Chimeric antigen receptor T-cell therapy is an immunotherapy in which the patient's own T cells are isolated in the laboratory, redirected with a synthetic receptor to recognize a particular antigen or protein, and reinfused into the patient.
Chimeric antigen receptor (CAR) T-cell therapy is an immunotherapy in which the patient’s own T cells are isolated in the laboratory, redirected with a synthetic receptor to recognize a particular antigen or protein, and reinfused into the patient. Clinical trials of CAR T cells directed to the CD19 antigen have shown impressive results in advanced B-cell malignancies at multiple academic centers over the last 3 years. We describe the technology of CAR T cells, findings at 5 academic centers, and toxicities associated with CAR T cell-treatment and their management. Although CAR T cells for B-cell malignancies are the most advanced in terms of clinical testing, CAR T cells are the basis of a new platform technology that is poised to be expanded to other hematologic and nonhematologic malignancies, especially as new targets are identified and manufacturing processes are streamlined.
Chimeric antigen receptor (CAR) T-cell therapy is an immunotherapy in which the patient’s own T cells are isolated in the lab, redirected with a synthetic receptor to recognize a particular antigen or protein, and re-infused into the patient (Figure). Over the last 5 years, at least 15 clinical trials of CAR T cell therapy have been published. A new wave of excitement surrounding CAR T cell therapy began in August 2011, when investigators from the University of Pennsylvania (Penn) published a report on 3 patients with refractory chronic lymphocytic leukemia (CLL) who had long-lasting remissions after a single dose of CAR T cells directed to CD19.1,2 Since then, the enthusiasm has only increased as investigators from the National Cancer Institute (NCI),3,4 Memorial Sloan Kettering Cancer Center (MSKCC),5 which allowed us to transition most of these patients to a standard- of-care allogeneic hematopoietic stem cell transplant (allo- SCT), and Penn6 have published impressive results in lymphoma and acute lymphoblastic leukemia (ALL). At large meetings, including those of the American Society of Hematology (ASH) and the American Association for Cancer Research (AACR), new sessions devoted to CAR T cell therapies have been organized and well attended, and all the major centers developing CAR T cells, including also Baylor College of Medicine, Fred Hutchinson Cancer Research Center, City of Hope National Medical Center, and MD Anderson Cancer Center, have discussed their latest results. Tellingly, what was considered a fringe field only 5 years ago has recently attracted investors, and several new partnerships and biotechnology companies have been formed around CAR T cell therapies (Table).
CARs are synthetic molecules that minimally contain: (1) an antigen-binding region, typically derived from an antibody, (2) a transmembrane domain to anchor the CAR into the T cells, and (3) 1 or more intracellular T cell signaling domains.7 A CAR redirects T cell specificity to an antigen in a human leukocyte antigen (HLA)-independent fashion, and overcomes issues related to T cell tolerance. Thus far, CAR T cells directed against CD19 are the only ones that have achieved a clinical proof-of-concept and are in later stages of clinical development. However, CAR-modified T cells are a platform for a therapy that could be extended to other hematologic and nonhematologic tumors, provided the CAR is specific to the desired proteins on the surface of the tumor cell. The CAR can be introduced into the T cells by a variety of techniques, some of which permanently modify the genome of the T cell (integrating vectors) and have the potential for long-term effects, and some of which are transient (nonintegrating). Examples of integrating methods are retroviral, lentiviral, and transposase-based vectors; examples of nonintegrating methods are adenoviral vectors and plasmid or RNA electroporation. Each method carries unique benefits and concerns related to cost of development, safety, and the durability of CAR expression. The intracellular domains included in the CAR (ie, CD28-z or 4-1BB-z) can modulate T cell properties, such as their ability to survive and proliferate.
Hematologic malignancies are a natural setting for initial testing of CAR T cell therapy: There is extensive knowledge of the surface antigens on leukemic cells (some of which are exclusively expressed in specific hematopoietic cell populations), T cells naturally home to hematologic organs and tissues, and investigators can reasonably and repetitively sample blood and marrow to monitor disease and the infused CAR T cells. To date, most studies have focused on the treatment of advanced B-cell malignancies, including CLL, ALL, and non-Hodgkin lymphoma (NHL), such as diffuse large B-cell lymphoma (DLBCL), using CAR T cells directed to the CD19 antigen found uniquely on B cells. The majority of these studies have utilized second-generation CARs containing a CD28 intracellular signaling domain (CD19-28z), while some studies have utilized a second-generation CAR with a 4-1BB intracellular domain (CD19-BBz). There are currently 5 centers conducting advanced clinical trials of CD19 CAR T cell therapy. Each center uses its own type of CAR, method of introducing the CAR into the T cells, and clinical study design.
