Development of TCR-T cell therapy targeting mismatched HLA-DPB1 for relapsed leukemia after allogeneic transplantation

概要

主論文の要旨

Development of TCR-T cell therapy targeting
mismatched HLA-DPB1 for relapsed leukemia after
allogeneic transplantation
同種移植後再発白血病に対する不適合ヒト組織適合性抗原DPB1を
標的としたT細胞受容体遺伝子改変T細胞の開発

名古屋大学大学院医学系研究科 総合医学専攻
微生物・免疫学講座

分子細胞免疫学分野

(指導：西川 博嘉 教授)
CAROLYNE BARAKAT

【Introduction】
Relapsed hematological malignancies after allogeneic hematopoietic stem cell transplantation
(allo-HSCT) remains a significant challenge, with the re-emergence of the primary disease
being the major cause of death in approximately 50% of cases. Human leukocyte antigen
(HLA) matching between recipient and donor is a critical component of allo-HSCT.
Alloreactive donor immune cells can induce an immune response against mismatched HLA
antigens in recipient tissues, leading to potentially life-threatening graft-versus-host-disease
(GvHD). Therefore, HLA-identical siblings are generally considered optimal options for alloHSCT. However, such donors are not always available, and allele-level HLA-A, -B, -C, -DRB1,
and -DQB1 (10 loci/10 loci)-matched unrelated donors are currently being considered. The
alloreactive T cells derived from the donor graft can, on the other hand, recognize and attack
residual malignant cells in the recipient’s body, causing a graft-vs-leukemia/lymphoma effect
(GvL), which is believed to be a good counterbalance of GvHD and responsible for the lower
rates of relapse observed in patients who receive allo-HSCT.
GvL-mediated adoptive cell therapy is a form of immunotherapy that involves the infusion
of donor-derived immune cells, mainly T cells, into a recipient after allo-HSCT to treat
relapsed or refractory hematological malignancies. Donor T cells can be further genetically
engineered to express T cell receptors (TCRs) or chimeric antigen receptors (CARs) that target
specific tumor antigens to confer the GvL effect. Recent studies have highlighted mismatched
HLA-DPB1 as a therapeutic target. Even in 10 loci/10 loci HLA-matched unrelated allo-HSCT,
more than 70% of the donor-recipient pairs additionally present mismatches in HLA-DPB1
antigens, and certain HLA-DPB1 mismatches have been shown to correlate with a lower risk
of relapse after allo-HSCT. Targeting mismatched HLA-DPB1 antigens carries a low potential
risk of GvHD if performed under appropriate conditions because HLA-DP expression is
restricted to hematopoietic cells and inflamed non-hematopoietic tissues to a lesser extent.
Furthermore, post-allo-HSCT, donor-derived normal hematopoietic cells are also spared
because they are ‘self’ for T cells targeting mismatched recipient-type HLA-DPB1 antigens.
【Research methods】
In this study, we sought to establish T cell clones specific for mismatched HLA-DPB1
antigens using post-HSCT donor-derived CD4+ T cells physiologically primed to mismatched
HLA-DPB1 after transplantation. We obtained blood samples from three patients (Allo-001, 004, and -009) diagnosed with hematological malignancies before (pre-) and after (post-) alloHSCT from HLA-DPB1 mismatched unrelated donors. We optimized the best protocol to
induce mismatched HLA-DPB1 specific T cells clones by comparing the stimulation efficiency
of CD4+ T cells isolated from 30 and 100 days post-allo-HSCT patients’ PBMCs with different
stimulators such as HLA-DPB1 mismatched pre-allo-HSCT (recipient origin) PBMCs, donor
B-lymphoid cell line (LCL) or leukemia cell line K562 transduced with recipient-specific allo-

-1-

HLA-DP α and β cDNA along with CD86. Allo-HLA-DP-specific response was tested at the
single-cell level by intracellular cytokine staining (ICS) for interferon (IFN)-γ. For enrichment
of allo-HLA-DP-specific T cells, an IFN-γ secretion assay followed by MACS sorting was
performed. The sorted cells were then cloned by limiting dilution. Growing clones were tested
for reactivity against non-hematopoietic HeLa cell line transduced with recipient’s HLA-DP α
and β cDNA together with HLA-DM and invariant chain by IFN-γ ELISA. Clones that showed
no reactivity against HeLa transfectants were selected for further detailed analysis.
Putative clones were tested for IFN-γ production against various leukemia cell lines and
primary leukemia blasts by ELISA. The potential cross-reactivity of putative clones against a
panel of U937 Leukemia cells transduced with various HLA-DP alleles commonly found in
the Japanese population was tested by IFN-γ ELISA.
For generation of genetically modified T cells, TCR-α and TCR-β gene sequences from
selected T cell clones were determined by RT-PCR and direct sequencing. The TCRα and
TCRβ genes were cloned into a retroviral vector and used to infect 5-day CD3/CD28-activated
T cells isolated from healthy donors. The resultant TCR-T cells were subjected to functional
and specificity analysis.
【Results and Discussion】
From three patients recruited in this study, we isolated multiple T cell clones specific for
HLA-DPB1*02:01, -B1*04:02, and -B1*09:01. We prioritized the HLA-DPB1*09:01-restricted
clone 2A9 because HLA-DPB1*09:01 is unique and its frequency is relatively high in the
Japanese population.
Detailed analysis confirmed clone’s 2A9 reactivity against 5/5 of leukemia cell lines in
HLA-DPB1*0901 restriction manner. Furthermore, clone 2A9 demonstrated reactivity against
primary myeloid leukemia blasts isolated from patients with leukemia, even those with low
HLA-DP expression, suggesting that clone 2A9 possess a high-affinity TCR and is promising
for clinical application.
TCR-T cells expressing 2A9 TCR retained the ability to trigger HLA-DPB1*09:01restricted recognition and lysis of various leukemia cell lines and showed multifunctionality 
indicated by multiples cytokines production in vitro. Notably, not only CD4+ fractions but also
CD4-negative fractions similarly produced IFN-γ, suggesting CD8+ TCR-T cells expressing
high affinity 2A9 TCR do not require CD4 accessory protein to stably ligate to HLADPB1*09:01-expressing cells.
Currently, analysis of TCR repertoire of other putative clones specific to other HLA-DPB1
alleles than HLA-DPB1*09:01 and preparation of TCR-T cells is ongoing.
【Conclusion】
In this study we proved that induction of mismatched HLA-DPB1-specific T cell lines and

-2-

clones from alloreactive post-allo-HSCT CD4+ T cells was feasible. Furthermore, we
demonstrated that redirection of T cells with a cloned TCR repertoire was readily possible
without increased risk of cross-reactivity. Like other single-antigen targeting therapies (e.g.,
chimeric antigen receptor T cells), one potential limitation of our approach is evasion of
antigen-loss variant tumors at genetic level. However, most relapse in HLA class IImismatched HSCT is reported to be due to epigenetic changes leading to downregulation of
class-II antigens which may be reversible. For future clinical applications, further efforts to
generate TCR panels that cover common HLA-DP alleles, establish in vivo therapeutic models,
and development of treatment to restore HLA class II expression are necessary.

-3-