Finding new lanes: Chimeric antigen receptor (CAR) T‐cells for myeloid leukemia

Abstract

Background

Myeloid leukemia represents a heterogeneous group of cancers of blood and bone marrow which arise from clonal expansion of hematopoietic myeloid lineage cells. Acute myeloid leukemia (AML) has traditionally been treated with multi‐agent chemotherapy, but conventional therapies have not improved the long‐term survival for decades. Chronic myeloid leukemia (CML) is an indolent disease which requires lifelong treatment, is associated with significant side effects, and carries a risk of progression to potentially lethal blast crises.

Recent Findings

Recent advances in molecular biology, virology, and immunology have enabled researchers to grow and modify T lymphocytes ex‐vivo. Chimeric antigen receptor (CAR) T‐cell therapy has been shown to specifically target cells of lymphoid lineage and induce remission in acute lymphoblastic leukemia (ALL) patients. While the success of CAR T‐cells against ALL is considered a defining moment in modern oncology, similar efficacy against myeloid leukemia cells remains elusive. Over the past 10 years, numerous CAR T‐cells have been developed that can target novel myeloid antigens, and many clinical trials are finally starting to yield encouraging results. In this review, we present the recent advances in this field and discuss strategies for future development of myeloid targeting CAR T‐cell therapy.

Conclusions

The field of CAR T‐cell therapy has rapidly evolved over the past few years. It represents a radically new approach towards cancers, and with continued refinement it may become a viable therapeutic option for patients of acute and chronic myeloid leukemia.

1. INTRODUCTION

Acute myeloid leukemia (AML) is a highly lethal disease and requires intensive chemotherapy for treatment, often in combination with hematopoietic stem cell transplantation (HSCT) and molecularly targeted therapies. A combination of two highly cytotoxic agents, cytarabine and daunorubicin, has been the standard of care for AML patients for over four decades. Despite use of highly toxic therapy, the current 5‐year overall survival (OS) of AML patients is less than 30%.1 Poor outcomes of AML patients highlight the fact that our armament against AML is restricted and there is an urgent need to discover newer therapeutic modalities. Advances in immunotherapy during the past two decades have revolutionized the field of anticancer therapy. Expanding understanding of viral vectors, progress in genetic engineering, and improvement in cell manufacturing techniques have led to the invention of T‐cells with novel receptors that can attack any desired cell type. These novel T‐cell receptors (TCR), called chimeric antigen receptors (CAR), are genetically engineered to combine the extracellular antibody binding and intracellular cell signaling properties of T‐cells. This has enabled oncologists to redirect the immense cytotoxic power of T‐lymphocytes towards specific types of cancer cells with remarkable efficiency. The long‐term vision behind this adoptive cell transfer (ACT) technology is to treat cancer patients by modulating their own immune system, thereby avoiding the exposure to highly toxic chemotherapy agents.

The most widely recognized triumph of ACT therapy has been its ability to induce remission in precursor B‐cell ALL patients.2, 3 The success of CAR T‐cells against ALL and subsequent FDA approval of Tisagenlecleucel has reignited the enthusiasm in adapting this therapy for broader antineoplastic application, specifically against myeloid leukemias and solid tumors. However, translation of this technology to develop a robust anti‐AML therapeutic platform has proven to be quite daunting. Despite challenges, numerous groups are evaluating the CAR T‐cell technology to target unique antigens in hopes of designing therapies that are safe and potent against AML.

1.1. Principles of CAR technology

The fundamental idea behind CAR T‐cell therapy is to introduce an artificial gene construct into the genome of T‐lymphocytes which can modify the structure of TCR as desired, thus changing their target antigens. Clinically, this is achieved by first obtaining healthy T‐lymphocytes from cancer patients by leukapheresis, followed by in‐vivo expansion and genetic modification, and finally, infusing the newly created CAR T‐cells back into patients. Once in circulation, chimeric receptors bind to their ligands on tumor cells and trigger a signal transduction cascade that directs the activated T‐cells to kill the target either directly or through release of highly potent cytokines, chemokines and proteases.

