CAR‐T Cells: Future Perspectives

Chimeric antigen receptor (CAR) T cell therapy has cured cancer in some patients for whom chemotherapy had failed. The development of CAR-T cells has been a decades-long journey from when the technology was first proposed in the late 1980s to the Food and Drug Administration (FDA) approval of Novartis's tisagenlecleucel in 2017, heralding the emergence of a multibillion dollar industry. The first CAR was reported in 1993, when Eshhar et al successfully combined the cytotoxic potential of a T cell with the specific targeting of an antibody in a single gene transfer.1 Their first-generation CAR linked an antibody single chain variable region with either the FcRγ or CD3ζ signaling domains. While T cells transduced with this receptor demonstrated antigen-mediated activation, the earliest clinical trials with first-generation CAR-T cells in solid cancers were disappointing. Optimization of the basic construct over time has crucially included the addition of costimulatory domains leading to improved T cell activation and survival, and the identification of favorable target antigens, most prominently CD19.2 These developments have led to positive clinical outcomes, including reports of cure in patients with B cell malignancies. The scope to further optimize CAR-T cell design and delivery raises the hope of a cure for many more people with malignancies, and heralds an exciting new era in cancer treatment. While researchers, physicians, patients, and investors alike are understandably seduced by such potential, there are still several major hurdles to overcome. For the vast majority of patients with blood cancer, and all with solid cancers, CAR-T cells are not yet proven to be effective, are too toxic, or are not available due to expense or geography. This perspective article aims to give an overview of where we are right now, and consider the issues that must be addressed as the field moves forward, in order to fulfill the promise of recent successes. CAR technology is a biologically and economically powerful tool. With power comes responsibility and the European Hematology Association (EHA), along with its sibling organizations has a responsibility to realize its potential through supporting scientific progress, cooperative working across national borders and medical specialties, clinical trials, education, training and patient advocacy in government and with leaders in the biotechnology industry. In 2019, HemaSphere, the journal of the EHA, will publish a series of review articles discussing the possibilities, current problems, and future developments of CAR-T cell therapy. By the time this article is published, the EHA in conjunction with the European Society for Blood and Marrow Transplantation will have made another important step in this direction by delivering the first European CAR-T Cell Meeting in Paris. Replacing chemotherapy in the treatment of B cell malignancies will be high on the agenda, but there are even greater rewards to be gained if we are ambitious in addressing the challenges of harnessing CAR-T cells for other malignancies and ensuring global access based on need rather than ability to pay. The current situation The majority of clinical trials using CAR-T cells are early phase studies in B cell malignancies. Trial activity increased dramatically in 2016 and continues at a rate of nearly 100 new trial registrations each year. The most common target is CD19, mostly alone, but increasingly in combination with other antigen targets (Fig. 1). While there is significant international co-operation, trials are overwhelmingly centered in the Northern hemisphere, with the vast majority registered in the United States and China (Fig. 2). In the United States alone, well over 1000 patients have now received CAR-T cells, and several studies have opened looking at the long-term effects in responders.Figure 1: An analysis of CAR-T trials registered at clinicaltrials.gov. (A) Distribution of trials by condition targeted. (Note that the majority of trials target more than 1 condition; therefore, the sum does not add up to total number of trials.) (B) Distribution of trials by target antigen. Note that 12 of the trials registered do not target a specific antigen, but manufacture a personalized CAR depending on the disease phenotype. (C) The number of trials by phase and year first posted. ∗2018 to 7th December. CAR = chimeric antigen receptor.Figure 2: The geographical distribution of chimeric antigen receptor-T trials registered at clinicaltrials.gov. Map created with mapchart.net ©.Data continue to accumulate supporting the efficacy and durability of responses to anti-CD19 CAR-T cell therapy in B cell malignancies. Response rates in acute lymphoblastic leukemia (ALL) are reported between 68% and 93%, in chronic lymphocytic leukemia between 57% and 71%, and in B cell lymphoma between 64% and 86%.3 Longer-term follow-up data from the ZUMA1 study indicate durable responses for the minority of patients who achieved a complete remission.4 Future work to better identify predictors of response will improve the risk–benefit balance and minimize unnecessary financial outlay for individual patients and healthcare systems. This could eventually result in those predicted to respond poorly to chemotherapy but well to CAR-T cells