Keywords : CAR-T cell; iPSC; T cell; iPSC-derived T cell; tumor
cell; therapeutic; off-the-shelf
Running title: Application of iPSCs for T cell and CAR-T cell
Generation
Introduction :
Adoptive cell therapy (ACT) is a new form of cancer treatment that
empowers T cells through genetic engineering to express
either T-cell receptors or chimeric antigen receptors (CARs) to
recognize and destroy the malignant cells (Z. Wang & Cao, 2020). ACT
manipulates and alters the capacity and anti-tumor response of
physiological T cells for a robust immune response against infected or
abnormal cells (Rohaan, Wilgenhof, & Haanen, 2019). ACT using CD19
CAR-T cells has led to outstanding responses in patients with certain
types of hematologic malignancies, such as large B-cell lymphoma and
B-cell acute lymphoblastic leukemia (B-ALL) (Milone & Bhoj, 2018). In
recent years, CAR-based therapy has demonstrated remarkable results
against tumor cells. Over the last decade, FDA approved three types of
CAR-T cells, namely Tisagenlecleucel (Kymriah), Axicabtagene ciloleucel
(Yescarta), Brexucabtagene autoleucel (Tecartus), for some patients with
relapsed or refractory leukemia and for patients with some kinds
of lymphomas (Maude et al., 2018; Neelapu et al., 2017; M. Wang et al.,
2020). Technically, a patient’s T cells are genetically engineered to
express artificial receptors, CARs, to redirect T cells specifically
toward the antigen of interest on cancer cells (Srivastava & Riddell,
2018). A CAR is a synthetic construct composed of an antigen recognition
domain derived from a monoclonal antibody, a hinge region, a
transmembrane domain, a co-stimulatory domain(s), and a TCR activation
domain (Figure 1A ). CAR gives T lymphocytes supraphysiologic
properties to identify and attack the preferred target cells in a quick
and robust manner (Sadeqi Nezhad et al., 2020; Yang, Jacoby, & Fry,
2015). When CAR interacts with the cognate tumor antigen, the activation
signal transmits to T cells and the activated T cells recognize the
target cells independently from HLA or MHC expression. This unique
feature makes CAR-T cells a versatile therapeutic option for different
patients and opens up a new avenue for adoptive immunotherapy (Abreu,
Fonseca, Gonçalves, & Moreira, 2020). In contrast, the TCR is an α/β
heterodimer that recognizes a short peptide from the MHC products. Each
α/b subunit contains variable and constant region domains followed by a
transmembrane region. TCR associates with a six-subunit complex called
CD3 dimers, such as CD3εγ, CD3εδ,
and CD3ζζ, to initiate an intracellular signaling (Figure 1B )
(Birnbaum et al., 2014; Rudolph, Stanfield, & Wilson, 2006).
Nevertheless, most of the CAR-based therapies lie within autologous
CAR-T therapy, and very few allogeneic CAR-T treatments resulted in as
promising as autologous CAR-T cells. The therapeutic efficacy of CAR-T
cells may encounter critical problems because the major T-cell source is
derived from patient’s own peripheral blood mononuclear cells (PBMCs).
These may affect the therapeutic outcomes due to some pivotal issues,
including difficulty in harvesting an adequate number of T cells,
time-consuming, and ineffective to treat patients with progressive and
lethal diseases in due time (X. Wang & I. Rivière, 2016). Current CAR-T
manufacturing platforms are costly and time-consuming. A typical
autologous CAR-T cell therapy includes three main phases: (a)
leukapheresis (cell isolation), (b) manufacturing, and (c) preparation
and administration. These processes, from isolation to administration,
takes between 2-3 weeks, depending on manufacturing methods. Thus, this
required time could be too long for patients at their late stage of
diseases. These hurdles in autologous CAR-T cells forced the researchers
to turn to the allogeneic source of T lymphocytes, T cells derived from
third-party donors. However, allogeneic CAR-T cells have confronted
challenges, such as alloreactivity or graft-versus-host disease (GVHD)
and human leukocyte antigen (HLA) mismatches in recipient patients
(Townsend, Bennion, Robison, & O’Neill, 2020). Gene-editing technology,
clustered regularly interspaced short palindromic repeats (CRISPR)/
CRISPR-associated protein 9 (Cas9) system has been introduced to remove
such barriers; but, more investigation is required to have a clear view
about genome editing technology in CAR-T cells. Allogeneic CAR-T cells,
such as CRISPR-edited allogeneic CAR-T cells, are not a promising
alternative option to autologous CAR-T cells, despite showing some
remarkable results (C. Li, Mei, & Hu, 2020; J. Liu, Zhou, Zhang, &
Zhao, 2019). Therefore, a new T-cell source is needed to address the
challenges in autologous and allogeneic CAR-T cells. Herein, induced
pluripotent stem cells (iPSCs) seem to have a potential capacity to be a
novel T-cell source for CAR-T manufacturing.
