based on results of the ZUMA-1 trial in October. Axi-cel was administered as a single infusion of modified autologous T cells at a target dose of 2 x 106 CAR-positive T cells/kg.
As of January 2017, 101 patients enrolled in the ZUMA-1 trial were treated with axi-cel. The study completed its primary endpoint of combined objective response rate (ORR) for patients with diffuse large B-cell lymphoma (DLBCL), primary mediastinal B-cell lymphoma (PMBCL), or transformed follicular lymphoma (FL) after 6 months’ follow-up. The majority of the population enrolled in the ZUMA-1 trial were patients with DLBCL (n = 77); the ORR and CR rates in this population were 82% and 49%, respectively.7 Among the patients with PMBCL/ FL (n = 24), the ORR was 83% and the CR rate was 71%.
Both axicabtagene ciloleucel and tisagenlecleucel are also being evaluated in clinical trials for additional indications (TABLE 1). Results from the ongoing phase I/II ZUMA-3 trial for axicabtagene ciloleucel in adults with relapsed/refractory B-cell ALL were reported at the 2017 ASCO Annual Meeting. Among 8 patients evaluable for efficacy, 6 achieved an MRD-negative CR or a CR with partial or incomplete hematopoietic recovery.8
The ongoing phase II JULIET trial is investigating the safety and efficacy of tisagenlecleucel treatment in adult patients with relapsed/refractory DLBCL. Preliminary results from JULIET were presented at the 2017 International Conference on Malignant Lymphoma. Of 85 patients who received a single dose of tisagenlecleucel, 51 were evaluable for tisagenlecleucel response with at least 3 months’ follow-up. For these patients, the ORR was 59% (95% CI, 44.2%-72.4%); 43% of patients had a CR, and 16% had a partial remission.9
Juno Therapeutics’ CD19-targeted CAR T-cell therapy, JCAR015, was originally slated to secure regulatory approval first. However, the development of JCAR015 was discontinued following 2 clinical holds placed on the phase II ROCKET trial, in which an adverse safety profile revealed an increased incidence of deaths due to cerebral edema, as was discussed yesterday in a presentation by Mark J. Gilbert, MD, of neurotoxicities seen in the ROCKET trial.
Despite the disappointing safety concerns with JCAR015, Juno Therapeutics is continuing clinical development of a substantial CAR T-cell therapy pipeline. The ongoing phase I TRANSCEND study is investigating the safety and efficacy of another pipeline CD19-targeted CAR T-cell therapy, JCAR017, and has enrolled patients with relapsed/refractory B-cell NHL, including DLBCL, PMBCL, FL, and mantle cell lymphoma. Interim data were presented at the 2016 ASH Annual Meeting of evaluable patients following treatment with fludarabine, cyclophosphamide, and JCAR017. The majority of the population (n = 25) had DLBCL. Of 20 patients treated with lower-dose JCAR017, the ORR was 80% and 60% achieved a CR.10
Juno’s experience highlights a complicating factor of clinical implementation of CAR T-cell therapiesthe potential for severe, sometimes fatal, treatment-related toxicities. Most prominent are cytokine release syndrome (CRS)—a severe immune reaction resulting from the release of cytokines from the targeted cancer cells and by immune effector cells recruited to the area—and neurotoxicity, resulting in brain swelling, altered consciousness, and seizures.
In the ELIANA trial, CRS occurred in 78% of patients, with 27% of grade 4 severity; however, there were no CRS-related deaths. Grade 3 neuropsychiatric events occurred in 15% of patients, but there was no incidence of cerebral edema. In the ZUMA-1 trial, adverse events of CRS and neurotoxicity of grade ≥3 severity occurred in 13% and 28% of patients, respectively; all cases resolved except for a grade 1 memory impairment. The first death from cerebral edema in this trial was recently reported, and although studies will continue a, patient consent forms will now be amended to reflect the risk.6,7
All CAR T-cell therapies developed to date cause these and other toxicities in at least some patients, but there is substantial variation in the severity and type of toxicity. A prospective registry and monitoring study were both proposed to assess the safety of tisagenlecleucel for up to 15 years after initiation to gain a better understanding of the toxicity profile and longterm effects of the approved treatment.11
Understanding the reasons for these toxicities is a central focus of ongoing research, in addition to the question of longterm safety relating to the risks of using viral vectors and the persistence of CAR T cells in the body after treatment.2,3
Alternative Targets Broaden Scope
The success of CD19-targeted CAR T cells has prompted exploration of other TAAs. Of particular note is the B-cell maturation antigen (BCMA), a member of the tumor necrosis factor receptor superfamily. Until recently, CAR T-cell therapies had generated mixed results in multiple myeloma, an incurable B-cell malignancy, but BCMA is expressed nearly universally on multiple myeloma cells, making it a promising target for CAR T-cell therapy in this tumor type.