receiving them upfront. Despite promising response rates in trials, applying this data to real-world patients is currently very challenging, partly as inclusion criteria favor better prognosis groups. Owing in part to concerns about increased rates of cytokine release syndrome (CRS) in more unwell patients with more inflammatory physiologies, combined with the risks of disease progression during the 2 to 3 weeks required for product manufacture, trials have largely excluded patients with rapidly progressive or symptomatic disease, those with poor performance status, raised inflammatory markers, fevers, cytopenias, or derangements in renal and hepatic biochemistry. Less than half of a historical cohort of chemorefractory patients with diffuse large B cell lymphoma (DLBCL) studied at 1 center would have fulfilled eligibility criteria for the ZUMA1 study. Those eligible patients had a significantly better survival with standard of care immunochemotherapy than the ineligible ones.5 This skewed trial subject population has created a false denominator issue, potentially overstating the efficacy and downplaying the toxicity of current products. As such, the relative magnitude of the benefit of CAR-T remains undetermined even in CD19 expressing malignancies. With future improvements in manufacture and product turnaround time, as well as refinement of CAR-T design and management of CRS, clinical trials should be better able to include more representative subjects. Randomized studies, of which so far only 1 has opened, will better establish the place for CAR-T cells in relation to existing potentially curative therapies in B cell malignancies. The situation for other hematological malignancies and solid tumors remains a long way behind, although recent advances in myeloma and an increasing range of potential antigen targets in solid cancers have led to a large number of trials opening to address this gap. Future perspectives For a variety of well-described reasons, CD19 expressing blood cancers appear most conducive to CAR-T cell therapy. High levels of tumor expression of the target antigen, ease of physical access to tumor cells through the blood and lymphatics, and the tolerability of the on-target off-tumor effect of B cell aplasia make CD19 a unique target. However, <5% of all new cancer diagnoses are CD19 expressing malignancies targetable by licensed products. The innovative strategies developed in CD19 expressing diseases to abrogate antigen-negative relapse, improve efficacy of tumor killing, improve CAR-T cell persistence, and increase control of activity and toxicity, are in parallel being pursued in efforts to bring CAR-T cell therapies to bear against other diseases. We will now look at the technical, logistic and economic challenges, ongoing innovations, and future perspectives on the journey to harness CAR-T cells against other hematological and solid malignancies. Some of the future challenges and possible solutions discussed in this section are outlined in Table 1.Table 1: Challenges in the Future Development of CAR-T CellsMyeloma Targeting the B cell maturation antigen (BCMA) in patients with myeloma is an area of great interest owing to promising early data. BCMA is expressed in nearly all cases of myeloma, and is not present on hematopoietic stem cells or nonhematological cells. Among small numbers of very heavily pretreated patients, overall response rates over 80% are being consistently reported in abstract submissions. The first published trial data of patients receiving anti-BCMA CAR-T cells reported 16 subjects with a median of 9.5 prior lines of therapy, of whom 13 responded, 11 with at least a very good partial response, with an overall median event-free survival of 31 weeks.6 Numbers remain too small and the follow-up too short to yet know how impactful this approach will be, but a strong proof of principle has emerged. As the technology is refined and delivered earlier in the therapeutic journey, we may see deep and durable responses that change the standard of care. Other antigen targets are also being explored, including CD19, CD38, CD138, SLAMF7, and kappa light chains in kappa-restricted myeloma (thus sparing nonmalignant lambda-expressing plasma cells). CD138 as a target has established proof of principle through the clinical efficacy of the naked monoclonal antibody Daratumumab, but thus far progress with anti-CD138 CAR-T cells has been hampered by concerns of on-target off-tumor activity, with CD138 being expressed on erythrocytes, salivary tissue, liver, and skin. Owing to the clonal heterogeneity of myeloma, and shifting clonal tides in response to chemotherapy, it seems likely that combinatorial antigen targeting will be particularly important. Mikkilineni and Kochenderfer reviewed the progress of CAR-T cell therapy in myeloma in Blood last year.7 As there remain no curative chemotherapy options in myeloma despite recent progress, there is significant potential for CAR-T cells to disrupt the treatment landscape. Conversely, innovations in noncellular immunotherapies, such as antibody-drug conjugates and bispecific antibody therapies, seem set to provide stiff competition and may prove to be far less expensive. Myeloid malignancies The curative potential of allogeneic hematopoietic stem cell transplantation in acute myeloid leukemia (AML) and myelodysplasia has long-established proof of concept for T cell-mediated immunotherapy in these diseases, but there are several major issues in extending this concept to the application of CAR-T cell therapy. Significant interpatient and intrapatient genetic and phenotypic heterogeneity means that there is no single AML-specific antigen to target. Furthermore, expression of AML antigens on normal healthy myeloid precursor cells risks prolonged and potentially fatal myeloablation following CAR-T cell activation. Expression of myeloid antigens on nonhematopoietic tissues risks additional on-target off-tumor effects. For example, CD33 is expressed on hepatic Kupffer cells, and hepatoxicity is a recognized side effect with the CD33 targeting antibody-drug conjugate gemtuzumab ozogomycin. The potential for long-term persistence of anti-CD33-directed CAR-T cells amplifies this risk. There are several proposed strategies to overcome these obstacles, including (1) identification of AML-specific antigen pairings required for initiation and maintenance of leukemogenesis that can be exploited by combinatorial antigen targeting; (2) early termination of CAR-T cell activity once remission is achieved with suicide constructs or transient CAR expression techniques; and (3) myeloablative CAR-T cell therapy followed by rescue allogeneic hematopoietic stem cell transplantation. Candidate target antigens explored thus far in preclinical models include CD33, CD123, Lewis-Y, CD44v6, FLT3 receptor, CLL-1, and the folate receptor β. Several of these are being taken forward in early phase clinical trials, from which published data so far is sparse. For further insights on this field, the reader is directed toward Sarah Tasian's recent review.8 T cell malignancies Developing effective CAR-T cells against T cell malignancies, for which chemotherapy is rarely curative, will be a huge challenge. Particular, unique obstacles include contamination of the autologous CAR-T cell product by malignant T cells carrying the CAR, as well as unwanted CAR-T directed death of fellow CAR-T cells (fratricide) and of healthy T cells owing to shared target antigen. This fascinating set of challenges is reviewed by Marion Alcantara alongside Carl June and others in an elegantly succinct article.9 Among the proposed solutions are alternative cellular vehicles for the CAR, of which CAR-NK cells appear to be gaining most traction in early phase clinical trials against both T cell malignancies and other cancers. Further data are keenly awaited from this promising emerging field. Solid tumors While immune checkpoint inhibitors have established proof of concept concerning activated T cell efficacy against solid cancers, outcomes from treating solid cancers with first-, second-, and third-generation CAR-T cell products targeting single antigens have been very disappointing. In a field with very few curative options for metastatic disease, there is clearly massive unmet clinical need, but sadly fewer patients have benefitted from CAR-T cells than review articles have been written on the subject. Among these, Long et al have produced an authoritative article outlining the particular obstacles to progress.10 There is a dizzying array of early phase CAR-T cell studies in solid cancers, targeting a wide variety of antigens. This reflects both the biological diversity of this field as well as the lack of persuasive data supporting any one CAR-T cell product against any particular solid tumor or tumor antigen thus far. There are some common themes that appear responsible for these disappointing results compared with blood cancers: (1) CAR-T cells face difficulty gaining access to target cells sitting within poorly vascularized tumor masses, walled-off to a certain degree by nonmalignant inflammatory cells and connective tissues. (2) On gaining access, CAR-T cells face a and tumor potential (3) For those CAR-T cells that do to a solid tumor and cytotoxic potential, there is a further issue, a lack of only is there a lack of expressed tumor antigens, but also a lack of with of antigens on tumor and healthy tissues. to CAR-T cell design the potential to and an and efforts are to increase the expression of antigens using or therapy with the of CAR-T may yet prove that the most and effective for CAR-T cells in solid tumors is not as a but in or a chemotherapy, or remission in disease with a high of does not appear as we are on the of clinical benefit with CAR-T cell therapies in solid tumors by the in blood cancers, and it is to be a of the right target and it in to a third-generation CAR-T will be With improved treatment for solid cancers an area of such vast unmet need, and a potentially for the current toward these issues should some CAR-T cells There have been a variety of innovations in the design of CAR-T cells, in an to improve efficacy and toxicity in hematological malignancies, and to the challenges of solid cancers. 