This article will discuss the
applicability of iPSCs in CAR-T cells and consider the challenges,
manufacturing process, and perspective of CAR iPSC-derived T cells.
iPSC Overview
Embryonic stem cells (ESCs) and human-induced pluripotent stem cells
(hiPSCs) are pluripotent stem cells (PSCs). They are similar in terms of
surface marker expression, self-renewal, proliferation, feeder
dependence, morphology, and in vivo teratoma formation (K.
Takahashi et al., 2007; Yu et al., 2007). However, subtle differences in
epigenetic profiling and transcriptomic were identified (K. Kim et al.,
2011; Parrotta et al., 2017). PSCs generate a multitude of
differentiated cells that are needed for clinical purposes and offer an
alternative source of cells to regenerative medicine (Wattanapanitch,
2019). ESCs are derived from the inner cell mass of the blastocyst. In
contrast, iPSCs are originated from different somatic cell types, such
as skin, cord blood cells, bone marrow cells, urine, and peripheral
blood cells (G. Liu, David, Trawczynski, & Fessler, 2020). In the
beginning, somatic cells were introduced with Yamanaka factors, such as
Oct3/4, Sox2, Klf4, and c-Myc, maintained a pluripotency status. Two of
these factors, c-Myc and Klf4, showed a tumorigenic potential, which
increased the risk of tumor formation upon the activation or
overexpression, and consequently, they were replaced with Nanog and
Lin28 (K. Takahashi et al., 2007; Kazutoshi Takahashi & Yamanaka, 2006;
Yu et al., 2007). Oct3/4 is a transcription factor whose expression
controls the fate of primitive inner cell mass and governs the
maintenance and regaining of stem cell pluripotency (Shi & Jin, 2010).
Oct3/4 is further regulated by Sox2 that governs the pluripotency, while
C-Myc functions on histone acetylation for chromatin decompensation,
allowing Sox2 and Oct3/4 to access their genome loci (D. Kim et al.,
2009; Scheper & Copray, 2009). Klf4 activates Sox2 and mediates an
anti-apoptotic response for iPSCs self-renewal (Nandan & Yang, 2009;
Niwa, Ogawa, Shimosato, & Adachi, 2009). Lin28, a highly conserved
RNA-binding protein, performs as a stem cell pluripotency factor via
controlling self-renewal and regulating mRNA translation, especially
let-7 miRNAs (Shyh-Chang & Daley, 2013). Finally, Nanog arranges the
transcriptional network with Oct3/4 and Sox2 and plays as an important
factor for the maintenance of the undifferentiated state of pluripotent
cells (Rodda et al., 2005).
Currently, somatic tissues undergo a reprogramming process using
pluripotency-associated transcription factors, such as Oct3/4, Sox2,
Nanog, and Lin28 (Wu, Zhang, Mishra, Tardif, & Hornsby, 2010). The
reprogrammed cells are known as human iPSCs (hiPSCs), and they can be
re-introduced into patients without causing severe immune rejections (X.
Liu, Li, Fu, & Xu, 2017; Yamanaka, 2020). hiPSCs attracted tremendous
attention in therapeutic purposes, including cancer treatment and
regenerative medicine, providing the opportunities to study the stem
cell properties and the embryonic development process (Y. Li, Hermanson,
Moriarity, & Kaufman, 2018; M. Themeli et al., 2013). hiPSCs pave the
way for clinical application through studying the disease biology, drug
screening, disease models, toxicity tests, and therapeutic purposes
(Fernandez, de Souza Fernandez, & Mencalha, 2013). iPSC lines should
meet some inclusion criteria set by the International Stem Cell Banking
Initiative for banking purposes (Crook, Hei, & Stacey, 2010). The
suggested characteristics and features for iPSC lines include but not
limited to: (i) pluripotency attitudes including positive for renewal or
pluripotent markers such as Nanog, Oct3/4, TRA-1-60, TRA-1-8; (ii)
detecting embryonic-like morphology; (iii) differentiation capacity both
teratoma formation (in vivo) and embryoid body formation (in vitro)
models; (iv) transgene-silencing following reprogramming; (v) assessment
of chromosomal abnormalities by karyotyping; (vi) identity testing by
short tandem repeat-PCR and DNA fingerprinting; and (vii) biological or
microbial contamination essay to ensure the sterility of culture (Crook
et al., 2010; Huang et al., 2019).
iPSC-derived T cells Manufacturing and CAR iPSC-derived T
cells Generation
The generation of CAR iPSC-derived T cells define into two main steps:
(a) iPSCs generation (Figure 2 ) and (b) introduction of the CAR
gene into iPSC-derived T cells. Each of which has been discussed
thoroughly in the following sections.
Principle of iPSC Generation
Somatic cell isolation
iPSCs are produced from most somatic cells and differentiate into most
target cells. Peripheral blood T lymphocytes are the possible choice of
somatic cells for CAR-iPSC generation (Vizcardo et al., 2013). For this
reason, the general platform starts from obtaining T cells either from
autologous or allogeneic sources by leukapheresis procedures. The
processed cells are then washed to discard anticoagulants or other
contaminated cells. T cells are enriched for a specific subset of cells,
such as CD3+, CD4+,
CD8+, CD25+, and
CD62L+ T cells. Finally,
a specific T cell population can
either be used for the next step or cryopreserved for future
investigation (Tumaini et al., 2013; Xiuyan Wang & Isabelle Rivière,
2016a).
