Chimeric Antigen Receptor T-Cell Therapy for the Community Oncologist

An immunotherapy that has shown promise in cancer treatment is chimeric antigen receptor-modified (CAR-modified) T lymphocytes. CARs are engineered receptors constructed from antigen recognition regions of antibodies fused to T-cell signaling and costimulatory domains that reprogram a patient’s T cells to specifically target tumor cells. The current landscape of CAR T-cell therapy, effective management of patients undergoing treatment, and the most suitable candidates for current trials are discussed.

Abstract

The field of cancer immunotherapy has rapidly progressed in the past decade as several therapeutic modalities have entered into the clinic. One such immunotherapy that has shown promise in the treatment of cancer is the use of chimeric antigen receptor (CAR)-modified T lymphocytes. CARs are engineered receptors constructed from antigen recognition regions of antibodies fused to T-cell signaling and costimulatory domains that can be used to reprogram a patient’s T cells to specifically target tumor cells. CAR T-cell therapy has demonstrated sustained complete responses for some patients with advanced leukemia, and a number of CAR therapies are being evaluated in clinical studies. CAR T-cell therapy-associated toxicities, including cytokine release syndrome, macrophage activation syndrome, and tumor lysis syndrome, have been observed and effectively managed in the clinic. In patients with significant clinical responses, sustained B-cell aplasia has also been observed and is a marker of CAR T-cell persistence that might provide long-term disease control. Education on CAR T-cell therapy efficacy and safety management is critical for clinicians and patients who are considering this novel type of treatment. In the present report, the current landscape of CAR T-cell therapy, the effective management of patients undergoing treatment, and which patients are the most suitable candidates for current trials are discussed.

Implications for Practice:

The present report describes the current status of chimeric antigen receptor (CAR) T lymphocytes as an immunotherapy for patients with relapsed or refractory B-cell malignancies. CAR T cells targeting CD19, a protein expressed on many B-cell malignancies, typically induce high complete response rates in patients with B-cell leukemia or lymphoma who have very limited therapeutic options. Recent clinical trial results of CD19 CAR T-cell therapies and the management of CAR T-cell-associated adverse events are discussed. The present report will therefore inform physicians regarding the efficacy and safety of CAR T cells as a therapy for B-cell malignancies.

The Immune System in Cancer and the Role of Immune Tolerance

It is well established that the immune system plays a central role in preventing both tumor initiation and progression. Histopathological and clinical observations have shown that several characteristics of lymphocytic infiltrates, including the number, type, and location, in primary tumors are prognostic indicators for both disease-free and overall survival in patients with a range of tumors [1]. Furthermore, the immune system can inhibit cancer development by protecting the host against infection from viruses, eliminating pathogens to resolve an inflammatory state, and targeting malignant cells that aberrantly express tumor antigens through a process known as tumor surveillance [2].

The immune response to malignant cells involves the presentation of tumor antigens to T cells directly by tumor cells or indirectly by antigen-presenting cells (APCs) known as dendritic cells. Antigens processed by dendritic cells can be presented to both CD8+ and CD4+ T cells. In addition to presenting antigens to T lymphocytes, APC signaling stimulates lymphocytes to proliferate and differentiate. Two concurrent signals are prerequisites for effective T-cell activation and differentiation: (a) the antigen presented by the APC and its recognition by the T-cell receptor (TCR), and (b) one or more costimulatory signals provided by molecules on the APC, which interact with receptors on T cells. The costimulatory signal that is the best characterized is the interaction between CD28, a costimulatory receptor expressed on the surface of the T cell, and B7 (either B7-1 or B7-2) costimulatory ligands, present on the surface of the APC. The CD28-B7-1/2 costimulation activates clonal expansion and differentiation of activated T cells [3]. Another costimulatory signal is the interaction of 4-1BB ligand, which is expressed on activated dendritic cells, macrophages, and B cells, with 4-1BB (also known as CD137), which is expressed on T cells [4]. With a sufficient quantity of the appropriate signals, cytotoxic T lymphocytes are then able to kill tumor cells, proliferate to greater numbers, and differentiate into memory cells [5, 6].