TCR is a hetero‐dimer anchored to the membrane of T‐lymphocyte. It is composed of alpha and beta subunits of proteins that work in conjunction with gamma, delta, and epsilon chains and the invariant CD3ζ chain. The structure of the CAR, in analogy with the TCR complex, can be functionally divided into three regions—the extracellular domain, which is responsible for antigen identification and overall affinity; the intracellular domain, which acts as the signal transduction moiety; and the transmembrane region, which connects the two domains. Most designs also include a hinge (spacer) region that assists in more efficient ligand binding. The evolution of CAR T‐cells can be best understood by picturing the sequential addition of intracellular domains that have enhanced their functionality and longevity (Figure 1). The first‐generation CAR T‐cells are composed of an extracellular domain which is usually a single chain variable fragment (scFv), a spacer, and the CD3ζ chain intracellularly.4 While this simplistic structure allowed the CAR T‐cells to recognize tumor antigens as targets, it had low cytokine production ability and lacked in‐vivo efficiency. With better comprehension of the underlying mechanism of TCR complex and costimulatory subunits, the design of CAR T‐cells underwent gradual but significant upgrades. The second generation of CAR T‐cells is characterized by presence of a second signal to stimulate T‐cell activation using CD28 molecule.5 Third‐generation CAR T‐cells have a wider range of signaling units like CD28, 41BB, OX40, DAP10, in conjunction with CD3ζ chain.6, 7 This further enhances the potency by increasing cytokine secretion and in‐vivo persistence. A further improvement in design, called CAR T‐cells redirected for universal cytokine killing (TRUCK), is currently being explored and is considered by some, the fourth‐generation of CAR T‐cells.8 These fourth‐Gen CAR T‐cells contain an inducible cytokine secretion mechanism and find application in the context of solid tumors. They secrete transgenic cytokines (like IL‐12 and IGF‐γ) and induce local cytotoxicity in the tumor stroma. This can potentially increase the anticancer effect against solid tumors because their enormous phenotypic diversity makes antigen‐targeting inherently difficult.

Figure 1.

Advancements in the design of Chimeric antigen receptors

Bispecific CARs (tandem CAR) and Compound CARs (dual CARs) that can identify two different antigens and have higher specificity with fewer side effects are also being developed.9, 10 This strategy reduces the risk of tumor cells becoming refractory due to antigen loss since each CAR T‐cell is targeting two antigens. Another tactic garnering attention is the use of inhibitory CAR (iCAR) that might allow fine‐tuning of cytotoxicity and minimize off‐target effects.11 Combining existing anticancer molecular therapies with CAR T‐cells has also been an active area of research. Check‐point inhibiting agents (using PDL1, CTLA4) and receptor tyrosine kinase inhibitors are examples of some therapeutic agents that can offer synergistic effect and mitigate toxicity when combined with CAR T‐cell therapies.12, 13

1.2. CAR T‐cells against myeloid leukemia

While CAR T‐cell therapy has emerged as a powerful tool in our fight against ALL, the path towards fulfilling its potential as an anti‐AML therapy is fraught with challenges. One major reason for the unprecedented success of CAR T‐cells against ALL is the relatively limited expression of CD19 on leukemia cells and a small subset of lymphoid hematopoietic cells. The ideal antigen to target by CAR technology should have high immunogenicity, should be expressed on all or almost all malignant cells, should not be expressed on noncancerous cells, and should be critical for cell differentiation and proliferation. Due to nonrestrictive expression of most AML antigens, targeting myeloid leukemia cells will also target myeloid cells and marrow precursor cells broadly. Therefore, most anti‐AML CAR T‐cell therapies lead to significant myelotoxicity and prolonged cytopenic phase. This immunosuppression puts the patient at risk of serious bacterial, fungal, and viral infections, and can be life threatening. Secondly, off‐target toxicity by CAR T‐cells has been a key hinderance in their application for treating AML. A study testing CLL1‐CD33 dual targeting CAR T‐cells found that a pediatric patient showed dramatic reduction in leukemia cells but developed prolonged pancytopenia and grade 3 neurotoxicity. Many other clinical trials have also been closed or revised due to significant hepatotoxicity and neurotoxicity reported by enrolled patients. Thirdly, since the effector functions of CAR T‐cell are mediated through strong cytokines and chemokines (IL‐2, IL‐6, C‐reactive proteins, etc.), the release of high amounts of inflammatory mediators frequently causes systemic effects like vasodilation, respiratory distress, hypotension, and end‐organ failure. There have many reports of these dramatic inflammatory reactions, collectively called cytokine release syndrome (CRS), causing significant morbidity and deaths. For instance, FDA had imposed clinical holds on two Phase I studies of CD123 targeting CAR T‐cells following the death of a patient that was attributed to CRS. Despite these obstacles, several CAR T‐cells are in pipeline, and many AML‐specific antigens are being targeted in preclinical and clinical trials, as summarized in Table 1.