Reprogramming Methods
T cell of choice is ready to
reprogram through the introduction of defined transcription factors
either Oct4, Sox2, Nanog, and Lin28 or Oct4, Sox2, Klf4, and c-Myc
(Nishimura et al., 2013). Generally, the generation of hiPSCs relies on
two gene delivery methods: (a)
integrative and (b)
non-integrative approaches. Integrating techniques includes
γ-retroviruses and lentiviruses. Although integrative gene delivery is
an efficient approach and has a stable transgene expression, it
integrates into the genome of host randomly and leads to mutagenesis and
tumor formation (Ben-David & Benvenisty, 2011; Blelloch, Venere, Yen,
& Ramalho-Santos, 2007). Hence, constructive techniques are required to
address problems in integrative gene delivery, and meticulous
consideration is needed when it comes to clinical studies. In contrast,
the non-integrative strategies constitute of liposomal magnetofection,
episomal vectors, adenoviral vectors,
Sendai virus vectors,
adeno-associated viral vectors,
plasmid transfection, synthetic
messenger RNA, minicircle vectors, and
transposon vectors (Ban et al.,
2011; Miyoshi et al., 2011; Narsinh et al., 2011; Sohn et al., 2013;
Stadtfeld, Nagaya, Utikal, Weir, & Hochedlinger, 2008; Warren et al.,
2010; Yusa, Rad, Takeda, & Bradley, 2009; H. Zhou et al., 2009; W. Zhou
& Freed, 2009). These delivery methods mitigate the possibility of
insertional mutagenesis and transgene occurrence due to minimal DNA
alterations and produce a safe iPSC line (Haridhasapavalan et al.,
2019). The non-integrating methods are more applicable for clinical use
than that of integrating strategies.
The hiPSC Characterization
The generated hiPSCs are
characterized by a set of markers that are exclusively related to their
pluripotency attitudes. These markers include stage-specific-embryonic
antigens (SSEA3 and SSEA4), tumor rejection antigens (Tra-1-60,
Tra-1-81), and the transcription factors (Oct3/4, Sox2, and Nanog).
Characterization can be done based on the surface markers (Tra-1-60,
Tra-1-81, SSEA3, and SSEA4) and the intracellular markers (Oct3/4, Sox2,
and Nanog) using flow cytometry assay. In addition, Immunohistochemistry
is another laboratory method useful for detecting both the surface and
intracellular markers in hiPSC cell lines (Abujarour et al., 2013;
Bharathan et al., 2017; Paik, O’Neil, Ng, Sun, & Rubin, 2018). Alkaline
phosphatase staining is also conducted as an early marker for hiPSC
establishment (Singh et al., 2012).
In vitro and vivo differentiation assay
The transduced T cells are cultured in an appropriate media for in vitro
and vivo differentiation assay. From in vitro perspective, the iPSC
colonies are cut mechanically and culture in suspension to form embryoid
bodies and develop the three germ layers known as endoderm, ectoderm,
and mesoderm. Each of these layers is positive for a specific marker,
including beta-III tubulin for ectoderm, alpha-fetoprotein or SOX17 for
endoderm, and smooth muscle actin for mesoderm (Kumazaki, Kurata,
Matsuo, Mitsui, & Takahashi, 2011; Secher et al., 2017). In vivo assay
evaluates the teratoma formation of hiPSCs in an immunocompromised mouse
whether the undifferentiated hiPSCs are capable of forming a teratoma
(Ito et al., 2019; Xiang et al., 2019).
Genetic analysis
Genetically unstable hiPSCs increase the risk of cancer or
tumorigenicity and decrease the potential therapeutic use (Ben-David &
Benvenisty, 2011). Genome stability is characterized by different
techniques, including karyotype, fingerprinting, whole-genome analysis,
array CGH, single nucleotide polymorphism, cancer predisposition
testing, residual vector, and HLA typing (Elliott, Elliott, &
Kammesheidt, 2010; Jang et al., 2019; Nazareth et al., 2013; Popp et
al., 2018; Quinlan et al., 2011; Ramos-Mejía et al., 2012). Karyotype
and fingerprinting techniques have been used widely in the genetic
integrity of hiPSCs (Taapken et al., 2011). Giemsa-banding (G-banding)
is a technique that produces a visible karyotype and used for chromosome
counts to detect aneuploidies and karyotypically abnormal hiPSCs
(Dekel-Naftali et al., 2012). Besides, the fingerprinting technique
analyses short tandem repeats (STR) loci to authenticate the hiPSCs and
help to recognize unwanted switch or cross-contamination (Sarafian,
Morato-Marques, Borsoi, & Pereira, 2018).