Tumors use several strategies to evade immune surveillance and elimination [2, 5]. The selective pressure applied by the immune system on tumor cells exhibiting genetic instability selects for cells that have lost immunogenic markers, resulting in a lack of T-cell recognition of the tumor. Furthermore, impaired antigen presentation due to mutations in major histocompatibility complex (MHC) genes or genes required for antigen processing is commonly observed in tumors. Tumors also establish an immunosuppressive state within the microenvironment by producing cytokines such as interleukin-10 (IL-10) and transforming growth factor-β [7–9]. It has also been shown that tumors often express negative costimulatory molecules that inhibit T-cell activity, including programmed death ligand-1 (PD-L1), the ligand for the T cell-expressed programmed cell death-1 (PD-1) receptor [10–13].

In an attempt to overcome immune tolerance, several therapeutic strategies have been investigated. Highly specific monoclonal antibodies against targets of interest can now be developed and thus might target tumor-promoting factors, such as growth factors or tumor antigens [14, 15]. For example, rituximab and ofatumumab (anti-CD20), trastuzumab (anti-HER2 [human epidermal growth factor receptor 2]), and bispecific T-cell engagers such as blinatumomab (anti-CD19) have been developed and have shown promise in the clinic [16–18]. Moreover, immunomodulators and checkpoint inhibitors have been investigated to potentially amplify an immune antitumor response [19].

An advantage of cell-based immunotherapy is the ability to overcome immune tolerance and generate immune memory. Advances in gene transfer technologies have led to the strategy of combining gene-based therapies with cell-based therapies [20, 21]. One method that was developed involved the adoptive transfer of bulk T lymphocytes obtained from peripheral blood that have been transduced to express tumor-specific TCRs [22–24]. This approach overcomes the limited repertoire of T cells that are specific to a patient’s tumor. However, several challenges to this approach exist, including human leukocyte antigen restriction and the requirement that the tumor cells adequately express MHC. Additionally, the transduced T cells could be subject to the same negative regulatory factors in the tumor microenvironment as the host’s normal T cells.

In contrast to TCR-modified T cells, the use of a recombinant receptor molecule known as a chimeric antigen receptor (CAR) to genetically reprogram a patient’s own T cells to specifically target the cancer is MHC independent [25, 26]. CARs can also be designed to incorporate costimulatory signals, which could make the CAR T cells less susceptible to negative regulation by the host. Furthermore, CARs can genetically reprogram both CD4+ and CD8+ T cells equally well, although TCR-transduced T cells might depend on either CD4 or CD8 coreceptor binding for efficient antigen recognition [27]. Genetically reprogrammed CD4+ CAR T cells have proved to be as cytolytic as CD8+ T cells [28–30].

CARs and CAR T-Cell Therapy

CARs combine an antibody fragment with intracellular signaling domains, generating a single chimeric protein (Fig. 1) used to reprogram immune cells into “hunter cells” that target tumors [31]. CAR T-cell therapy combines the specificity of an antibody with the cytotoxic and memory functions of T cells [32]. CAR specificity comes from the extracellular domain, which is derived from the antigen-binding site of a monoclonal antibody. A single-chain variable fragment is generated by linking the variable region heavy and light chains. A flexible peptide linker is included, and this extracellular domain is fused to an intracellular domain via a transmembrane sequence. The intracellular domain attempts to recapitulate the normal series of events by which T cells are activated (Fig. 2). The iterative process of modifying the intracellular domain led to the development of first-, second-, and, recently, third-generation CARs. First-generation CARs consisted of only the TCR complex CD3-zeta chain domain and antigen recognition domains [25]. The use of first-generation CARs was examined in clinical trials that included patients with several cancers, including lymphoma and ovarian cancer, and showed only modest efficacy, primarily as a result of insufficient T-cell persistence in vivo [33–36]. Second-generation CARs were subsequently developed that incorporated costimulatory domains, such as CD28 or 4-1BB (CD137), to augment CAR T-cell survival and proliferation, leading to increased antitumor activity [37–40]. So-called third-generation CARs that combine the CD3-zeta domain with ≥2 costimulatory domains are still early in development and have not yet been shown to be an improvement over second-generation CARs. A number of CAR T-cell therapies with differences in receptor design, the vector used for gene delivery, and manufacturing process have been developed, and many have proceeded to clinical trials [41]. Thus far, many of the CAR T-cell therapies generated have been directed at CD19, a cell-surface protein whose expression is restricted to B cells and B-cell precursors and is expressed on the surface of most B-cell malignancies [42].