Table 1.

Recent developments and clinical trials of myeloid‐targeting CAR T‐cells

Targeted Antigen	Function/Expression	Recent Clinical Development	NCT Number
NKG2D	Expressed on transformed and infected cells	Autologous NKG2D‐DAP10‐CD3ζ CAR T‐cells administered to 12 patients showed sustained hematological response20. Dose‐escalation trial (THINK study) is evaluating safety and clinical activity of NKG2D based CAR T‐cells21. One patient with refractory AML achieved sustained remission with NKG2D CAR T‐cell treatment followed by HSCT22.	NCT02203825, NCT03018405
Lewis Y antigen	Glycosphingo‐lipid expressed on certain tumor cells	LeY CAR T‐cells infusion caused reduction in residual disease in two (out of four) patients in five patients with relapsed AML26.	NCT01716364
CD33	Transmembrane protein expressed in AML cells	CD33‐41BB‐CD3ζ CAR T‐cells resulted in transient response in a patient (NCT01864902)31. Dose escalation study testing CD33‐CAR‐T cells is currently recruiting (NCT03126864).	NCT03126864, NCT01864902, NCT02799680, NCT03473457, NCT02958397
CD123	Transmembrane protein expressed in AML cells, low expression on myeloid cells	CD123CAR‐CD28‐CD3ζ‐EGFRt+ T cells are being tested in AML patients following lymphodepletion43. Anti‐CD123 CAR T‐cells being tested for the treatment of myelodysplastic syndrome39. Compound CAR T‐cells targeting CD33 and CD123 show impressive in‐vitro efficacy against AML44.	NCT02159495, NCT02623582, NCT03672851, NCT03114670, NCT03766126, NCT03190278, NCT03556982, NCT03796390, NCT03473457, NCT03190278
FLT3	Receptor tyrosine kinase expressed on hematopoietic precursor cells and signals	Anti FLT3 CAR T‐cells using 41BB and CD28 co‐stimulating domains are effective against AML cell lines in‐vitro and in xenografted NSG‐SGM3 mice52,53. FLT3 ligand containing CAR T‐cells also demonstrate impressive anti‐AML cytotoxicity in‐vitro and in‐vivo55,56.	Preclinical studies Xenograft studies
Folate receptor	Expressed on numerous carcinomas and healthy tissues including lung tissue	Anti‐FRβ‐CD28‐CD3ζ CAR T‐cells showed activation and cytokine release upon co‐incubation with human AML cell lines70.	Preclinical studies Xenograft studies
CLEC12A	Transmembrane glycoprotein expressed on myeloid lineage cells	CLL1‐41BB‐CD28‐CD3ζ CAR T‐cells are cytotoxic to CLL1+ AML cell lines, patient samples, and in a xenograft model58,59,60.	Preclinical studies Xenograft studies
Others
TIM3	Expressed on T lymphocytes and regulates Th1 mediated cell immunity	Early preclinical studies
CD38	Glycoprotein expressed on surface of immune cells, bone marrow precursor cells and leukemia cells	CD38‐41BB‐CD3ζ CAR T‐cell are effective against AML cell lines expressing high levels of CD3863,64.	NCT03222674, NCT03473457
CD7	Immunoglobulin superfamily protein, NK cell and T cell marker, expressed in AML cells	CAR T‐cells targeting CD7 are protective in mouse xenograft models49.
CD25	Expressed on surface of immune cells, IL2 receptor	Early preclinical studies
CD32	Expressed on surface of B lymphocytes, Fc‐g receptor	Early preclinical studies
CD44	Surface glycoprotein involved in lymphocyte activation, hematopoiesis, and cell‐cell interactions	Early preclinical studies
CD47	Membrane receptor glycoprotein involved in cell‐cell interaction, proliferation, and apoptosis	Early preclinical studies
CD56	Neural cell adhesion molecule expressed on surface of neurons, skeletal muscle cells, natural killer cells, and some types of T‐cells	Multi‐CAR T cells recognizing many AML antigens will be tested in phase‐1 clinical trial.	NCT03222674, NCT03473457
CD117	Receptor tyrosine kinase expressed on the surface of hematopoietic stem cells	Multi‐CAR T cells recognizing many AML antigens will be tested in phase‐1 clinical trial.	NCT03222674, NCT03473457
Muc‐1	Extracellular glycoprotein present on normal cells and prevents infections. Also found on cancerous cells	Multi‐CAR T cells recognizing many AML antigens will be tested in phase‐1 clinical trial.	NCT03222674
IL1RAP	Co‐receptor for the IL1 and IL33 receptors expressed on inflamed and malignant cells	CAR T‐cells targeting IL‐1RAP that can selectively target quiescent CML stem cells are being tested in preclinical studies76.	NCT02842320