Sterility Testing
Cell culture undergoes sterility testing for possible contamination by
bacteria, endotoxins, mycoplasma, and viruses. The most common method
for bacterial or mycoplasma detection is a PCR-based technique and LAL
(limulous amoebocyte lysate) assay for endotoxin verification (Sullivan
et al., 2018).
T cell differentiation from iPSCs
Human T lymphocyte development from hiPSCs is the key step to acquire
the desired number and high-quality T cells required for clinical
application. HiPSCs are manipulated by introducing different cytokines
or chemokines to form mesoderm from which hemogenic endothelium
generates a transient specialized endothelial precursor of all
hematopoietic progenitors in the human embryo (Farkas, Simara, Rehakova,
Veverkova, & Koutna, 2020; Julien, El Omar, & Tavian, 2016).
The in vitro derivation of this specialized endothelium provides
hematopoietic stem and progenitor
cells (HSPC) that has the capability to create T cell lineage
(Figure 3 ). Eventually, the
CD8αβ+/CD4+ double-positive cells
are produced and then differentiate into either CD8+or CD4+ T lymphocytes (Maeda et al., 2016; Minagawa et
al., 2018; Maria Themeli et al., 2013).
Principle of off-the-shelf CAR iPSC-derived T cellGeneration
The developed T cells from iPSCs (iPSC-derived T cells) are enriched for
a specific cell population, such as CD8+ T cells. The
pure iPSC-derived T cells are then activated through the interaction
between their TCR or co-stimulatory receptors and T cell stimulation
factors such as mAbs, interleukins (IL-2, IL-7, 1L-15), anti-CD3/CD28
antibody-coated magnetic beads, soluble CD3 antibody, artificial
antigen-presenting cells (K562 cell lines), plate-bound antibody, and
adhesion molecules (CD2) (Cheadle et al., 2012; Xiuyan Wang & Isabelle
Rivière, 2016b). Next, there are two choices to generate off-the-shelf
CAR iPSC-derived T cells. First, the CAR transgene and the CRISPR/Cas9
system introduce into activated iPSC-derived T cells separately. In this
stage, the CAR-transgene is delivered into activated iPSC-derived T
cells through either viral (lentiviral or retroviral) transduction or
non-viral (electroporation of naked DNA and plasmid-based transposon/
transposase) methods. Afterward, the CRISPR/Cas9 system is introduced
into CAR modified-T cells through one of the following ways. The use of
plasmid DNA encoding the Cas9 protein and sgRNA from the same vector.
The second format is to deliver the mixture of the Cas9 mRNA and the
sgRNA. The last format is the use of Cas9 protein and sgRNA complex,
known as ribonucleoprotein, which showed to be beneficial compared to
the other two systems (Luther, Lee, Nagaraj, Scaletti, & Rotello,
2018). CAR iPSC-derived T cells are expanded by different commercial
procedures using supplemental factors and cytokine supplementation (Ou
et al., 2019). Finally, CAR iPSC-derived T cells are ready to be
introduced into the recipient patient through IV injection or
intratumoral administration (Figure 4 ).
Generation of T lymphocytes from iPSC Technology: A New T-cell
Source
In contrast to the remarkable success in fighting B cell malignancies,
the improvement of immunotherapy towards T cell malignancies is far from
a promising therapeutic approach. T cell malignancies are a
heterogeneous group of diseases that affect the clonal growth of T cells
at various stages of development, resulting in either T-cell leukemias
or T-cell lymphomas (Palomero & Ferrando, 2017). T cell malignancies
are the most common form of hematologic disorders seen in pediatric
patients. T cell acute lymphoblastic leukemia (T-ALL) accounts for
nearly 12%-15% of childhood ALL cases (Raetz & Teachey, 2016). One of
the pivotal problems in patients with T cell malignancies is the low
number of normal T cells and proliferative exhaustion, which impede the
adoptive T cell immunotherapy from a successful treatment. Besides,
almost all the current adoptive T cell immunotherapies are conducted in
autologous-based T cells, which are costly, laborious, time-consuming,
and depend on the count and quality of the patient’s T cells (Iyer,
Bowles, Kim, & Dulgar-Tulloch, 2018; Papathanasiou et al., 2020).
Having an alternative source of T cells with optimized physiological
features would greatly improve the scope of immunotherapy. To this aim,
T lymphocytes can be generated from induced pluripotent stem cells
(Nagano et al., 2020).