Figure 1.

CAR structure.

Abbreviations: APC, antigen-presenting cell; CAR, chimeric antigen receptor; mAb, monoclonal antibody; MHC, major histocompatibility complex; scFv, single-chain variable fragment; TCR, T-cell receptor.

Figure 2.

Mechanism of action of CAR-modified T cells.

Abbreviations: AP-1, activator protein 1; CAR, chimeric antigen receptor; Fas-L, Fas ligand; IFN-γ, interferon-γ; IL-2, interleukin-2; LAT, linker of activated T cells; MTOC, microtubule-organizing center; NFAT, nuclear factor of activated T cells; NF-κB, nuclear factor κ-light-chain-enhancer of activated B cells; P, phosphate group; ZAP70, ζ-chain-associated protein kinase 70.

In general, the process of manufacturing and delivery of CAR T cells involves the following (Fig. 3): (a) leukapheresis—a patient’s T cells are isolated and harvested through apheresis; (b) T-cell activation—antibody-coated beads serving as artificial dendritic cells are used to activate the isolated T cells; (c) transduction—activated T cells are genetically reprogrammed ex vivo by the lentiviral-mediated transduction of a construct encoding the CAR; (d) expansion—genetically reprogrammed T cells undergo further ex vivo expansion; (e) chemotherapy—before T-cell infusion, the patient receives a preparative lymphodepleting regimen; and (f) CAR T-cell infusion—genetically engineered T cells are infused into the patient.

Figure 3.

Overview of CTL019 therapy in the clinic.

Patient Selection

The patient populations examined in the clinic in which CAR T-cell therapy was successful have primarily been those with CD19+ B-cell malignancies, namely relapsed/refractory (r/r) acute lymphoblastic leukemia (ALL), chronic lymphocytic leukemia (CLL), and B-cell lymphomas. Outside the context of CAR T-cell therapy, although the initial treatment for most pediatric patients with ALL is successful, an unmet need exists for the 20% of the patients who develop a relapse after achieving an initial complete response (CR) to conventional therapies [43]. In contrast, adults with ALL fare much worse, with lower remission rates and higher toxicity rates. After relapse, the outcomes for patients with ALL are generally poor, with 5-year overall survival rates of 7%–10% in adults and ≈30% in children using currently available therapies [43–45].

Currently, no standard of care is available for the treatment of adult or pediatric patients with r/r ALL. With respect to CLL, the prognosis is especially poor for patients with high-risk disease, refractory disease, or early relapse [46]. Allogeneic hematopoietic stem cell transplant (alloSCT) is a potentially curative option for r/r ALL and CLL, but not all patients are eligible, as the upper age limit at most transplant centers is 60–70 years, and patients with significant comorbidities are also ineligible [47, 48]. Furthermore, patients with CLL have a high morbidity and mortality rate after alloSCT. The nonrelapse mortality rates range from ≈15%–50%, depending on whether myeloablative or reduced-intensity conditioning is used [49, 50]; ≈50% of patients develop chronic graft-versus-host disease [51]. High infection rates can also lead to mortality [52]. Despite recent therapeutic advancements, a great need still exists to improve the outcomes of patients with ALL and CLL, in particular, those with r/r disease. This is also the case for patients with r/r CD19+ non-Hodgkin lymphoma (NHL), including diffuse large B-cell lymphoma (DLBCL) and follicular lymphoma (FL), for which several CAR T-cell therapy trials are underway. Thus, a significant number of patients might benefit from CAR T-cell therapy.