1.2.1. NKG2D

Natural Killer Group 2D (NKG2D) is a transmembrane protein expressed on NK cells and CD8+ T cells that trigger cytotoxicity upon recognition of its ligand. NKG2D ligands such as MHC class I polypeptide‐related sequence A and B (MICA/B) and UL16 binding protein (ULBP) 1 to 6 are expressed on the surface of inflamed cells and several types of tumors (like ovarian cancer and AML).14, 15 Studies have found that leukemia cells of about 75% of AML patients express at least one NKG2D ligand at the surface.16 Building on this concept, several different variants of NKG2D‐based CAR have been developed, and in‐vivo mechanistic and cytotoxicity studies have yielded exciting results in multiple preclinical studies as well as clinical trials.17, 18, 19, 20, 21, 22 In one study, a single intravenous administration of low dose autologous NKG2D‐DAP10‐CD3ζ CAR T‐cells (flat dose 3 × 10⁷ cells) without prior chemotherapy was tested in 12 patients with relapsed/refractory AML, myelodysplastic syndrome, or multiple myeloma. There were encouraging signs of activity with one AML patient showing hematologic improvement for 3 months post‐treatment.20 A dose‐escalation trial (THINK study, NCT 03018405) is evaluating safety and clinical activity of NKG2D‐based CAR T‐cells.21 In another clinical trial, one 52‐year‐old male patient with refractory +8/del7(q22q36), FLT3/NPM1 wild‐type AML has successfully achieved sustained remission after receiving allogenic HSCT following NKG2D CAR T‐cell treatment.22 NKG2D directed CAR T‐cells have established a new paradigm where a natural ligand has been employed to redirect T‐cells and achieve antileukemia effect.

1.2.2. Lewis Y antigen

Lewis Y antigen (LeY) is a difucosylated carbohydrate antigen that is overexpressed on hematological cancers like AML as well as some nonhematological malignancies.23, 24 Scientists therefore took advantage of this selective expression and created CAR T‐cells targeting LeY in AML patients.25, 26 LeY CAR T‐cell infusion was tested in five relapsed AML patients in a clinical trial.26 Out of the four patients who were evaluable, all demonstrated persistence of CAR T‐cells and two had reduction in residual disease but eventually all of them relapsed.

