Immunotherapy for B-Cell Neoplasms using T Cells expressing Chimeric Antigen Receptors From antigen choice to clinical implementation *

Immunotherapy with T cells expressing chimeric antigen receptors (CAR) is being evaluated as a potential treatment for B-cell neoplasms. In recent clinical trials it has shown promising results. As the number of potential candidate antigens expands, the choice of suitable target antigens becomes more challenging to design studies and to assess optimal efficacy of CAR. Careful evaluation of candidate target antigens is required to ensure that T cells expressing CAR will preferentially kill malignant cells with a minimal toxicity against normal tissues. B cells express specific surface antigens that can theoretically act as targets for CAR design. Although many of these antigens can stimulate effective cellular immune responses in vivo, their implementation in clinical settings remains a challenge. Only targeted B-cell antigens CD19 and CD20 have been tested in clinical trials. This article reviews exploitable B cell surface antigens for CAR design and examines obstacles that could interfere with the identification of potentially useful cellular targets.

overcome these limitations, a variety of approaches are being developed to augment the ways by which antibodies kill malignant cells. 3n interesting approach to enforce tumour antigen recognition is to generate genetically engineered T cells to express highly active chimeric antigen receptors (CAR) to kill leukaemic cells without the requirement for restriction of the major histocompatibility complex (MHC) molecules.These genetically engineered T-cell constructs combine the antibody-derived antigen-binding motif, known as a single-chain fragment of variable regions (scFv) that recognises the target antigen, with the intracellular functional domains of T-cell receptor (TCR)-related signalling molecules, the CD3ζ-chain, and T-cell costimulatory molecules such as CD28, CD134, and CD137. 4,5Using lentiviral transduction or electroporation, the genetic constructs are introduced into the autologous peripheral T cells collected by leukapheresis procedures [Figure 1].7][8] Generally, before an infusion of genetically engineered T cells, patients receive lymphodepleting chemotherapy to eliminate immunosuppressive cells and other lymphoid immune cells that produce growth factors such as inteleukin-7 (IL-7) and IL-15. 9,102][13] It has been reported that the day of adoptive T-cell transfers in relation to preconditioning regimens is critical as T cells given at day two after stem cell transplantation show superior persistence as compared to those given in later days. 14ntil recently, development of this personalised T-cell-based therapy was limited due to the lack of gene transfer techniques and efficient T-cell culture systems.Therefore, previous clinical applications with T cells expressing CAR showed modest clinical activities, as these engineered T cells demonstrated a limited expansion in vivo and rapidly disappeared from circulation. 6,7,15,16With the most recent wave of genomic technology, great progress has been made in improving gene transfer technology and in developing methods to enhance T-cell effector functions.][19][20][21][22] As for any targeted immunotherapy, the success of novel genetically engineered T cells relies heavily on the characteristics of the target cell surface antigen.Since the list of potential candidate antigens is expanding, the choice of a suitable target becomes more crucial for both the design and optimal efficacy of CAR.In addition, a critical evaluation of candidate target antigens is required to ensure that T cells expressing CAR will preferentially kill malignant cells with little toxicity against normal tissues.Therefore, this article will briefly review the B-cell surface antigens, which may be exploited as targets for CAR design to treat B-cell neoplasms.It will also examine potential obstacles that could interfere with the identification of these cellular targets, which might be useful for CAR immunotherapy.