Ideally, patients appropriate for CAR T-cell therapy will display the following characteristics: good performance status, adequate organ function and physiological reserve to tolerate pronounced fevers and accompanying symptoms, and a lack of other suitable low-risk treatment options. Several considerations must be considered when identifying a patient suitable for CAR T-cell therapy. CAR T-cell therapy should be used in patients with cancers that express a target surface antigen such as CD19 to sustain potential benefit. In some circumstances, patients will have very few or none of their own functional T cells, such as after multiple treatment cycles with lymphotoxic agents; thus, these patients would likely not have adequate T cells for collection and generation of CAR T cells to benefit from treatment. Also, some patients have T cells that cannot grow in vitro or are unable to sufficiently multiply. Patients with a history of significant autoimmune disease might not be good candidates, because either they require immunosuppressive treatment for their autoimmune disease or they might be at increased risk of autoimmune disease exacerbation after infusion of activated T cells. Patients must also display adequate venous access for the initial apheresis procedure, although the placement of a temporary apheresis catheter is possible and could be justified in some cases. Therefore, if the patient is relatively healthy, the B-cell malignancy is well controlled or in remission, and a suitable donor is available, alloSCT could be considered appropriate rather than CAR T-cell therapy. Because CAR-T cell therapy has shown significant efficacy in a patient population with advanced r/r disease, it is these patients for whom inclusion in trials of CAR T-cell therapy is currently most appropriate. It is as yet unknown whether CAR T-cell therapy will be beneficial as a front-line therapy.

Because CAR-T cell therapy has shown significant efficacy in a patient population with advanced r/r disease, it is these patients for whom inclusion in trials of CAR T-cell therapy is currently most appropriate. It is as yet unknown whether CAR T-cell therapy is beneficial as a front-line therapy.

Clinical Trial Data: Efficacy and Safety Summary

University of Pennsylvania and Children’s Hospital of Philadelphia

As of January 2016, more than 200 patients, including those with ALL, CLL, and NHL, have been treated at the University of Pennsylvania (Penn) and Children’s Hospital of Philadelphia (CHOP), in partnership with Novartis Pharmaceuticals, with CTL019, a CD19 CAR that is transduced by a lentivirus and contains a 4-1BB costimulatory domain. In the recent analysis of the CTL019 experience presented at the European Hematology Association annual meeting in 2015, it was reported that 45 of 48 pediatric patients (94%) with r/r ALL achieved a CR [53]. The median follow-up in that study was 8 months. In a preliminary analysis of 30 patients, 76% had a response that lasted >6 months [54]. In both adult and pediatric patients with r/r ALL, CTL019 achieved a high remission rate, even in patients for whom previous alloSCT had failed [54]. Ten patients in the study experienced relapses, 5 of whom relapsed with CD19-negative disease. No events were reported after 12 months, and only 3 of 39 patients (8%) subsequently underwent SCT [55]. Data from the phase II study evaluating CTL019 in 24 patients with advanced r/r CLL were reported at the American Society of Hematology (ASH) annual meeting in 2014. An overall response rate (ORR) of 42% was observed, with 5 of 24 patients achieving a CR, 5 patients achieving a partial response, and 70% of the patients having a response lasting >9 months [56]. Moreover, CTL019 therapy also induced responses in some patients with advanced r/r FL (ORR 100%) and r/r DLBCL (ORR 50%) [57, 58]. The longest ongoing clinical response durations in that study were >350 days for FL and approximately 400 days for DLBCL [58]. The median progression-free survival for patients with DLBCL who were treated with CTL019 was 90 days, and all patients with FL who were treated with CTL019 had clinical responses of at least 150 days [57, 58]. In patients with r/r ALL, CTL019 sequences remained detectable by quantitative polymerase chain reaction for at least 2 years in patients with sustained remissions [54]. The persistent CTL019 cells are functional and could provide ongoing protection against tumor recurrence/relapse, and the persistence of this therapy has been associated with response in some patients with advanced CLL and ALL.