1.2.3. CD33

CD33 (sieglec 3) is a transmembrane receptor which is expressed on about 85% of AML cells but is also frequently present on normal myeloid progenitors and myelocytes.27, 28 Due to expression on cells of myeloid lineage, adapting CD33 as a target for anti‐AML therapy has been a long sought‐after goal. Many groups have developed CAR T‐cells targeting CD33 which are currently undergoing clinical trials.29, 30, 31, 32, 33 In one clinical trial, antiCD33‐41BB‐CD3ζ CAR T‐cells were administered to a patient in escalating fractions, resulting in transient response (NCT01864902).31 Another study is currently testing a safe and tolerable dose of CD33 CAR T‐cells on patients with relapsed/refractory AML (NCT03126864). One frequently encountered side effect of this therapy is the myelotoxicity due to shared expression of CD33. Novel ways to circumvent this problem are also being studied, like engineering a resistant population of hematopoietic stem cells (HSC) with reduced or no expression of CD33 in the CAR T cell recipient.34 It is expected that CD33 targeting by CAR T‐cell therapy will remain an exciting area of research in the coming years.

1.2.4. CD123

A transmembrane cytokine receptor, CD123, helps transmit the signal of interleukin‐3 (IL‐3) and plays a central role in immune system.35, 36, 37 It was one of the first antigens that were found to be preferentially expressed in myeloid leukemia cells.38 Multiple centers have created and tested distinct CD123 targeting CAR T‐cells and antibodies with initial results suggesting this to be a promising approach for AML patients.39, 40, 41, 42, 43, 44 T‐cells expressing CD123‐specific CAR and truncated EGFR (antiCD123‐CD28‐CD3ζ‐EGFRt+ T cells) are being tested in AML patients following lymphodepletion.43 Early results from this trial have been encouraging with many patients showing complete and partial responses. Another group is testing anti‐CD123 CAR T‐cells for the treatment of myelodysplastic syndrome and has found encouraging results during in‐vitro and xenograft studies.39 Compound CAR T‐cells possessing discrete domains for targeting CD33 and CD123 simultaneously have also demonstrated impressive in‐vitro cytotoxicity against AML.44 Cumulatively, these results demonstrate that anti‐CD123 targeted CAR T‐cell therapy has been steadily refining and laboratory‐based efforts to increase its efficacy and reduce toxicity could make it clinically viable in time.

1.2.5. Other targets for AML

In addition to the abovementioned targets which are being actively investigated in preclinical and clinical trials, there are several other candidates that can theoretically be adapted for AML CAR T‐cell therapy (Table 1). T‐cell immunoglobulin mucin‐3 (TIM3) is expressed in high level in AML cells of most subtypes (except M3) and has minimal expression on healthy HSC, making it an attractive candidate for CAR T‐cell therapy.45 This concept has been tested using anti‐TIM3 antibody in xenograft models reconstituted with human leukemia stem cells (LSC) or normal HSCs, where it exerted a potent antileukemic effect while sparing normal hematopoiesis.46, 47 CD7 is expressed in up to 30% of AML patients, and it is associated with poorer prognosis and therapy resistant disease.48 CD7‐edited CD7‐targeting CAR T‐cells have demonstrated robust cytotoxic effect against AML and were also found to be protective in mouse xenograft models.49

Fms like tyrosinase kinase‐3 (FLT3, CD135) is a cytokine receptor belonging to the receptor kinase III family. It is expressed on the surface of most myeloid leukemia cells, hematopoietic progenitor cells as well as mature cells of myeloid lineage.50 FLT3 receptor mediated signaling is important for the normal development of HSC and progenitor cells.51 FLT3 receptor is encoded by the FLT3 gene which is altered in up to 30% of AML patients. Anti FLT3‐41BB‐CD3ζ CAR T‐cells showed impressive cytotoxicity against AML cell lines in‐vitro and in xenografted NSG‐SGM3 mice.52 Similar results were also found by another group which developed a second‐generation anti‐FLT3 CAR T‐cells with CD28 as the co‐stimulating domain.53 Crenolanib has been shown to increase surface expression of FLT3 receptors which enhanced the potency of CD4+ and CD8+ CAR T‐cells against FLT3‐ITD+ as well as wild type AML cells.54 CAR T‐cells containing the human FLT3 ligand have also shown impressive anti‐AML cytotoxicity in‐vitro and in‐vivo.55, 56