Bluebird Bio and the Chinese company Nanjing Legend Biotech are pursuing the development of rival CAR T-cell therapies for the treatment of relapsed/refractory multiple myeloma. Both therapies created a buzz at this year’s ASCO Annual Meeting.
Nanjing Legend Biotech’s LCAR-B38M is uniquely engineered, in which the antigen-binding region of the CAR is a non-scFv structure targeting 2 regions of the BCMA antigen. Researchers hope this design will allow for more effective antitumor activity. To address safety concerns, LCAR-B38M is also being administered as 3 smaller infusions rather than a single large infusion.12 Nineteen patients have been treated with LCAR-B38M to date, 7 of whom were monitored for more than 6 months. The ORR was 100%; 95% of patients (n = 18) reached near CR without relapse, including 7 patients who achieved CR and were MRD-negative at a median follow-up of 6 months. CRS was observed in 74% of patients but was mostly mild or manageable. Two patients experienced grade 3 or 4 CRS; the patient with grade 4 CRS recovered fully after treatment.12
In Bluebird Bio’s phase I trial, bb2121 was administered to 11 patients with relapsed/refractory multiple myeloma in 4 escalating dose cohorts (doses of 5, 15, 45, and 80 x 107 CAR T cells). The ORR in the 6 evaluable patients treated with doses of ≥15 x 107 CAR T cells was 100%, including stringent CR in 2 patients and MRD-negative responses in 2 patients. There were no reported adverse events of CRS or neurotoxicity of ≥grade 3, and there were no dose-limiting toxicities.13
Looking Beyond B-Cell Malignancies
The unprecedented clinical efficacy of CAR T-cell therapy in patients with hematologic malignancies has not translated into solid tumors to date. However, the safety and efficacy of CAR T-cell therapy for solid tumors, the major cause of cancer-related mortality, have been the focus of a growing number of clinical trials.
Solid tumors are thought to present a number of challenges to the effectiveness of CAR T-cell therapy, including in selecting appropriate target antigens given the substantial heterogeneity even within an individual tumor. CD19 has proven to be a success story in part because of its near uniform expression on B-cell malignancies and minimal presence on normal B cells.
The success of drugs targeting members of the human epidermal growth factor receptor (HER) protein family makes them obvious targets for CAR T-cell therapy in solid tumors. Extracellular glycoproteins such as mucin-1, mucin-16, and mesothelin, as well as other cell surface proteins like fibroblast activation protein and L1 cell adhesion molecule, are also being explored for targeting moieties (TABLE 2).14-17
Positive data have been demonstrated in trials of HER2- targeted CARs in patients with sarcoma. Findings from a small phase I study were presented at the 2017 ASCO meeting. Among 6 patients with refractory or metastatic HER2-positive sarcoma, 1 patient achieved a CR, 2 had stable disease, and 3 had progressive disease. The median overall survival in 2 patients who were still alive was 14.2 months.18
CARsgen Therapeutics is developing CAR T cells that target glypican-3 (GPC-3), a cell surface protein that is highly expressed in hepatocellular carcinoma (HCC). Phase I clinical trial data, presented at the 2017 ASCO meeting, showed that it was a safe and feasible treatment for Chinese patients with relapsed/refractory GPC-3positive HCC.19This study has been terminated for reasons not related to safety, efficacy,or feasibility. A phase II trial in this indication is planned, and a phase I/II trial in pediatric solid tumors is ongoing.