Characteristics of an Ideal Target Antigen
An important step in designing a CAR is the selection of a good candidate antigen.The target antigen should possess certain characteristics to justify scientifically its clinical applicability.First, the suitable antigen should ideally be expressed by malignant cells but not on normal cells.While this seems to be a reasonable criterion for antigen selection, it may not be a prerequisite in every case, especially for B-cell neoplasms where most potential targets are also expressed by some cells of normal tissues.In these cases, the target antigen should be expressed in higher amounts on malignant cells relative to normal cells.Second, the target antigen should be expressed on the surface of the malignant cells.This reasonable cell surface expression is important to avoid immune escape of malignant cells, which are endowed with a strong predilection to down-regulate antigen expression. 23Third, beside its immunogenicity, the target antigen should not be internalised by malignant cells to ensure the induction of immune effector cell functions.Fourth, an ideal antigen should exhibit negligible shedding and only a small amount of secretion with a limited potential for mutation to avoid compromising its surface molecule exclusively expressed by B cells and follicular dendritic cells among cells of the hematopoietic system. 25,26Of interest, its expression is lost during terminal B-cell differentiation into plasma cells.CD19 is involved in growth regulation of B cells through the B-cell antigen receptor.It is also required for the development of marginal zone B cells, germinal centre formation, T-dependent antibody response, and B-cell memory maintenance. 26Compared to other B-cell surface molecules, CD19 is expressed at high levels on essentially all B-lineage leukaemias and lymphomas, making this antigen an excellent tumour-associated antigen for CAR development.

CD20 ANTIGEN
Following the identification of surface immunoglobulin, CD20 was the first cell surface differentiation antigen of human B cells to be identified by a monoclonal antibody.CD20 is a 33-37-kDa integral membrane protein with four transmembrane domains and an extracellular disulfide bond. 27It is expressed on pre-B and mature naïve B cells but not on plasma cells or early pro-B cells.At the primary sequence level, CD20 shows little sequence homology with CD19 but therapeutic utility.Finally, the antigen of interest should preferentially be involved in key functions of the target cell such as growth-regulating signalling, immune recognition, or apoptosis.

Potential Candidate Antigens for CAR Design in B-Cell Neoplasms
B cells express a variety of specific surface antigens that can theoretically act as targets for CAR design.Although many of these antigens showed the ability to stimulate effective cellular immune responses in vivo, their implementation in clinical settings remains a big challenge.The ones with the most promise in B-cell neoplasms are summarised in Figure 2, and their characteristics are briefly described in Table 1.

CD19 ANTIGEN
Because of its promising proprieties in B-cell development, pioneering work in CAR design focused on the CD19 antigen.This molecule was first identified as the B4 antigen on normal and malignant human B lymphocytes. 24This 95-kDa type I transmembrane glycoprotein is a cell both antigens interact with MHC class II molecules on B cells.Current studies suggest that CD20 is a component of a signal transduction complex that is involved in the growth regulation of B lymphocytes following activation. 28In addition to its restricted B-lineage expression, CD20 is present on the majority of B-cell lymphomas, providing a strong rationale for CAR design.

CD22 ANTIGEN
The CD22 antigen was originally identified as a B-cell-associated adhesion molecule that appears to function in the regulation of B-cell activation.It is a member of the sialic-acid binding immunoglobulinlike lectin (SIGLEC) family.SIGLEC proteins are a subset of the immunoglobulin super-family of cell recognition molecules that bind to sialic acidcontaining glycans of the cell surface. 29CD22 is a 135-140-kDa B-cell-specific transmembrane sialoglycoprotein, expressed on all B cells except on memory B cells and plasma cells. 30It regulates multiple B-cell functions including the cellular activation thresholds and survival.In this role, CD22 modulates B-cell dependent immune responses, prevents autoimmunity, and controls the homing of B cells back to the bone marrow.It also modulates T-cell signalling upon binding to its ligand on T cells. 31CD22 is strongly expressed by many B-cell lymphomas and leukaemias, particularly hairy cell leukaemia, where it is universally highly expressed. 32Given that CD22 is found on normal B cells at low levels but not on plasma cells, these  on activated monocytes and eosinophils.The expression of CD30 is highly induced on B cells by mitogen activators, or as a result of viral infection with human T-cell lymphotropic viruses (HTLV), human immunodeficiency virus (HIV), or EBV. 37f interest is the finding that CD30 expression is absent on most cells outside the immune system.Interaction of CD30 with its ligand CD153 induces the recruitment of signalling proteins, which leads to the transduction of several biological signals such as cell death, proliferation, activation, and differentiation depending on the cellular context. 37n addition to the malignant disorders cited above, CD30 is also expressed on 20% of B-cell lymphomas and some multiple myeloma cells.Thus, because of its restricted profile expression in normal conditions, the CD30 antigen appears to be an attractive target for CAR development.