Human C‐type lectin‐like molecule‐1 (CLL1 or CLEC12A) is a type II transmembrane glycoprotein expressed on myeloid cells and the majority of AML blasts.57 Importantly, CLL1 expression is absent in HSC, which makes it a good therapeutic target for AML treatment. In preclinical studies, lentivirally transduced antiCLL1‐41BB‐CD28‐CD3ζ CAR T‐cells specifically lysed CLL1+ cell lines and primary AML patient samples in vitro.58, 59 Strong antileukemic activity was observed using CLL1 directed CAR T‐cells in a xenograft model of disseminated AML.60 CLL‐1 based ACT, therefore, has the potential to be used against AML either in combination with conventional chemotherapy or as standalone treatment. Similarly, expression of CD38 on a subset of AML cell population and relative absence of healthy mature HSCs has been shown in several studies.61, 62, 63, 64, 65 AntiCD38‐41BB‐CD3ζ CAR T‐cell was found to be cytotoxic against AML cell lines expressing high CD38 levels in a dose‐dependent manner.63, 64 Intriguingly, this group also found that concomitant treatment with ATRA enhanced CD38 expression on AML cells and augmented the antileukemia effect of CAR T‐cells.

The folate receptor family consists of four folate binding proteins (α, β, γ, and δ) and mediate the process of folate uptake by endocytosis. Since folates play a central role in one‐carbon transfer and nucleic acid metabolism, they are integral for malignant tumor cells that have high rate of proliferation. Targeting tumors through folate receptors was first conceptualized over 2 decades ago when anti‐FRα CAR T‐cells for ovarian tumors were developed.66 Folate receptor beta (FRβ) is expressed in about 70% of primary AML patients making it a lucrative target for leukemia CAR T‐cell therapy.67, 68, 69, 70 Preclinical studies using anti‐FRβ‐CD28‐CD3ζ CAR T‐cells demonstrated significant cytokine release upon co‐incubation with human AML cell lines, suggesting T‐cell activation and proliferation.70 Another interesting finding was the augmentation of FRβ expression upon exposure of AML cells to ATRA.

Many studies have found increased expression of CD25, CD32, CD44, CD47, and CD96 on LSC as compared with healthy myeloid cells and HSC.71, 72, 73, 74 In a Japanese study aimed at identifying LSC gene signature in AML patients, 53% participants were found to express either CD32 or CD25 or both.73 Furthermore, these CD32+ or CD25+ cells were found to be necessary for initiating AML and were relatively quiescent and chemotherapy resistant in vivo. Healthy HSCs depleted of CD32+ and CD25+ cells were able to reconstitute long‐term multilineage hematopoietic repertoire in vivo, signifying the potential safety of treatments targeting these molecules. The safety of CD38, CD56, CD117, or Muc‐1 CAR T cells is also currently under investigation in a phase I clinical trial (NCT03222674) that is currently recruiting children and adults with relapsed/refractory AML in China.

1.2.6. Chronic myeloid leukemia

The anticancer potential of CAR T‐cells is not limited to aggressive AML alone. The indolent form of myeloid leukemia, called chronic myeloid leukemia (CML), is one of the most prevalent malignancies of the modern world. Multiple antigens with selective expression on CML cells are known, and new ones are being constantly investigated for CAR T‐cell targeting.75, 76 One such example is IL1RAP, a co‐receptor for the IL1 and IL33 receptors, which is highly up‐regulated on CML cells. In one study, selective killing of CML mononuclear cells was achieved by using anti‐IL1RAP monoclonal antibodies demonstrating the feasibility of targeting this antigen.75 Recently, a French group has developed CAR T‐cells targeting IL‐1RAP and showed that it can selectively target quiescent CML stem cells in preclinical studies.76

This therapy has the potential to offer a permanent cure from CML which is otherwise considered a life‐long pathology. However, researchers have only scratched the surface of the nascent CML CAR T‐cell landscape so far, and it remains to be seen if any of these therapies can become clinically feasible in future. One reason for such slow progress is the fact that the morbidity and mortality from CML has declined substantially since the introduction of tyrosine kinase inhibitors about 20 years ago.