In 2010, three research groups reported the
reprogramming of PBMCs with four
transcription factors to produce induced pluripotent stem cells (Loh et
al., 2010; Seki et al., 2010; Staerk et al., 2010). Staerk et al.used FUW-M2rtTA and the individual doxycycline-inducible lentiviruses
encoding Oct4, Sox2, c-Myc, and
Klf4 to reprogram PBMCs. This system led them to an unsuccessful
reprogramming, which might be due to poor penetration of the plasmid in
PBMCs. To address this problem, they exerted doxycycline-inducible
lentivirus encoding all four factors Oct4, Klf4, Sox2, and c-Myc from a
polycistronic expression cassette. This system resulted in higher
penetration efficacy and led to the generation of
iPSC-derived PBMC colonies. One of
the significant findings is that IL-7 plays an important role in
reprogramming efficiency and cell expansion in cell culture, compared to
IL-3, IL-6, G-CSF, and GM-CSF. iPSC-derived PBMCs demonstrated the
pluripotency markers, such as Oct4, Nanog, and Tra1-8, and displayed a
normal karyotype. In vitro differentiation analysis revealed that
iPSC-derived PBMC lines showed all three lineage markers, such as
mesodermal, endodermal, and ectodermal. These molecular and
morphological characteristics of reprogrammed PBMCs are indicative of
pluripotency. Furthermore, TCR gene rearrangement detection was
performed in iPSC-derived PBMCs to determine the origin of iPSC as a
mature T cell. Despite negative results of
TCR gene rearrangement assays in
most PBMC-derived iPSC lines, one iPSC line was positive for TCR gene
rearrangement, indicating the possibility of iPSC generation from human
terminally differentiated circulating T cells (Staerk et al., 2010).
Subsequently, in the same year, two other groups reported the generation
of human iPSCs from peripheral blood mononuclear cells (Loh et al.,
2010; Seki et al., 2010). Although both studies used the same
transcription factors (Oct3/4, Sox2, Klf4, and c-Myc) to generate
iPSC-derived PBMCs, the vectors exerted in these studies were different.Seki et al. used a temperature-sensitive mutated Sendai virus
(SeV), a non-integrating vector. This vector only replicates in the
cytoplasm of target cells and does not integrate into the host genome.
SeV showed that even a low sample size, 1 ml, is sufficient to infect
the cell of interest. It indicated a rapid removal of residual viral
genomic RNA from infected cells, signifying a promising choice to
mitigate tumor formation risk associated with inactivation of tumor
suppressor genes or oncogene (Seki et al., 2010). In contrast, Loh
et al. introduced the four transcription factors with integrating
lentiviruses and found a satisfactory result. Lentiviral vectors are
integrating vectors, meaning that viral DNA translocates into the
nucleus and integrates into the host genome. This process of integration
requires more time to produce the interest results, compared to
non-integrating vectors (Loh et al., 2010). More importantly, both
studies demonstrated the possibility of hiPSC generation from peripheral
blood cells. Of note, iPSC-derived PBMCs had human embryonic stem cells
(ESCs)-like characteristics, including morphology,
pluripotency-associated transcription factors, DNA methylation,
differentiation potential, expression of surface antigens, and detection
of all three embryonic germ layers (Loh et al., 2010; Seki et al.,
2010). Likewise, Netsrithong et al . have generated two iPSC
lines, MUSIi011-A and MUSIi011-B, from peripheral blood T lymphocytes
(Netsrithong et al., 2019). Further success can be seen in reprogramming
melanoma tumor-infiltrating lymphocytes (TILs) containing
patient-specific tumor-reactive repertoire. iPSC-derived TILs
demonstrated all the features of embryonic stem cells and had the
capability to differentiate in vitro and in vivo. Interestingly, this
study showed that the SeV vector is able to generate iPSCs from
terminally differentiated and exhausted TILs expressing a high level of
inhibitory receptors, such as PD-1. However, a low reprogramming
efficiency in TILs was seen, which might be due to the exhausted or
differentiation status of the cells. These iPSCs can produce many
phenotypically defined, expandable, specific, and functional polyclonal
T cells with the same specific-TCR for cancer immunotherapy (Saito et
al., 2016). In consistent with these studies, Nagano et al.further corroborated that the development of T cells from iPSC
technology is feasible and can be applied for large-scale T cell
generation (Nagano et al., 2020).
CTLs detect antigenic peptides presented by MHC class I expressed by
infected or malignant cells. These cells, by their direct contact or in
a TCR-dependent manner, attack the target cells and provide a safe
harbor to other cells. Interestingly, CTLs could also destroy tumor
cells through an MHC-independent manner attacking autologous tumor cells
without affecting other cells within the area (Kelly et al., 2002; Liem
et al., 2019). Cancer immunotherapy using transfer of TCR genes (Johnson
et al., 2009; Morgan et al., 2006) or adoptive transfer of CTLs (Chapuis
et al., 2013; Rosenberg et al., 2011) has been facing an important
problem in harvesting enough antigen-specific CTLs for therapy.Maeda et al . used iPSC technology to clone and expand CTLs in an
adequate number for therapeutic purposes. They harvested WT1-specific
CTLs from healthy volunteers. WT1 antigen is expressed by acute myeloid
leukemia (AML) cells. For the establishment of iPSC lines, SeV vectors
encoding Yamanaka factors and SV40 were introduced to the WT1-specific
CTLs. Despite a successful transfection and establishment of three
WT1-iPSC lines with normal karyotype and ESCs-like characteristics, this
study shed light on other important features of CTLs. When iPSCs are
generated from WT1-specific CTLs, they inherited rearranged TCR
genes-specific for WT1, highlighting the same expression of WT1-specific
TCRs in iPSCs. It is crucial to determine whether the
WT1-CTL-iPSC is a CD8αβ
heterodimer or CD8αα homodimer because a normal functional CTL is a
CD8αβ heterodimer (CD4- CD8α+CD8β+) (Figure 1C ). One of the key steps to
have a regenerated CD8αβ T cell is to use double-positive
CD4+/CD8+ T cells and remove
double-negative T cells in mixed cultures because double-negative cells
kill CD4+/CD8+ T cells. Moreover,
WT1-CTL-iPSCs demonstrated satisfactory outcomes regarding safety issues
and anti-tumor activities in mice with leukemia (Maeda et al., 2016).