CD40 ANTIGEN
Like CD30, the CD40 antigen belongs to the TNFR superfamily and is a surface receptor of 45-50-kDa, playing a central role in T-cell-dependent B-cell activation and proliferation 36 CD40 is expressed during all stages of B-cell development and differentiation, whereas its ligand (CD40L or CD154) is mainly expressed on activated T cells.In addition to its expression on all antigenpresenting cells such as dendritic cells, follicular dendritic cells, and monocytes, CD40 is also expressed on non-haematopoietic cells including endothelial cells, fibroblasts, and epithelial cells .38In fact, CD40 is involved in the amplification and regulation of inflammatory responses by providing critical signals required for the optimal activation of immune cells.These key signals resulted in optimal antigen presentation, expansion of antigen-specific memory and effector T cells, proliferation of B-cell responses, immunoglobulin class switching, and secretion of cytokines. 39Currently, the CD40-CD40L immunostimulatory effect is used to enhance antigen presentation in gene transfer protocols.High levels of CD40 expression were detected on a wide range of malignant cells including B-cell neoplasms, bladder cells, and carcinoma cell types, among others.It has been reported that in Hodgkin's lymphoma, hairy cell leukaemia, chronic lymphocytic leukaemia, and follicular lymphomas, CD40 activation contributed to tumour survival and resistance to chemotherapy.In HIV-related characteristics make CD22 an attractive target for CAR development.

CD23 ANTIGEN
The CD23 antigen was initially defined as the low-affinity receptor for immunoglobulin (Ig)-E, also known as FcεRII.It is a 45-kDa type II transmembrane glycoprotein expressed on B cells but not on immature bone marrow B lymphocytes.
Its expression was also reported on various cell types including monocytes, eosinophils, and platelets. 33D23 participates in multiple regulatory functions either as a membrane-bound glycoprotein or as a freely soluble protein.Among these functions, CD23 regulates the production of immunoglobulin-E, the antibody that mediates the allergic reactions, and B-cell growth, as underlined by its overexpression and abnormal regulation in two B-cell neoplasms resulting from the clonal expansion of mature B cells (i.e.chronic lymphocytic leukaemia and small cell lymphocytic lymphoma). 34In addition, the Epstein-Barr virus (EBV)-transformed B cells express high levels of CD23 and its expression is used to delineate classical Hodgkin's lymphoma from mediastinal diffuse large lymphoma. 34In combination, these properties support CD23 credibility as an interesting target for CAR design in B-cell neoplasms.

CD30 ANTIGEN
Based on its presence on Reed-Sternberg cells, which are characteristic of Hodgkin's lymphoma, CD30 was identified by means of the monoclonal antibody Ki-1 in 1982.Since then, CD30 has been used as a diagnostic marker for Hodgkin's lymphoma and allowed the description of a new group of malignant disorders named CD30-positive lymphoproliferative disorders. 35The CD30 antigen is a member of the tumour necrosis factor receptor (TNFR) superfamily which includes CD40, CD70, CD95, and CD134, among others. 36CD30 is a type I glycoprotein with a molecular weight of 102-120-kDa.Structurally, an 85-kDa form, which is a product of the proteolytic cleavage of the CD30 antigen, could be detected in the blood of patients with CD30-positive lymphoproliferative disorders or autoimmune diseases. 37Under normal conditions, CD30 expression is detected in activated B, T and natural killer cells.In addition, lower levels of expression were reported lymphomas, CD40 binding triggered the growth of B cells via the activation of the nuclear factorkappa-B (NF-κB) pathway that in turn stimulated HIV replication. 38,39Because of the key role of CD40 in mediating growth of B-cell neoplasms, this antigen could be a promising target for CAR development.