University of Pennsylvania
Memorial Sloan Kettering and Fred Hutchinson Cancer Centers
Baylor Center for Cell and Gene Therapy
National Cancer Institute
MD Anderson Cancer Center
Johnson & Johnson
The University of Pennsylvania has open trials to test CAR T cell therapy for the treatment of pediatric and adult ALL, as well as CLL and NHL (NCT01626495, NCT02030834, NCT02030847, NCT01551043). The first report from the university came from a study of 3 patients with relapsed/refractory CLL who were treated with CAR T cells; 2 of the patients achieved a complete response (CR), and 1 had an impressive partial response (PR).2 Notably, the CAR T cells expanded inside the patients and mediated their effects, both responses and toxicities, in a relatively delayed fashion (several weeks after infusion).2 In pediatric ALL, the first 2 patients reported on achieved complete remissions; one was durable (and remains ongoing) and one was transient due to outgrowth of a subpopulation of tumor cells that did not express CD19, known as “antigen escape.”6 Importantly in this study, the major toxicity of cytokine release syndrome was aborted rapidly and completely with administration of tocilizumab, an anti-IL-6 receptor antibody.
The Penn group presented 4 abstracts updating the results of their ongoing clinical trials at ASH 2013. In 1 report, 22 pediatric and 5 adult patients with resistant/recurring (r/r) ALL were treated with CD19-BBz CAR transduced T cells.8 This cohort experienced an 89% CR rate to the therapy, although there were 6 who relapsed following an initial response. The CAR T cells were found to persist in the blood, marrow, and cerebrospinal fluid of patients, and the persistence correlated with response. Two abstracts reported the results of 2 separate CAR trials for patients with r/r CLL: in a dose-escalation trial, 18 patients were treated and an overall response rate (ORR) of 39% (3 CR and 4 PR) was seen; the infused dose did not seem to have an impact on efficacy or toxicity.9 In the second trial, an ORR of 57% was reported, and all of the responding patients experienced a cytokine-release syndrome and B cell aplasia.10 Finally, in an abstract presenting the immune monitoring of all patients treated at Penn, the investigators reported that robust in vivo expansion of CAR T cells correlated with responses and B cell aplasia, and cytokine release syndrome was concurrent with peak CAR T cell expansion.11
The NCI currently has 3 trials testing CD19-28z CAR T cells for the treatment of B cell malignancies, including relapsed/ refractory B cell malignancy (NCT009243264), relapsed B cell malignancy post-allogeneic stem cell transplantation (SCT; NCT0108729412), and pediatric B-cell malignancy (NCT01593696). Complete and partial responses lasting several months have been observed in the 30% to 67% range in all 3 trials, with the highest response rates in ALL. In the relapsed/ refractory setting, patients receive lymphodepleting chemotherapy with high-dose cyclophosphamide and fludarabine before CAR T cell infusion, and IL-2 after CAR T cell infusion; both pretreatment with lymphodepleting chemotherapy and posttreatment cytokine therapy are thought to improve the persistence, and therefore efficacy, of the CAR T cells. Pediatric patients receive lymphodepleting chemotherapy prior to CAR T cells, but no IL-2 after CAR T cells. Because of safety concerns regarding graft-versus-host disease, post-allogeneic patients receive donor-derived CAR T cells without lymphodepletion or IL-2.
The first MSKCC report of CAR T cells was mixed but telling in that the single patient to achieve a clinical response was the patient with ALL; the other 8 patients reported all had refractory CLL and none of them achieved a CR.13 Most patients tolerated CAR T cell infusions well, while 1 (a patient with CLL) died shortly after transfusion (possibly caused by sepsis). In the first follow-up report of the MSKCC ALL trial, rapid and deep remissions were observed: 5 relapsed patients with ALL received treatment with CAR T cells and all of them responded.14 The duration of response to CAR T cells was not evaluable because the patients were treated with allogeneic SCT when they achieved negative minimal residual disease (MRD) status. In the second follow-up report on the MSKCC ALL trial, reporting on 16 patients, a CR rate of 88% was observed; again, those who were eligible went on to allogeneic transplant (NCT01044069).15 At ASH 2013, the investigators reported on a new study of previously untreated CLL patients, where CARs are administered as consolidation after pentostatin/cyclophosphamide/ rituximab chemotherapy. Of the 7 patients treated to date, 3 who had PR after chemotherapy achieved CR, 2 maintained PR, and 2 had progressive disease for an ORR of 43%.16 Additional CAR T cell trials at MSKCC are investigating the efficacy of CAR T cells in the post-autologous SCT setting for lymphomas (NCT01840566), in the setting of relapsed ALL after allogeneic SCT (NCT01430390), and in pediatric relapsed ALL (NCT01860937).
Investigators at Baylor College of Medicine in Houston have 3 ongoing clinical trials of their CAR T cells. Unlike the other groups discussed here that are using bulk T cells to make CAR product, the Baylor group is also using virus-specific T cells for their CAR T cells; in addition to carrying the CAR construct, the T cells are specific to adenovirus, Epstein-Barr virus (EBV), cytomegalovirus (CMV), and/or BK virus, all of which can cause complications in the posttransplant setting. In addition to their potential ability to control viremia, the rationale for using virusspecific T cells includes enrichment of long-lived central memory T cells and continued antigen stimulation of these T cells through their virus-specific T cell receptor. One difficulty with virusspecific CAR T cells is that it takes longer to generate a product for a patient. In an update presented at ASH 2013, the group reported that their post allogeneic SCT trial (NCT00840853) had enrolled and infused 9 patients with ALL or CLL with the dualtropic CARs and a 22% ORR was achieved, with 1 CR and 1 PR.17
Image courtesy of Novartis Pharmaceuticals Corporation; reprinted with permission. Copyright © 2014 Novartis Corporation.