As of July 2014, of the 97 evaluable patients treated with CTL019, grade 3–5 cytokine release syndrome (CRS) occurred in 64% of patients with ALL and 27% of patients with CLL/NHL [59]. In particular, severe CRS occurred in 11 of the 12 adult patients with ALL, which resolved in 2 and was successfully treated in 6 [59]. However, 3 patients developed refractory CRS resulting in death. All 3 patients with refractory CRS had infectious complications and a significant disease burden and had received higher doses of CTL019. All these factors could have contributed to the worsening of CRS in these 3 patients.

Memorial Sloan Kettering Cancer Center, Fred Hutchinson Cancer Research Center, and Seattle Children’s Hospital

In conjunction with the Memorial Sloan Kettering Cancer Center, Fred Hutchinson Cancer Research Center, and Seattle Children’s Hospital, Juno Therapeutics has investigated the use of multiple CD19 CAR T-cell therapies in B-cell malignancies, with promising results. Similar to CTL019, one of Juno Therapeutics’ CD19 CAR T-cell therapies (known as JCAR017 or JCAR014, depending on the trial) is transduced by a lentivirus and contains a 4-1BB costimulatory domain. JCAR014 therapy resulted in a CR in 5 of 7 adult patients (71%) with r/r ALL at Fred Hutchinson Cancer Research Center, and JCAR017 induced a minimal residual disease-negative CR in 20 of 22 pediatric patients (91%) with r/r ALL at Seattle Children’s Hospital [60, 61]. In clinical studies at Memorial Sloan Kettering Cancer Center, a CD19 CAR transduced by a gammaretrovirus and containing a CD28 costimulatory domain known as JCAR015 induced CR in 29 of 32 adult patients (91%) and in 2 of 4 pediatric patients with r/r ALL [62, 63]. Disease-free survival lasting >1 year was observed in 30% of patients with r/r ALL who were treated with JCAR015, and 17 of 22 patients (77%) included in a preliminary analysis proceeded to SCT after JCAR015 therapy [64]. Both JCAR015 and JCAR014 had efficacy in patients with r/r NHL; however, the CR rates reported were lower for NHL than for ALL. Only 1 of 9 patients (11%) with r/r NHL who were treated with JCAR014 had a CR [60]. In some analyses, CRS was observed in all responding pediatric patients treated with JCAR015 and JCAR017 [63, 65]. In adult patients with r/r ALL, CRS was associated with tumor burden, and none of the patients with minimal residual disease at the time of JCAR015 infusion developed CRS [64]. One adult patient treated with JCAR014 died of CRS-related events [60].

U.S. National Cancer Institute

Clinical success has also been observed in trials at the National Cancer Institute (NCI) in partnership with Kite Pharma using a CD19 CAR known as KTE-19. KTE-C19 contains a CD28 costimulatory domain and is transduced with a retroviral vector. At the ASH 2014 annual meeting, data from the NCI were presented showing that 5 of 8 patients (63%) with r/r DLBCL responded to KTE-C19, with 1 patient having a CR [66]. The duration of response in these patients ranged from 1 to more than 7 months [66]. The group at the NCI had previously reported a 57% CR rate using a higher dose of both lymphodepleting chemotherapy and CAR T cells in patients with r/r DLBCL [67]. The major safety issue associated with KTE-C19 using the lower dose regimen was short-term neurotoxicity [66]. In pediatric patients with r/r ALL, therapy with KTE-C19 resulted in a CR in 14 of 20 patients (70%), and grade 4 CRS was observed in 14% of the patients [68].

An overview of the reported clinical data of CAR T-cell therapy in hematologic malignancies from the 2014 ASH meeting, the 2015 American Association for Cancer Research meeting, the 2015 American Society of Clinical Oncology meeting, and the 2015 European Hematology Association meeting is provided in Table 1.