3 of the of CAR with increasing of additional to This is discussed in more in a recent review article published in this Chimeric antigen receptor (CAR) developments in CAR (A) The CAR is made up of an or and signaling or more costimulatory domains. The and domains can be made from a variety of constructs as each with and costimulatory for which there are also several have 3 of leading to improvements in CAR-T persistence, and Further developments to either improve efficacy further or to (B) to improve efficacy include of antigens, for which work through of more than 1 antigen leading to CAR activation, and which of a common such as to which domains can be Other are at the tumor such as and have the ability to to the tumor through of either or and may be further to only in these to minimize on-target off-tumor effects. (C) Other to toxicity include further developments in or and to on-target off-tumor effects. with or suicide for of the CAR should toxicity prove despite medical therapy. = approach interest is with a to increase a variety of tumor and antigen-negative This has led to the emergence of T cells, based on the and of off-tumor effects can be by both antigens to be present for CAR activation, or which will in the of one antigen only if the other is not Targeting one antigen another can and antigen-negative relapse, and can be achieved by either 2 of CAR-T cells, 2 the or by the These are being in clinical trials, and phase 1 data are to emergence on the most common combination of CD19 and The potential for further has been with the development of or These the to target new antigens as and combined with new for may give the ability to of malignancies. a variety of to an This cytokine by cell for cytokine of which the studied on target antigen, so shifting the tumor in favor of and tumor cell Further developments have the tumor to be to and CAR-T cells, with the addition of to or to and in the of the immune potential of CAR-T cells the of This has been addressed with the addition of suicide which can be if required for of the CAR-T cell This principle has been with where CAR expression is on ongoing with the potential to control of CAR-T cell activation, to response and The be to the normal of the immune with CAR-T cells which respond to malignant disease, during and and on of the target antigen at disease As well as in progress in gene the the CAR to be at a specific genetic both increasing efficacy and concerns with to This technology also for to be such as or target antigens in T cell some of the discussed in a of the CAR an With so many parallel developments CAR-T it will some time to the to optimize both efficacy and The of continue to change as progress in this area and we more about the of the we have CAR = chimeric antigen CAR-T cell field will not in is likely that tumor responses will be achieved and by disease or combination of CAR-T cells with other such as small and chemotherapy, and a of potential There are several trials which do not a specific CAR but will design a CAR against a number of possible antigen targets on the of the is that studies will be required to identify the way to this array of treatment options in the era of personalized CAR-T manufacture In parallel with developments in CAR design and ongoing advances in the manufacture will be to and manufacture at time, and in this should access to patients with more disease to those far from current of manufacture, and to those in less CAR-T cell manufacture using on such as the T cell may significantly CAR-T cell In an cells be to a from which of the CAR could be with a product emerging a few step CAR-T cell could in with a T and the In addition to the challenges of this would major and control but is a fascinating approach to addressing this is the of allogeneic CAR-T cells. gene to minimize the risks of or can be rapidly and are also available to those to or of autologous cells. The scope for products be and efficacy and are far from but the of potential and the ability to on a large eventually the huge financial to existing autologous products. With 2 products available for licensed in and there is a in delivering CAR-T cells. is to that this financial is to current concerning this treatment. have in the order of of on the that CAR-T cell products will be but this should not from an of the on which we treatment this treatment does up to its possibilities, a will be required concerning access to curative either being a for the or a right for those in This is also a of great huge financial could give the scientific huge to realize the potential of this emerging field, and forward in basic cancer and While the balance of and access is not a new in the of financial and the potential of the therapeutic by CAR-T cells for and patient advocacy to work with biotechnology researchers, and to a CAR-T cell that is as and effective as 3 of there is now a degree of about the of CAR-T cell therapy. The of change in this field challenges for and in the right CAR-T products for the right patients at the right time, potentially in the right combination with other existing and emerging responses in some patients with chemorefractory B cell malignancies are but are no for long-term follow-up data from with standard of care in representative patient In this the of CAR-T therapies remains even in CD19 expressing blood cancers and place in the treatment of other is and and access to the 2 products are huge competition and of technology and should to more although this has not to be the even with The ability to and the immune with increasing levels of to provide therapy is far that which can be achieved with treatment this great now is to realize this hope for the benefit of