Exploring T cell mediated immunotherapy against therapy resistant leukemic stem cells

Abstract

The cancer stem cell (CSC) model suggests that cancers depend on continuous replenishment from rare and distinct CSCs, but their existence has been challenging to prove for many malignancies. The CSCs in myeloid leukemias, known as leukemic stem cells (LSCs) have been well characterised and shown to selectively escape treatment, thus representing the cellular source of relapse. Relapse following initial periods of clinical remission represents the most significant threat to leukemic patients as the available treatment options at this stage are sparse due to the development of resistance. There is therefore a need for strategies facilitating earlier detection of an impending relapse to initiate preemptive treatments early, as well as for new targeted therapies to eliminate LSCs.

The cellular identity of LSCs was recently assigned to the hematopoietic stem cell (HSC) compartment in patients with low to intermediate risk myelodysplastic syndromes (MDS), for which the only curative treatment option is allogeneic stem cell transplantation (allo-HSCT). Detailed characterisation of LSCs in terms of phenotype, function, and treatment resistance is thus now possible, with the aim to ultimately identify novel targets allowing for targeted elimination of LSCs. In study I, we performed LSC-directed screening for measurable residual disease (MRD) in patients with MDS after allo-HSCT. By flow cytometric purification of the hematopoietic stem and progenitor cells, the MRD-sensitivity was enhanced 97- fold compared to conventional screening methods on unfractionated bone marrow cells. Consequently, in our patient cohort, targeted screening of the leukemia initiating cells led to the detection of impending relapses on average ten months before clinical diagnosis. In the clinic, this would allow for prescribing preventative treatments earlier while disease burden remains low which could improve outcome.

A potentially effective cancer treatment option is adoptive T cell therapy, using T cells that carry cancer-specific T cell receptors (TCR). TCR T cells can, unlike chimeric antigen receptor (CAR) T cells, recognise intracellular antigens but despite the theoretical large range of targetable antigens, very few are immunogenic. Consequently, no TCR T cell therapy is yet clinically approved for treatment against hematological malignancies, highlighting the need for new potential targets. In study II we showed that terminal deoxynucleotidyl transferase (TdT), which is normally transiently expressed during early B and T cell development and overexpressed in >80% of patients with B and T acute lymphoblastic leukemia (ALL), is a promising target for TCR based immunotherapy in context of HLA-A*02:01 (HLA-A2). T cells targeting TdT, identified through the blood of healthy donors, showed very efficient elimination of TdT+ leukemic cells, while sparing healthy hematopoietic stem and progenitor cells and mature lymphocytes in clinically relevant mouse models. Thus, TdT TCR T cells are a novel and promising immunotherapy option for patients with B- and T-ALL.

Myeloid leukemias are, to a greater extent than the lymphoid leukemias, characterised by somatic recurrent mutations which generate cancer-specific neoantigens representing a group of unique attractive therapeutic targets. In study III, an HLA-A2 restricted TCR with high specificity against a recurrent neoantigen generated from the FLT3-D835Y mutation in acute myeloid leukemia (AML) was identified and shown to exhibit great anti-leukemic effect restricted to the FLT3-D835Y mutated cells while non-mutated cells were spared. Interestingly, the FLT3-D835Y TCR T cells also had the potential to eliminate the LSCs in vitro as shown by loss of leukemia initiating formation in mice following co-cultures.

The results from study II and III revealed the therapeutic potential of TCR T cells, therefore, in study IV we proposed TCR T cells as a targeted therapy against LSCs. Identification of antigens that mediate specific elimination of all LSCs across many patients has been challenging due to large inter- and intra-patient heterogeneity. Myeloproliferative leukemia protein (MPL) has an important role in lifelong maintenance of HSCs and is therefore expressed on all HSCs and consequently also all LSCs in patients with low to intermediate risk MDS. We suggest TCR T cells targeting MPL presented on HLA-A2 as an immunotherapeutic approach with clinical relevancy to treat an impending relapse following an haploidentical allo- HSCT. MPL TCR T cells would potentially also represent a novel research tool to study the dependency of HSCs and LSCs in normal and malignant hematopoiesis.