Table 1.

CAR T-cell therapy: overview of recent clinical data in hematologic malignancies reported at ASH 2014, AACR 2015, ASCO 2015, and EHA 2015

Expert Opinion on Patient Management During CAR T-Cell Therapy-Associated Adverse Events

The initial step in the CAR T-cell therapy process involves apheresis to collect peripheral blood mononuclear cells from which the patient’s T cells can be enriched. Although apheresis is generally regarded as a safe procedure, some patients experience adverse effects, including fatigue, nausea, dizziness, feeling cold, and tingling in the fingers and around the mouth. More serious complications, including abnormal heart rate and seizures, can occur but are extremely rare. Typically, inhibitors of angiotensin-converting enzyme are held for 24 hours to avoid bradykinin release syndrome [69]. A recent retrospective analysis of 15,763 apheresis procedures found only 59 moderate to severe adverse events (AEs) (0.37%), including dizziness or fainting episodes (0.12%), citrate toxicity (0.02%), a combination of dizziness/fainting and citrate toxicity (0.11%), vascular injuries (0.07%), and miscellaneous events (0.04%) [70]. For an adequate cell collection, apheresis should occur before the administration of lymphodepleting chemotherapy.

In our experience with hematologic malignancies, CAR T-cell infusion can be performed as an outpatient procedure; however, toxicities from therapy often require hospitalization. Therefore, after CAR T-cell infusion, patients should be observed for a few hours to ensure no acute reaction develops, and if no acute reaction occurs, the patient can return home. With further follow-up, it is possible that patients will develop prodromal symptoms that can occur hours to days after infusion, and these are associated with low-grade fevers, fatigue, or anorexia.

The primary CD19-directed CAR T-cell therapy-associated AEs that have been observed include CRS [54], neurotoxicity, macrophage activation syndrome (MAS)/hemophagocytic lymphohistiocytosis [71–74], and tumor lysis syndrome (TLS) [54]. An expected consequence of CD19-directed CAR T-cell therapy, CRS is a systemic inflammatory response produced by cytokine elevations associated with T-cell expansion and proliferation. Thus, CRS might also correlate with the efficacy of CD19 CAR T-cell therapy. In patients with ALL, the severity of symptoms has significantly correlated with the baseline tumor burden before CAR T-cell infusion [54]. The diagnosis of CRS is based on clinical symptoms, including fever, one of the first manifestations of CRS, but also myalgia, nausea, anorexia, transient hypoxia, encephalopathy, and transient hypotension [75]. In the CTL019 experience, CRS for patients with ALL typically occurred 1–14 days after CAR T-cell therapy infusion, whereas CRS usually occurred 14–21 days after infusion in patients with CLL [54].

Patients who develop severe CRS require intensive care, including vasopressor support for hypotension [54]. It was shown that severe CRS manifested at ≈1 day after infusion compared with 4 days after infusion for nonsevere cases. In patients who are neutropenic after receiving chemotherapy, it is difficult to determine whether the fever is a result of CRS or of lymphodepletion; therefore, patients should receive supportive care per institutional guidelines and treated for neutropenic fever. In all patients, systemic inflammation markers have been elevated. Compared with patients without CRS, patients with severe CRS displayed higher peak levels of these systemic markers, including IL-6 [54]. The marked IL-6 elevation after CAR T-cell therapy and rapid reversal of severe CRS with the IL-6 receptor antibody tocilizumab have been previously reported; therefore, tocilizumab has been incorporated into severe CRS management [54]. To this end, a novel CRS grading system and management algorithm have been developed based on the clinical experience of patients treated at Penn and CHOP (Table 2) [76].

Table 2.