The CARs were also found to exhibit antiviral activity in vivo. The CR lasted 3 months, the PR lasted 8 months, and stable disease lasted 8 months. Other open CAR trials at Baylor include a competitive repopulation study in which patients with NHL and CLL are being treated with 2 CARs at the same time to evaluate differences in the CAR signaling domains (NCT01853631) and a trial combining CAR T cells with checkpoint blockade (ipilimumab) (NCT00586391). Initial results from this type of competitive repopulation design have been published and demonstrated that second-generation CARs engrafted and expanded better in patients than first-generation CARs.18
MD Anderson Cancer Center in Houston is conducting 4 phase I clinical trials to test CAR T cells for treatment of B-cell malignancies. Abstracts describing the progress of these studies were recently presented at ASH 2013.19,20 A total of 32 patients have been enrolled into the 4 trials, most having ALL or NHL; although the numbers are small, the remission rates thus far appear to be high (37% - 50%) and durable for at least 6 months. In 1 trial, CAR T cells are administered with standard chemotherapy for CLL (NCT01653717); in the other 3 trials, CAR T cells are administered after autologous SCT for lymphoma (NCT00968760), or after allogeneic SCT (NCT01497184), or umbilical cord SCT (NCT01362452) for leukemias.
Although it is early in the clinical experience with CAR T cells, there appear to be 3 major types of toxicities: B cell aplasia, cytokine release syndrome (CRS), and tumor lysis syndrome. Of note, all of these are directly related to the efficacy of CAR T cells as a therapeutic agent, as most patients who respond to CAR T cells experience these toxicities to some degree.
B cell aplasia is a consequence of the effective targeting of the CD19 antigen. Fortunately, B cells are not a life-sustaining type of tissue, and it is for this reason that most investigators have used CD19 as the initial target for CAR T cells. Targeting normal tissues is probably the largest hurdle in expanding CAR T cells to other tumors. The fact that CAR T cells discriminate only on the basis of target expression and not on tumorigenic mutations or aberrant signaling pathways is both their strength and weakness, as it is probably not necessary to target driver mutations, but it is important to rule out expression of the antigen on life-sustaining tissues before administration of CAR T cells.21
Expansion of the CAR T cells in the body is associated with CRS. Manifestations of CRS include fever, hypotension, hypoxia, and neurologic changes. Additional clinical and biochemical changes may occur, such as disseminated intravascular coagulation and/or transaminitis associated with marked elevations in ferritin and C-reactive protein, a constellation of findings similar to macrophage activation syndrome or hemophagocytic lymphohistiocytosis (HLH). Neurologic complications, including seizures, aphasia, and mental status changes have also been reported, and appear to be reversible in the vast majority of cases.5. Cytokine release syndrome can occur regardless of the tumor subtype, and has been observed in ALL, CLL, and lymphomas. Initially, investigators encountering severe CRS managed it with high-dose steroids (>100 mg daily of prednisone equivalent); however, aborting the CRS with steroids is not an ideal intervention because steroids impair CAR T cell function and may be insufficient to manage the syndrome.6 Based on the results of immune monitoring studies that linked CRS to significant elevations in multiple serum cytokines, including IL-6, subsequent patients who exhibited severe CRS have been treated with tocilizumab, an IL-6 receptor—blocking antibody. Tocilizumab dramatically improves the CRS syndrome and thus far does not appear to affect CAR T cell function or tumor responses.
Because CAR T cells have mostly been developed in the academic setting, each group has developed its own manufacturing processes and CAR designs. One of the big challenges moving forward will be optimizing the complex manufacturing and distribution of CAR T cells such that patients all over the world can receive this promising therapy.22 It will also be interesting to learn with time which construct of CAR T cells allows for the greatest responses and least toxicity. CAR T cells may someday be effective enough to challenge allogeneic bone marrow transplant as definitive therapy for lymphoid malignancies, and perhaps even other hematologic malignancies as new targets for myeloid and plasma cell malignancies are defined and tested in clinical trials.
ABOUT THE AUTHORS
Affiliations: Sagar B. Kudchodkar, PhD, is scientist, and Marcela V. Maus, MD, PhD, is assistant professor and director of translational medicine and early clinical development, from the Translational Research Program, Abramson Cancer Center, University of Pennsylvania, Philadelphia, PA.
Address correspondence to: Marcela V. Maus, MD, PhD, Translational Research Program, Abramson Cancer Center, University of Pennsylvania, Smilow Center for Translational Research, 3400 Civic Center Blvd, Bldg 421, 8th Fl, Rm 154, Philadelphia, PA 19104-5156; phone: (215) 746-5554; fax: (215) 573-0077; e-mail: firstname.lastname@example.org.
Disclosures: Drs. Kudchodkar and Maus have received sponsored research support from Novartis; Dr. Maus is supported by NCI K08CA166039.