More than 7 years have passed since the regression of advanced lymphoma was first reported in a patient who had undergone the infusion of T cells engineered to express a chimeric antigen receptor (CAR) targeting the CD19 antigen expressed on the surface of both normal and malignant B cells.1 Subsequent trials of CD19-targeted CAR T-cell therapy showed a complete response in some patients with relapsed or chemotherapy-refractory hematologic cancers for which there were no effective therapies.2-5

Figure 1. Figure 1. Chimeric Antigen Receptor T-Cell Therapy. Panel A shows the general steps in chimeric antigen receptor (CAR) T-cell therapy, in which white cells are first isolated from the patient by means of leukapheresis and then are taken to a Good Manufacturing Practice (GMP) production facility, where T cells are activated and genetically engineered with a retrovirus encoding the CAR. CAR T cells are further expanded, harvested, and finally reinfused into the patient, who has undergone a preparative lymphodepletion regimen before infusion. In the ZUMA-1 study by Neelapu et al., the entire process took a median of 17 days. Panel B shows the differences in design of the anti-CD19 CARs used in the study by Neelapu et al. and those used in the study by Schuster et al. TCR denotes T-cell receptor.

This personalized therapeutic approach entails the removal of peripheral-blood T cells from a patient, followed by in vitro activation, genetic modification, and expansion of the T cells under Good Manufacturing Practice conditions, and finally the infusion of the cells back into the patient (Figure 1A). Because of the challenging toxic effects and complexity of this promising therapy, it has been unclear whether the approach can be used to treat a large number of patients in clinical settings other than highly specialized academic centers.

Two studies — a multicenter, phase 2 trial (ZUMA-1) by Neelapu et al.6 and a smaller case-series study by Schuster et al.7 — the results of which are now published in the Journal, validate the efficacy of CD19-targeted CAR T-cell therapy in patients with refractory lymphomas. In the two studies, patients received autologous T cells that were genetically engineered to express second-generation anti-CD19 CARs that were composed of an extracellular antigen-binding domain derived from the single-chain variable fragment of the CD19-specific antibody FMC63, along with two intracellular T-cell signaling domains, one derived from the T-cell receptor CD3-ζ chain and the other from either CD286 or 4-1BB7 costimulatory molecules to promote further T-cell activity (Figure 1B).

Despite the differences in CAR design and lymphodepletion regimens that were used in the two studies, the clinical results were quite similar. Neelapu et al. report a 54% rate of complete response in 101 patients with refractory large B-cell lymphoma after treatment with the anti-CD19 CAR T-cell therapy axicabtagene ciloleucel. Schuster et al. report a 57% complete response rate in 28 patients with refractory B-cell lymphomas, including double-hit lymphomas, who were treated with the anti-CD19 CAR T-cell product CTL019. The durability of these responses remains to be determined, but National Cancer Institute investigators recently reported that 4 of 5 patients who had a complete response after anti-CD19 CAR T-cell therapy remained disease-free at the latest follow-up at 3.2 to 4.7 years; in the 4 patients with a complete response, recovery of normal B cells was observed in 3 patients.8

These two studies reinforce the risks of unique and often serious toxic events that have been observed in earlier trials. Three treatment-related deaths (3.0%) were documented in ZUMA-1, and one death (3.6%) was reported by Schuster et al. The cytokine-release syndrome (which most commonly included fever, hypoxia, and hypotension) was observed in 93% of the patients in the ZUMA-1 trial (13% with events of grade 3 or higher) and in 57% of the patients in the study by Schuster et al. (18% with events of grade 3 or higher). Neurologic toxic effects (which most commonly included encephalopathy, confusion, and tremor) were observed in 64% of the patients in the ZUMA-1 trial (28% with events of grade 3 or higher) and in 39% in the study by Schuster et al. (11% with events of grade 3 or higher). The majority of these side effects were reversible, with no clinical sequelae.

ZUMA-1 is a landmark study because it involved 22 institutions and showed that a personalized gene-engineered T-cell product could be rapidly generated at a centralized cell-manufacturing facility and safely administered to patients at transplantation-capable medical centers. T-cell products were generated with a 99% success rate, and patients received CAR T cells a median of 17 days after leukapheresis — a rapidity that is critical, given the aggressive nature of refractory large B-cell lymphomas. ZUMA-1 thus provides a blueprint for treating a large number of patients with gene-engineered T cells.

Despite the impressive clinical results, approximately half the patients with refractory or relapsed large B-cell lymphomas will not have a durable response after anti-CD19 CAR T-cell therapy. The reasons for this variation in response are not completely understood, but a number of strategies are being investigated to enhance the therapeutic efficacy of CAR T cells against lymphoma. These strategies include using CARs that target different or multiple B-cell antigens (e.g., CD20 and CD22) to combat CD19-antigen loss variants and optimizing CAR design by using different costimulatory domains and transmembrane and hinge regions. Targeted insertion of anti-CD19 CARs into specific genetic loci, such as in the gene encoding the T-cell receptor alpha constant domain (TRAC), with the use of CRISPR–Cas9 gene-editing technology9 and the introduction of anti-CD19 CARs into defined, more potent T-cell populations have been shown to enhance the antitumor activity of CAR T cells in preclinical models. Combining anti-CD19 CAR T-cell therapy with other agents, such as Bruton tyrosine kinase inhibitors and inhibitors of immune checkpoint pathways (e.g., programmed death 1 [PD-1] protein or its ligand [PD-L1]), may also increase therapeutic efficacy. However, because the mechanisms of the toxic effects associated with this therapy are not completely understood, strategies that enhance the potency of these products run the risk of inadvertently worsening the already daunting toxic effects.

The recent approval of anti-CD19 CAR T-cell therapy for the treatment of relapsed or refractory acute lymphoblastic leukemia and large B-cell lymphomas by the Food and Drug Administration sets a new standard of care for the patients who receive these therapies. However, the approval also comes with substantial economic challenges because of the high cost of care, a challenge that will grow as the indications for these therapies expand in the future. Policies will need to be developed to ensure that eligible patients receive these potentially curative therapies.

CAR-engineered T cells are a revolutionary treatment for patients with advanced blood cancers. Unfortunately, CAR-engineered T cells have been largely ineffective in patients with metastatic solid cancers, which account for the majority of cancer-related deaths. One major factor contributing to this lack of efficacy is the paucity of suitable cell-surface molecules that are expressed by solid cancers for CARs to target. This roadblock can be overcome by engineering T cells with T-cell receptors that can recognize tumor-specific antigens derived from intracellular proteins, such as neoantigens arising as a consequence of somatic mutations expressed by tumors,10 which are found in the majority of patients with metastatic epithelial cancers. T cells that are genetically modified to express T-cell receptors targeting mutated neoantigens may represent a new wave of promising and highly personalized gene-engineered T-cell therapies for patients with metastatic solid cancers.