Novel CRS grading system for CTL019 [76]

Neurological toxicity is an adverse event that has been observed with CD19 CAR T-cell therapies and with blinatumomab [54, 66, 67, 77–83]. In our experience with the Penn and CHOP trials, neurotoxicity occurred either concurrently with the high fever during CRS or after the fever had resolved. Symptoms, which were self-limiting and lasted 2–3 days, included aphasia, confusion, delirium, and hallucinations [54]. Other centers have also observed seizures and obtundation. Neurotoxicity caused by CAR T-cell therapy did not appear to correlate with CRS severity, was not prevented or ameliorated by tocilizumab, and resolved over 2–3 days without apparent long-term sequelae [54]. Instead, neurologic toxicity can be treated with antiepileptic medications and/or steroids. No clear imaging findings have emerged to correlate with the clinical toxicity; however, an analysis of cerebrospinal fluid has shown that CAR T cells do enter this space. The exact mechanism of the neurologic toxicity remains unclear but is an area of active investigation.

MAS/hemophagocytic lymphohistiocytosis is another CAR T-cell therapy-associated AE and typically occurs concurrently with, or shortly after, CRS. MAS is typically characterized by excessive activation of well-differentiated macrophages, can occur outside the context of CAR T-cell therapy, and has specific diagnostic criteria. MAS associated with CAR T-cell therapy has characteristic laboratory patterns, including highly elevated ferritin, C-reactive protein, and D-dimer. In a small number of patients, profound hypofibrinogenemia has been observed, which can be associated with bleeding. Patients have also experienced transaminitis and elevated triglycerides. In severe cases, tocilizumab can effectively treat MAS and is typically given for concurrently occurring CRS in the context of CAR T-cell therapy.

Effective CAR T-cell therapy results in the rapid destruction of tumor cells and thus TLS has also been observed [84, 85]. Most hospitals have standard protocols to effectively manage TLS. Monitoring includes testing calcium, potassium, phosphorus, creatinine, and uric acid levels, and this testing should occur 2–3 times per week, depending on the laboratory results. TLS is associated with hyperkalemia, hyperphosphatemia, and hyperuricemia, and the goal of the laboratory tests is to prevent the development of TLS. Management strategy generally includes fluid hydration, prophylactic allopurinol, and rasburicase. Hemodialysis could be needed in certain cases.

Another long-term effect of CTL019 therapy is B-cell aplasia [54]. Sustained B-cell aplasia is a marker of continued CTL019 persistence and is observed in long-term responders. It is possible that the extended persistence of CTL019 could lead to long-term disease surveillance and also prevent relapse. The CTL019-mediated elimination of normal CD19-expressing precursors and maturing B cells is an on-target, off-tumor toxicity and can be managed by periodic immunoglobulin administration per age and in accordance with location- and disease-based guidelines [86].

Future Directions and Considerations

To broaden the application of T-cell therapy, it is desirable to develop techniques that increase the efficiency and/or promote safety at each step of the process. Although challenges remain in manufacturing single-patient product lots in which the raw material is derived from a patient’s own cells, technologies for isolating, culturing, and increasing the safety and efficacy of CAR T-cell therapy are improving at a rapid pace. In the longer term, the development of universal T-cell products (i.e., “off-the-shelf” CAR T cells) from allogeneic donors, coupled with a knockdown of human leukocyte antigen genes, could lead to therapies in which one lot of the drug will treat many patients [87]. However, it remains to be seen whether universal T-cell products are as potent and durable as autologous CAR T-cell products. In either case, generating a streamlined process by improving current manufacturing techniques and examining testing standards can lead to cost and timeline improvements.