CD52 ANTIGEN
The CD52 antigen, also known as the Cambridge pathology (CAMPATH)-1 antigen, is a surface glycoprotein with a molecular weight of 21-28-kDa.CD52 comprises a short peptide sequence of twelve amino acids and a large complex N-linked oligosaccharide, and is attached to the cellular membrane by a glycosylphosphatidylinositol (GPI) anchor. 40It is abundantly expressed on virtually all B and T lymphocytes and, to a lesser degree, on monocytes and natural killer cells.It is also expresses on a large proportion of lymphoid B-cell malignancies including lymphomas and chronic lymphocytic leukaemia. 41The response of B lymphocytes, following its activation via the CD52 antigen, varies according to both the state of cellular maturation and the way the antigen is presented to cells and results in clonal expansion, differentiation, anergy, or deletion by apoptosis.It has been reported that the cross-linking of CD52 on B-cell lines resulted in profound growth inhibition, corroborating the notion that this antigen might be used as a target for CAR design.

CD70 ANTIGEN
The CD70 antigen is a member of the TNFR superfamily, like the CD30 and CD40 antigens. 36D70 is a type II transmembrane glycoprotein with a molecular weight of 50-kDa.It is transiently expressed by activated T and B lymphocytes, monocytes, and dendritic cells.It is also expressed on stromal thymic cells of the medulla, but not on other normal cells or tissues. 42The interaction of CD70 with its ligand CD27 regulates the expansion and differentiation of memory and effector T cells whereas in B cells, CD70-CD27 co-stimulation promotes B-cell expansion through several signalling pathways including spleen tyrosine kinase, phosphoinositide 3-kinase, and NF-κB. 43f interest, CD70 expression is documented in many different types of lymphomas, leukaemias, and carcinomas.CD70 is abundantly expressed by B-cell derived non-Hodgkin lymphoma including diffuse large B-cell lymphoma, follicular lymphoma, Burkitt's and mantle cell lymphomas, and B-cell lymphocytic leukaemia as well as Waldenström macroglobulinaemia, multiple myeloma, and chronic lymphocytic leukaemia. 42It is also highly includes CD80-, CD86-, B7-H3, B7-H4, the inducible costimulator ligand, and the programmed death-1 and death-2 ligands. 46These molecules are transmembrane, or glycosyl-phosphatidylinositol (GPI)-linked proteins, playing key roles in immune responses by providing co-stimulatory or coinhibitory signals upon binding their receptor.
In addition to its well-known role in regulating T-cell activity, recent studies suggest that CD80 also regulates B-cell functions. 47Cross-linking of CD80 on lymphoma cell lines resulted in growth inhibition and up-regulation of pro-apoptotic molecules. 48CD80 is transiently expressed on the surface of activated B cells but is constitutively expressed on a variety of non-Hodgkin lymphomas, including follicular lymphoma, making it a good target for CAR design.

Immunoglobulin Light Chains
Another potential antigen for CAR design is the kappa (κ) light chain (LC) of the Ig molecules.Ig is formed by two identical heavy chains and two identical LCs, lambda (λ) and κ.B lymphocytes express surface monoclonal Igs with either κ or λ light chains.Since expression of λ/κ LCs is clonally restricted, and B-cell neoplasms such as low-grade non-Hodgkin lymphoma and chronic lymphocytic leukaemia are themselves clonal, the malignant cells in a given patient will express either κ or λ LC. 49 Therefore, targeting the LC expressed by malignant B-cells should spare the normal population of B lymphocytes that express the reciprocal LC.Thus, LC λ or κ may be considered an attractive antigen for CAR development.