iPSC-derived T cells store a pre-rearranged TCR gene in the genome that
stemmed from antigen-specific CD8+ T cells.Kaneko et al. revealed the role of RAG1 (recombination activating
gene1) and RAG2 in both TCRα and TCRβ rearrangement during thymopoiesis.
The additional TCRα rearrangement at CD4/CD8 double-positive stage
during iPSCs differentiation to T cells is associated with
antigen-specificity loss. This extra TCRα rearrangement may lead to an
off-target effect and reduce the stability of the same TCR as it was
present in the original T cell clone. They used the CRISPR/Cas9 system
to knock-out RAG2 during the iPSCs differentiation to CD8αβ T cells.
This strategy led them to maintain the antigen-specificity of TCRs with
the same avidity and affinity in iPSC-derived CD8αβ T cells. Meanwhile,
the anti-tumor activity of RAG2–/– iPSC-derived
CD8αβ T cells was examined both in vitro and in vivoconditions, which successfully decreased the tumor growth and increased
the survival of treated NSG mice. This study provided a new approach to
acquire a great number of TCR-specific T cells from iPSCs with the same
potent anti-tumor activity, which would be beneficial for efficient and
safe immunotherapy (Table 1) (Minagawa et al., 2018).
Generation of CAR-T cells from iPSC Technology
In this decade, CAR-T cell therapy has been increasingly carried out in
clinical settings worldwide. The data from Clinicaltrails.gov
demonstrated that over 650 CAR T-related studies were initiated around
the world.
The T cells currently used for CAR-T development are predominantly
derived from patient’s PBMCs, rarely from umbilical cord blood and less
from third-party healthy donors (Depil, Duchateau, Grupp, Mufti, &
Poirot, 2020). It is well-defined that the current manufacturing process
of CAR-T cells is laborious and intensive, and it requires a carefully
selecting donors with adequate autologous T cells. This labor-intensive
generation of T-cell lines from the patients consumes time and may not
be applied to those who have insufficient T cells. These include
HIV-infected or T cell leukemia patients and leukemia patients with
relapse after allogeneic hematopoietic cell transplantation and those
the use of allogeneic donor lymphocyte infusions is challenging (Deol &
Lum, 2010; Matutes, 2007; Perdomo-Celis, Taborda, & Rugeles, 2019;
Yanada et al., 2020). Having a rapid access to unlimited T-cell sources
with optimized physiological features would remarkably improve the
success of genetically engineered-T cells. In this case, genetic
engineering of iPSC-derived T cells with CARs would be a promising
strategy to harvest iPSC-derived T cells unlimitedly and genetically
engineered them to generate phenotypically defined, expandable, and
functional T cells for therapeutic purposes. Themeli et al. have
isolated PBMCs from a healthy donor and transduced them by two
retroviral vectors, each encoding two of the reprogramming factors
(Oct3, Sox2, Klf4, and c-Myc). The transduced T cells under standard
laboratory conditions and care resulted in T-iPSC colonies, and
subsequently, were introduced with a lentiviral vector encoding CD19-28z
CAR. The transduced iPSCs are then differentiated into T cells that
express both the CAR and an endogenous TCR. Findings revealed a
successful generation of CD19-T-iPSC lines for targeting CD19 antigen in
a xenograft mouse model of Raji human CD19+ Burkitt lymphoma cells.
Laboratory analysis showed that CD19 CAR-iPSC-derived T cells had a low
expression of exhaustion markers, such as PD1, CTLA-4, and LAG3.In vivo cytotoxic assay displayed a potent anti-tumor activity
where the tumor growth was delayed and led to complete tumor regression.