In addition to manufacturing improvements, research is ongoing to identify, validate, and target lineage-specific antigens or tumor-associated antigens in other cancers. Theoretically, CAR T cells can result in deleterious effects by damaging healthy organs and structures when tumor-associated antigens are expressed on normal cells. Thus, the potential on-target but off-tumor effects must be thoroughly investigated for each candidate antigen. Additionally, the benefit of therapy versus this potential cost must be considered for each disease before considering CAR technology. Preclinical models are currently investigating CARs coexpressed with other genes to improve various characteristics of the modified T cell, such as safety, persistence, and effector functions. For example, one strategy involves engineering CAR T cells to also express the NKG2D receptor coupled to CD3-zeta [88]. This receptor engages several stress ligands that are known to be upregulated on tumor cells, and evidence suggests that the genetically reprogrammed T cells also target inhibitory regulatory T cells (Tregs) for destruction. Additionally, CAR T cells can be further modified with additional transgenes that can express specific cytokines to promote the recruitment of tumor-associated macrophages, thereby enhancing both the antigen-presenting and tumor lytic activity of T cells [89]. It has been shown in preclinical models that CAR T cells modified to secrete IL-12 displayed increased antitumor activity through a resistance to Treg-mediated immune suppression, a reduction in Tregs, and induction of B-cell aplasia [90]. Furthermore, preclinical data have suggested that IL-12 recruits natural killer cells and macrophages but also can reprogram a number of myeloid cell populations within the tumor microenvironment to be immunostimulating [91].

Because permanent expression of a CAR requires the insertion of a promoter and transgene DNA into genomic DNA, a possibility exists, however unlikely, that insertional mutagenesis could occur. However, no serious AEs attributed to insertional oncogenesis in genetically modified T cells have been reported to date [92]. It has been proposed that transfection with messenger RNA coding for a CAR leading to transient expression could pose an even more remote risk of genotoxicity [93–95]. Another area of theoretical concern is that the modified T cells could become capable of autonomous proliferation, independent of binding tumor-associated antigen. Moreover, it is also possible that modified T cells could lead to graft-versus-host disease. Although these events have not been observed in the clinic, careful consideration and examination of each new iteration of these technologies should be undertaken. Finally, the use of other therapies, such as lymphodepleting chemotherapy concomitant with CAR T-cell therapy, can lead to other toxicities. However, these effects can be alleviated using different protocols for dose escalation of chemotherapy or the infusion of CAR T cells over multiple days rather than one.

To date, it has been more difficult to demonstrate CAR T-cell activity in patients with solid tumors. However, numerous CARs have been generated against cell-surface molecules expressed on solid tumors, including HER2, GD2, prostate-specific membrane antigen, and mesothelin, and many of these are currently being studied in clinical trials [96]. Currently, >50 phase I trials are in development for CAR T-cell therapy for indications that included glioma, neuroblastoma, and sarcoma, in addition to hematologic malignancies. It is unclear whether currently available CAR T cells will be as effective against solid tumors as they have been in hematologic malignancies, and it has been suggested that combinations with immune checkpoint inhibitors will be necessary to overcome the immunosuppressive microenvironment of solid tumors [97, 98]. It is clear that additional research is required in this area.

To date, it has been more difficult to demonstrate CAR T-cell activity in patients with solid tumors. However, numerous CARs have been generated against cell-surface molecules expressed on solid tumors, including HER2, GD2, prostate-specific membrane antigen, and mesothelin, and many of these are currently being studied in clinical trials.

Conclusion

The field of cancer immunotherapy has rapidly developed and expanded during the past decade. Several immunotherapies have entered into the clinic and have shown promising results. The fusion of gene therapy, cell therapy, and immunotherapy shows significant promise for the future of treating and potentially curing patients with advanced malignancies.

Author Contributions

Conception/Design: Marcela V. Maus, Bruce L. Levine

Provision of study material or patients: Marcela V. Maus, Bruce L. Levine

Collection and/or assembly of data: Marcela V. Maus, Bruce L. Levine

Data analysis and interpretation: Marcela V. Maus, Bruce L. Levine

Manuscript writing: Marcela V. Maus, Bruce L. Levine

Final approval of manuscript: Marcela V. Maus, Bruce L. Levine

Disclosures

Marcela V. Maus: Inventor on some patents related to CAR T cells (IP), Novartis, Inovio (C/A), Novartis (RF); Bruce L. Levine: Novartis (IP, H, RF).

(C/A) Consulting/advisory relationship; (RF) Research funding; (E) Employment; (ET) Expert testimony; (H) Honoraria received; (OI) Ownership interests; (IP) Intellectual property rights/inventor/patent holder; (SAB) Scientific advisory board