Improving the Clinical Applicability of Targeted Antigens for Designing CAR
As discussed above, several candidate antigens can be targeted for designing CAR.These antigens are typically normal cell constituents that are over-expressed in B-cell neoplasms.The potential contribution of each antigen to CAR development and implementation is still debated.For instance, what determines which antigens should be first targeted?An additional question expressed on Reed-Sternberg cells in Hodgkin's lymphoma and other non-lymphoid tumour cells such as nasopharyngeal carcinoma, astrocytoma, glioblastoma, and renal cell carcinoma (RCC) as well as in EBV-and HTLV-associated malignancies. 42iven the differential expression pattern of CD70 in leukaemic cells as compared to normal B cells, this antigen represents a prominent target for CAR development.

CD74 ANTIGEN
The CD74 antigen, also known as an invariant chain (Ii), is a conserved type II membrane glycoprotein with a molecular weight of 31-41-kDa.The Ii on the cell surface is present in four isoforms and controls several aspects of the immune system. 44In addition to its role as a chaperone molecule for MHC class II molecules, CD74 is involved in several signalling pathways.CD74 is fundamental to the correct folding and transport of MHC class II molecules to endosomes and lysosomes as well as in antigen loading.It is directly involved in the maturation of B cells through the NF-κB pathway. 44It also serves as the high-affinity receptor for the proinflammatory cytokine known as the macrophage migration inhibitory factor, which activates innate immune cells such as macrophages and monocytes through kinase signalling pathways. 44Under normal conditions, CD74 is expressed on MHC class II positive cells such as B cells, monocytes, macrophages, thymic epithelium, Langerhans cells, activated T cells and dendritic cells.Under the inflammatory processes, CD74 is also detected in high levels on endothelial and epithelial cells as well as on a variety of malignant cells. 45Its overexpression is observed in several forms of cancer and many B-cell neoplasms, including non-Hodgkin lymphoma, chronic lymphocytic leukaemia, multiple myeloma, and follicular lymphoma.In many of these cancers, increased CD74 expression is associated with tumour progression and poor clinical outcomes. 45Compared to normal tissue, the relatively high expression of CD74 combined with the rapid internalisation of CD74 on ligation makes it an attractive target for CAR design.

CD80 ANTIGEN
Among other potential antigens for CAR design is CD80, a 60-kDa transmembrane glycoprotein.This antigen is a member of the B-family that is which B-cell neoplasms should be studied first?In this regard, defining shared features of target antigens is of paramount importance in designing and implementing CAR in B-cell neoplasms.CAR targeting CD19 (CART19) and CD20 (CART20) antigens were the first molecules to be used in B-cell malignancies.1][52][53][54][55][56] In animal models, CART19 and CART20 also eradicated systemic cell tumours established in severe combined immunodeficient (SCID) mice that were expressing CD19 or CD20. 51,52These preclinical experiments have supported the notion that under appropriate conditions, T cells can be therapeutically redirected to eliminate malignant B cells and have justified the translation of these genetically engineered T cells to clinical settings.8][19][20][21][22] Taken together, CART19-and CART20-based clinical trials are at the advanced planning stage in several centres around the world, and the results of these will be of importance to further establish the clinical efficacy of this therapeutic approach.[59][60] With respect to the therapeutic potential of tested CAR, it is worthwhile to note that suboptimal cytotoxic effects of these genetically engineered T cells vary among targeted antigens.Although several explanations may be proposed, one possible reason is the difference in expression levels of the targeted antigens in malignant and normal B cells.For example, it has been reported that normal B cells express high levels of CD20 when compared to CD22, with copy numbers per cell of approximately 150 x 10 3 and 30 x 10 3 , respectively. 61,62This high antigen intensity of CD20 on B cells may lead to anergy or deletion of engineered T cells, as demonstrated in adoptively transferred antigenspecific T cells in transgenic mice. 63Malignant B cells also showed a substantial variability over time in the expression intensity of different targeted antigens.The expression levels of targeted antigens on B cells may vary according to type of B-cell neoplasms, stage of the disease, and also from patient to patient. 61,62,64Thus, it is of importance that optimisation of CAR design benefits from the quantification of information about expression levels of targeted antigens on malignant B cells for each patient to increase anti-tumour activities.Therefore, the challenge now is to develop methods that can precisely measure the expression levels of targeted antigens on malignant and normal B cells in an accurate and reproducible way.Additionally, these standardised methods should be suited to clinical settings, easy to run, and non-invasive.One such method is flow cytometry, a robust and well-established technique that is regarded as an excellent immunophenotyping tool, as multiple molecules and surface antigens can be detected in individual cells. 65At present, it is not known to what extent heterogeneous antigen expression by individual malignant B cells represents a problem for CAR immunotherapy.However, it is quite likely that malignant B cells with low levels of targeted antigens may escape immunotherapy with engineered T cells.Since the targeted antigens are distinct molecules, the possibility exists for combining CART19 or CART20 with CART22, CART23, CART70, or CARTκ potentially to yield additive or synergistic activities against malignant B cells.In addition, to eliminate completely all malignant B cells, effective immunotherapy with CAR requires targeting surface antigens, which are expressed with great specificity by the earliest leukaemic progenitor.In this regard, engineered T cells should successfully penetrate deeper into the tissue and home to the sites with residual leukaemic B cells.Besides their capacity to proliferate and survive in vivo, the efficacy of engineered T cells to eradicate residual leukaemic B cells may also be dependent on the choice of the epitope targeted by the potential antigens.Clearly, the position of the targeted epitope itself, particularly its position within the antigen, has a major impact on the efficacy of engineered T cells activation independent of the scFv binding efficiency. 66