1.3. Future directions

The field of CAR T‐cell based anticancer therapy has seen significant development and refinement during the past decade. Extensive global focus and collaborative research in core fields of basic sciences have led to better molecular characterization of tumors and improvements in techniques of synthetic biology. These gradual advances have slowly but steadily led to a revolution of sorts in adoptive T‐cell based therapies. Years of arduous efforts are finally starting to make ripples across the clinical arena as refractory AML patients now have the option to participate in an increasing number of CAR T‐cell clinical trials. The success of anti‐CD19 CAR T‐cells has generated enormous interest among researchers and fueled the dream to find a similar CAR T‐cell therapy for the deadly AML and the everlasting CML. Nevertheless, numerous challenges need to be overcome before such CAR T‐cell therapies can make the critical leap from bench to bedside.

One of the biggest hurdles is to find the optimum surface antigen which will enable targeting of myeloid leukemia cells and LSCs while sparing noncancerous cells. As molecular and genetic typing of AML patients becomes routine, it is expected that our knowledge of leukemia surface antigens will continue to expand rapidly which may lead to discovery of novel target antigens. Secondly, strategies that can maximize CAR T‐cell potency and minimize the risk of resistance by finding rational combinations with small molecule drugs, like checkpoint inhibitors and tyrosine kinase inhibitors, need to be explored. Thirdly, any significant clinical development rests on our ability to overcome the antigen escape phenomena, due to loss of target antigens on leukemia cells, induced by CAR T‐cell treatments. One way to evade this is by development of bispecific and multispecific CAR T‐cells that can target more than one AML surface antigen and enhance the specificity of T‐cell cytotoxicity. Lastly, a major inadvertent side effect of anti‐AML CAR T‐cell therapy is the myeloablation due to collateral damage to healthy myeloid tissues in AML patients. While HSCT can reconstitute patient's marrow, it is fraught with its own long‐term complications like graft rejection and graft vs host disease (GVHD). As mentioned earlier, inhibition of the target antigens in healthy HSC, which can later be employed to reconstitute the marrow, is an area of ongoing research. An innovative idea proposed to terminate CAR T‐cell cytotoxicity and minimize off‐target toxicity is to employ either kill‐switches or antibody‐mediated depletion.77 While this can certainly allow rapid elimination of CAR T‐cells from the recipient, it is debatable whether it will also limit the anticancer potency of the system. As more patients undergo CAR T‐cell therapy, it will enhance our understanding of these complications and provide insights to many unanswered questions like—can patients be rescued from myeloablated state after receiving CAR therapy without needing HSCT?

It is exciting to see that the number of research groups and academic institutes involved in AML CAR T‐cell therapy is progressively increasing despite the massive challenges. While no single AML antigen has so far been sufficiently developed to be called a real clinical breakthrough, it is expected that immense efforts and determination of scientists will continue to push the boundaries of our current knowledge and capabilities. We sincerely hope that this review would enrich the literature, generate new discussions, and guide those interested in myeloid leukemia directed CAR T‐cell therapies. To conclude, we believe that CAR T‐cell therapy is a revolutionary shift from the conventional anticancer chemotherapy, and with continual modifications it could eventually become a valuable therapeutic approach for countering AML, and possibly CML.

ACKNOWLEDGEMENT/GRANTS/FINANCIAL SUPPORT

This work was supported by Oklahoma Center for the Advancement of Science and Technology grant awarded to J.Z.

CONFLICTS OF INTEREST

Suraj Pratap and Joe Zhao have filed patent application related to CAR technology.

Authors have no other conflicts of interest to declare.

AUTHOR CONTRIBUTIONS

All authors had full access to the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Conceptualization, S.P. and J.Z.; Methodology, S.P.; Formal analysis, S.P.; Resources, S.P. and J.Z.; Writing—Original draft, S.P.; Writing—Review and editing, S.P. and J.Z.; Funding acquisition, J.Z.

Pratap S, Zhao ZJ. Finding new lanes: Chimeric antigen receptor (CAR) T‐cells for myeloid leukemia. Cancer Reports. 2020;e1222. 10.1002/cnr2.1222