The need to expand the impact of CAR T-cell therapy by addressing challenges posed by solid tumors, as well as to improve its safety and efficacy, is driving the development of several innovative strategies for next-generation CAR design. One strategy to overcome TAA heterogeneity in solid tumors is to broaden the specificity of CAR T cells, either by transferring more than 1 population of CAR T cells targeting different antigens or by producing CARs that target multiple TAAs, so-called tandem or bispecific CARs. Several of the latter are currently being evaluated in clinical trials (NCT03097770, NCT03097770, NCT03271515, NCT03271632).14,20
The biopharmaceutical company Endocyte is collaborating with Seattle Children’s Research Institute to develop a novel approach that could create a universal CAR T-cell system capable of targeting any TAA. Their CAR T cells are engineered to bind to small adaptor molecules composed of a fluorescein isothiocyanate (FITC) molecule conjugated to a ligand that targets overexpressed receptors on tumor cells. Instead of the CAR T cells binding to a TAA, an anti-FITC receptor on the surface of the CAR T cell will bind to the FITC end of the adaptor molecule, which allows the activation of the targeted TAA on tumor cells by the free ligand component of the adaptor molecule.21
Another unique hurdle to CAR T-cell therapy in solid tumors is the tumor microenvironment, which presents physical, chemical, and biological barriers preventing entry to the tumor site. Many solid tumors are surrounded by a dense stroma and abnormal vasculature that impede access. The hypoxic, nutrient-deprived, acidic microenvironment is inhospitable for infiltrating T cells. In addition, both the tumor and the microenvironment exert immunosuppressive effects that hinder T-cell migration to the tumor site, and inhibitory immune checkpoints that switch off infiltrating T cells.
Armored CARs and TRUCKs (T cells redirected for universal cytokine killing) are being explored in clinical trials. These are designed to deliver a payload, such as cytokine interleukin-12 (IL-12), on antigen binding, which can help to make the tumor microenvironment more hospitable to immune cells. In a mouse model of ovarian cancer, IL-12secreting CAR T cells were found to be more effective and improved survival.22TRUCKs expressing catalase to protect T cells from oxidative stress, or heparanase, and to improve T cell penetration through the stroma, are also being explored.23
To overcome the ability of tumor cells to suppress the antitumor immune response through exploitation of immune checkpoint pathways, combinatorial trials of CAR T-cell therapy and immune checkpoint inhibitors are underway. CAR T cells that express antibodies against the PD-1 and cytotoxic T-lymphocyte antigen-4 are in early clinical trials.17Efforts to make CAR T-cell therapies safer are also under way, in particular seeking ways to control CAR T-cell activity once the cells have been infused into the patient, to minimize their impact on healthy cells. One such strategy being explored is to introduce a safety switch into the CAR design, such as suicide genes or inducible apoptotic molecules.24Another is the development of so-called inhibitory CARs, designed to provide a T-cell inhibitory signal when engaging an antigen on healthy cells.14,16
A particularly exciting research avenue is the application of state-of-the-art genome-editing technologies to the production of CAR T cells. The CRISPR/Cas9 system could be used to create a coveted “off the shelf” CAR T-cell therapy from universal donor cells, which would help to overcome some of the complexities associated with the manufacturing of CAR T cells.
The use of donor T cells is limited by the presence of endogenous TCR and human leukocyte antigens on their surface, which can activate an immune response and result in GvHD. Genome editing is being used to knock out the TCR and HLA genes to bypass these problems, in addition to knocking out genes that can dampen the antitumor immune response, such as PD-1.
The first human trial of CRISPR immunotherapy was recently given the green light by the FDA and will evaluate the feasibility and safety of knocking out the PD-1 gene in the T cells of patients with lung cancer (NCT02793856). Another CRISPR study seeking FDA approval will knock out the PD-1 and TCR genes in NY-ESO-1targeted TCR-engineered T cells. CRISPR could also be used to improve the delivery of CAR genes. While viral delivery randomly inserts the CAR into the T-cell DNA, CRISPR could more precisely place it in a specific part of the genome. The potential benefit of this approach has been demonstrated in a preclinical study, in which CRISPR was used to insert a CAR into the T-cell receptor alpha constant locus, which generated a more uniform expression of the CAR and produced CAR T cells that outperformed traditional ones in mice with ALL.25-28