Long term Potential Toxicities of Targeted B Cell Antigens
In order to establish T cells expressing CAR as an effective immunotherapy for B-cell neoplasms, several issues still need to be overcome.Some of these issues include the long-term potential toxicity to normal cells, long half-life, potential crossreactivity, immuno-toxicity, and genotoxicity of CAR.

TOXICITY TO NORMAL CELLS
An important concern with T cells expressing CAR is the long-term toxicity resulting from the depletion of normal human cells, which are important for homeostatic functions.This toxicity depends mainly on the distribution of the targeted antigens in normal tissues, as recently demonstrated in a clinical trial using T cells expressing CARG250 (CARTG250), a prominent RCC antigen. 67In early studies, monoclonal antibodies against this antigen showed a good safety profile in patients with RCC, but when scFv specificity was tested in CARTG250, severe hepatotoxicity occurred due to the reactivity of engineered T cells against the G250 antigen expressed at very low levels on the biliary epithelium.Similarly, based on the widely used humanised mAb trastuzumab in cancer patients, a third generation of T cells expressing CAR was developed against the tumour antigen ERBB2.The genetically engineered T cells displayed a robust anti-tumour activity in animal models, but strongly reacted with the ERBB2 antigen expressed at low levels on lung epithelial cells, resulting in sudden death of the patient five days after treatment. 68Collectively, these results demonstrate that the safety of targeted antigens with mAbs does not translate into the safety of T cells expressing CAR because these engineered T cells with high-avidity receptors can respond to cells expressing their targeted antigens at levels too low to be detected with conventional means.Although in B cell neoplasms, depletion of normal B cells following an engineered T-cell infusion could be medically managed by Ig transfer and antiviral drugs, the potential risk of a negative impact on other organ functions due to low levels of antigen expression should be carefully evaluated.

LONG HALF-LIFE
The issue of the long half-life of T cells expressing CAR remains a major problem.Unlike mAbs, which have defined half-lives, engineered T cells may potentially survive and function for years.Therefore, the risks versus benefits should be clearly balanced for any given patient and a given CAR design.As a cautionary note, early therapy with CART19 resulted in the death of a patient who was suffering from chronic lymphocytic leukaemia with a cytokine storm, supporting further the direct cytotoxic potency of these engineered T cells. 69

POTENTIAL CROSS-REACTIVITY
The possibility of unwanted cross-reactivity is becoming increasingly important in CAR design.Engineered T cells may react with more than one antigenic determinant if some modifications occur when transfecting these T cells with viral vectors.This cross-reactivity may arise because the targeted antigens share an epitope with other tissue antigens, or because they have an epitope which is structurally similar to one used for designing CAR.It may also result from an epitope or antigen spreading, a process in which engineered T cells may expand their immune responses beyond the primary epitope to target other epitopes on normal cells.While these cross-reactivities are difficult to predict, they may lead to severe toxicity, especially since new targeted antigens are being evaluated with second or third generations of CAR.These latter are endowed with potent cytotoxic activities by killing cells which express just a few antigen molecules.