Of note, this study generated iPSC-derived T cells with endogenous TCR,
which signifies the possibility of alloreactivity involving transplant
rejection and GVHD. Phenotypic profiling indicated that the generated T
cells from iPSCs are CD8α+CD8β-,
meaning that these lymphoid cells possibly originated from a fetal
cell–like hematopoietic stem cell intermediate committed to innate-like
lymphopoiesis rather than a CD8αβ heterodimer (CD4-CD8α+ CD8β+) T cell (M. Themeli et
al., 2013). This study was the first to report the genetic manipulation
of human iPSC-derived T cells with artificial receptors, CARs, with
therapeutic potential.
iPSC-derived T cell on Its Way to Off-the-shelf Product:
Challenges and Solutions
It is well-established that autologous CAR-T cell therapy is more
efficient, persistent, and safe with no allogeneic reaction for cancer
patients (Brentjens et al., 2011; Neelapu et al., 2017; Ritchie et al.,
2013). However, autologous CAR-T cell therapy may not be as effective as
it should be in some critical conditions, and it poses complexity and
difficulties for researchers and clinicians. The concerns that
autologous CAR-T cells could bring are included but not limited to the
following limitations. (I) CAR-T cell manufacturing takes nearly 2-3
weeks; during this period, patients may diagnose with rapid progression
of disease requiring immediate attention, and the possibility of
laboratory fiasco in the manufacturing process may prevent the patients
from receiving an appropriate dose of CAR-T cells (Maude et al., 2018;
Schuster et al., 2017). (II) CAR-T cells from autologous source are
produced in a single batch with limited quantity, meaning that redosing
of CAR-T cells may not be applicable for some patients (Rouce et al.,
2015). (III) collecting T cells from cancer patients with a prior line
of treatments increase the chance of isolating dysfunctional T cells due
to disease burden or previous use of heavy medications, such as
chemotherapy drugs (Azzaoui et al., 2016; Das, Vernau, Grupp, &
Barrett, 2019). (IV) leukapheresis products from cancer patients,
especially leukemia patients, may contain aberrant T cells, and CAR gene
may be unintentionally introduced into malignant T cells to express CAR
protein. This phenomenon leads to resistance through masking of the CD19
epitope (Ruella et al., 2018). (V) CAR-T cell efficacy is associated
with functional capacity of autologous T cells. It is essential to
determine the cell subset, whether it is effector memory T cells or
short-living effector T cells, or central memory T cells (Garfall et
al., 2019; Gattinoni et al., 2011). Although autologous CAR-T cells have
revolutionized the therapeutic landscape in oncology, the use of
allogeneic CAR-T cells from third-party donors has several distinct
advantages over autologous approaches (Hu et al., 2019). These
superiorities include pre-prepared CAR-T cell products for emergency
use, more high-quality products, applicable to develop two different
CAR-T cells for combination therapy, suitable for redosing purposes, and
reduce the cost of the manufacturing process. However, allogeneic CAR-T
cells are not deprived of problems. They may cause life-threatening GVHD
and may be rapidly eliminated by the host immune system, reducing the
anti-tumor activity and therapeutic outcomes (Kochenderfer et al.,
2013).
iPSC lines provide an unlimited source of T cells for therapeutic
purposes, but these iPSC-derived T cells are not considered
off-the-shelf products. They cannot be applied for third-party patients
due to the presence of endogenous TCR or human leukocyte antigen (HLA)
mismatch (M. Themeli et al., 2013). The most common TCR-associated
adverse effect is GVHD, in which the donor T cells identify the host
antigens as foreign, and subsequently, destroy them. GVHD risk depends
on several factors. The most important one is the HLAs, which are highly
polymorphic and variable (Taylor, Peacock, Chaudhry, Bradley, & Bolton,
2012). Among, HLA class I (HLA-A, -B, and -C) and HLA class II (HLA-DP,
-DQ, and -DR) are the most important molecules involving in immune
rejection. The former molecule is expressed by all nucleated cells and
present the processed antigens inside the infected cells to
CD8+ cytotoxic T cells. In contrast, the latter
present antigens outside the cells and is expressed by
antigen-presenting cells, such as dendritic cells and macrophages, to
activate CD4+ T cells for a further adaptive immune
response (Wieczorek et al., 2017). Xu et al. developed
immuno-compatible iPSCs from third-party donors using the genome editing
technology, CRISPR/Cas9. First, they converted HLA heterogenous iPSCs
into HLA pseudo-homozygous iPSCs, by knocking out a single allele of
HLA-A (A1) and HLA-B (B7) simultaneously. The HLA-AB-iPSCs were co-cultured with CD8+ T cells. They
remarkably did not stimulate CD8+ T cell proliferation
and could escape from T cell cytolytic activity, meaning no HLA mismatch
between CD8+ T cells and HLA-AB-iPSCs. This finding strongly suggested that homozygous iPSCs can be
generated from HLA heterozygous healthy individuals (third-party
donors), which would provide an unlimited source of HLA homozygous iPSCs
for therapeutic application, especially for CAR-T cell therapy. In
addition to HLA-A/B knockout, HLA-C was further removed in iPSCs,
resulting in triple knockout iPSCs (HLA-A-,
B-, and C-).