POTENTIAL IMMUNOTOXICITY
A key issue for the use of engineered T cells is the potential immunotoxicity of CAR.It has recently been reported that immune responses against both viral vector and scFv binding epitopes can be detected in patients receiving engineered T cells. 70,71As a consequence, these immune responses may compromise the persistence of engineered T cells and also limit their anti-tumour activities.It is therefore essential to monitor carefully a host's immune responses against all transgene products, including vectors themselves in order to increase the efficacy of engineered T cells.

POTENTIAL GENOTOXICITY
Insertion of suicide genes into the protein constructs could be envisaged to further limit long-term CAR toxicities.However, introduction of genetic material may lead to genotoxicity; that considered, so far there has been no reported evidence that retrovirally engineered mature T cells induce mutagenesis. 72,73Since their first description, engineered T cells have been used in more than one hundred patients with no malignant transformation reported in any case, contrasting with what was observed with genetically modified haematopoietic stem cells. 74To mitigate CAR toxicities, approaches such as a dose-escalating strategy or splitting T-cell doses over two or more days may be applied, particularly when new targeted antigens are first evaluated. 8,75Another approach is to administer high-dose corticosteroids, which could effectively eliminate engineered T cells as was demonstrated in RCC patients receiving CARTG250. 67

Conclusion
As engineered T-cell designs continue to improve, the selection of therapeutic targeted antigens has become crucial to the development of CAR.Lessons learned so far from pilot clinical trials in patients treated with T cells expressing CAR clearly emphasise the importance of the choice of the right antigen.Intense investigation is underway to discover more candidate antigens not only to expand the repertoire of potential targets for CAR design but also to identify those antigens that will provide selected killing of malignant B cells while sparing normal lymphocytes.A critical issue remains in identifying those targeted antigens that could be specifically expressed on B cells in individual neoplasms but not necessarily be used across all categories of patients with B cell malignancies.
By further understanding the characteristics of potential targeted antigens, it may be possible to identify key epitopes that will help design new CAR with the goal of providing maximal clinical efficacy of engineered T cells.It might also be expected that multiple antigenic epitopes could be combined and targeted to further improve the killing properties of CAR therapy.
Equally important, research should also focus on identifying subsets of patients who are likely to respond to a specific target antigen.This will provide a rational basis for target antigens to be exploited in well-designed multicentre studies to establish engineered T-cell immunotherapy as a viable option in the treatment of B-cell neoplasms.One might also argue that the selection of T-cell populations that could be used for CAR design, conditioning regimens, and infused T-cell doses may be different for each targeted antigen.Thus, it is quite possible that the only improvements in all these aspects will lead to significant progress that will ultimately advance the field of CAR therapy.This will require collaboration across different academic groups representing different disciplines in order to provide faster information that could be quickly translated into the CAR design.In the meantime, it is hoped that ongoing clinical trials will continue to generate increasing enthusiasm for the application of this personalised T-cell-based immunotherapy in B-cell neoplasms.

Figure 1 :
Figure 1: Clinical application of T cells expressing chimeric antigen receptors.Following a leukapheresis procedure, autologous T cells are isolated, transduced with a lentiviral vector and re-infused to patient after lymphodepletion with chemotherapy.

Figure 2 :
Figure 2: B-cell surface antigens that may be exploited as potential targets for chimeric antigen receptor design.

Table 1 :
Characteristic features of potential target B-cell antigens for chimeric antigen receptor design.