HLA-ABC- iPSCs could escape from T cell cytolytic
activity. But, the strategy of HLA-C depletion is not interesting and
cannot be used for a large population and may raise the concern of NK
cell response due to lack of MHC-I on iPSCs; therefore, the HLA-C7 was
retained in HLA-AB- iPSCs. Interestingly, the
HLA-AB-C7+ iPSCs demonstrated
outstanding results both in vitro and vivo , where they
evade the NK cell activity and survived against both T and NK cells
cytotoxic activities, respectively. These data proposed that HLA-C is a
pivotal factor for suppressing or evading the NK cell activity. Besides,
HLA-AB-C7+ iPSCs were further
manipulated through depletion of MHC class II trans-activator gene
encoding HLA-II because CD4+ T cells recognize HLA-II
and secrete cytokines to accumulate a wide range of immune cells. This
provides a better donor-host matching for therapeutic and clinical
applications (Xu et al., 2019). When iPSCs are considered for T cell
generation as an unlimited source for off-the-shelf CAR-T cell
development, endogenous TCR expression and HLA mismatch should be
addressed. This study clearly showed the possibility and feasibility of
genetic manipulation in iPSC lines; therefore, depletion of endogenous
TCR and HLAs in iPSCs are crucial for the development of off-the-shelf
iPSC-derived T cells.
Clinical Translation of Human CAR iPSC-derived T cells: Future
Perspective
Over the last two decades, iPSCs and CAR-T cells have opened up a new
avenue for regenerative medicine and cell-based therapy, respectively.
Since their discoveries, a wide range of protocols and studies have been
developed. Both iPSC-derived cells and CAR-T cells have been used in a
multitude of small and large pre-clinical and clinical trials to treat a
variety of diseases. The use of iPSC-derived T cells for CAR-T cell
generation has just begun and is in a preliminary stage of development.
Pre-clinical studies showed the feasibility and safety of iPSC-derived T
cells and displayed them as a novel T-cell source for CAR-T cell
development. The combination of these new technologies, CAR-T cell and
iPSC, has not been investigated in clinical trials; therefore, the
clinical use of CAR iPSC-derived T cells requires thorough
consideration. This section discussed the hurdles and perspective to
clinical translation of CAR iPSC-derived T cells, including major
clinical requirements, product manufacturing, quality standard,
off-the-shelf product, timelines, and side effects.
First, it is important to develop a controlled, robust, and reproducible
manufacturing platform adherent to Good Manufacturing Practice (GMP).
More Pre-clinical studies, possibly under Good Laboratory Practice
(GLP), is required to implement a defined cultured system and equipment
because undefined media or materials bring a risk of contamination and
may alter the phenotype or function of the cells. Pre-clinical studies
provide an opportunity to assess the off-target effects, feasibility,
product safety, and the potential in vivo challenges. Due to the
complexity of iPSC and CAR-T cell generation, studies need to comply
Current GMP (cGMP) regulations to provide sufficient evidence of safety,
identity, potency, and purity of the products. It is well-established
that the identity of iPSC and CAR-T cell products is commonly
characterized by defined markers and gene expression. Therefore, having
a predefined protocol to evaluate or analyze the features of iPSC and
CAR-T cell products would enhance the success of the manufacturing
process and therapeutic outcomes. For example, the most important
post-treatment safety concern for iPSCs is the formation of either
malignant teratoma or benign teratoma (Cunningham, Ulbright, Pera, &
Looijenga, 2012). Thus, it is crucial to ensure that the infused
products have a low risk of tumorigenicity.
More importantly, by the advent of iPSC-derived T cells as a new T-cell
source, several challenges need to be pre-clinically addressed before
embarking on multicentral clinical trials. First, the physiological
function and phenotypic maturity of iPSC-derived T cells have to be
characterized, as well as the anti-tumor activity or cytokine secretion
compared to physiological T lymphocytes. Despite an established
manufacturing protocol for clinical scale production, further assessment
concerning the potential risk of tumorigenesis and differentiation
capacity need to be considered (Nianias & Themeli, 2019). Furthermore,
there may be genetic and epigenetic variations among different iPSC and
CAR T products. The key factors, including DNA methylation and histone
modification, may alter the phenotype of products and may lead to
mitochondrial mutations, chromosomal aberrations, epigenetic variance,
and genetic diversification. Hence, culture conditions and manufacturing
strategies must be optimized to mitigate such phenomena (Duncan,
Gluckman, & Dearden, 2014; Keller et al., 2018; Yadav et al., 2020).
Breakthroughs in genome editing technology paved the way to generate CAR
and off-the-shelf iPSC-derived T cells with a robust anti-tumor
response, limited off-target effect, more remarkable persistence, better
homing or trafficking, and more compatible to histocompatibility
barriers (M. Themeli et al., 2013; Xu et al., 2019). Encouragingly,
iPSC-derived T cells have the potential to become off-the-shelf T cells
using the CRISPR/Cas9 system. The allogeneic iPSC-derived T cells
undergo a genome editing process to be deprived of TCR and HLAs. The
resulted T cells would be negative for TCR and HLA proteins and could be
used for different third-party patients (Xu et al., 2019). However, this
approach has not been investigated thoroughly and is in early stages;
therefore, for clinical purposes, several pivotal issues, such as
off-target effect, feasibility, safety, and manufacturing approach, have
to be investigated.