Characterization and targeting of leukemia propagating cells

Abstract

Patients with hematopoietic malignancies have a large, unmet medical need. Allogeneic hematopoietic stem cell transplantation remains the only curative option, but many patients are not eligible for such treatment. Many conventional therapies are also associated with a high degree of toxicities and despite initial remission, relapses are common with worse prognosis. There is an urgent need to better understand the mechanism driving leukemic transformation and to develop novel efficient therapies that specifically target tumor cells while sparing normal cells to circumvent unwanted side effects.

Recent advances in sequencing technology have revealed which somatically acquired oncogenic mutations are involved in driving the disease progression in most leukemias. Notably, the initiating mutations seen in myeloid malignancies have also been identified in expanded clones in aged healthy individuals. This age-related phenomenon is referred to as clonal hematopoiesis (CH) and confers an enhanced risk for malignant transformation where the expanded clones are suggested to represent a pre-malignant state. Mouse models could be of great benefit to better understand how CH contribute to leukemic transformation, but their relevance has been questioned. In Paper I, we show that mutations associated with human CH also spontaneously occur and promote clonal expansion in aged healthy mice. The detected mouse CH mutations were rare in bone marrow cells during steady state but became more frequent after being transplanted into second recipients, which allowed for expansion of the clones through stress or additional aging. Our results support the relevance of using aged and genetically modified mice to model the impact of CH in normal hematopoiesis and leukemic transformation that hopefully will lead to novel mechanistic insights and the development of early therapeutic interventions that can reverse the pathogenic effects of CH.

Cancer stem cells (CSCs) are believed to be required and sufficient for the propagation of cancer and have been linked to therapy resistance, in part through their quiescence nature making them difficult to target with conventional cell-cycle dependent therapies. CSC selective escape from therapeutic targeting could explain why a large fraction of patients relapse many years after seemingly successful treatment. The identity of CSCs is best established in myeloid malignancies such as myelodysplastic syndrome (MDS), where patients are rarely cured by chemotherapy and relapses are frequent after initial remissions. In Paper II, we show that a commercially available growth factor agonist, Romiplostim, activates primary MDS-CSCs and in a mouse model for MDS, we show that administration of Romiplostim prior to chemotherapy agents effectively enhanced the elimination of CSCs while having minimal impact on the normal hematopoietic stem cell compartment. This study provides proof-of-principle for the clinical relevance of Romiplostim-induced activation of CSCs to reduce their therapy resistance and hopefully reduce the risk for patient relapses after initial remission.

Adaptive cell transfer with engineered T cells expressing a T cell receptor (TCR) against specific tumor antigens have become a promising and rapidly growing field for cancer treatment. Unlike chimeric antigen receptors, TCR-T cells can recognize both intracellular and surface antigens which enables targeting of a broad range of antigens. TCR T cells can efficiently eliminate tumor cells but can also cause severe toxicity in patients if the target antigen is expressed on healthy cells. The choice of antigen is therefore critical for tumor clearance but also from a safety perspective. Terminal deoxynucleotidyl transferese (TdT) is overexpressed in the majority of ALL and its expression is restricted to the lymphoid lineage during a specific stage of B and T cell differentiation and could therefore serve as a good target with minimal off-target toxicities. In Paper III, we show that TCR T cells targeting TdT efficiently eliminate primary ALL cells both in vitro and in vivo while sparing normal cells. Our results indicate that this TCR therapy is efficient and safe and could serve as an alternative therapy for children and adults with ALL when standard therapies fail.

The ideal oncogenic targets are exclusively expressed on tumor cells to minimize any off-target toxicities. Neoantigens derived from recurrent oncogenic mutations represent candidate targets unique to the cancer cells, however, most driver mutations do not yield immunogenic neoantigens presented on the surface of the target cells. In Paper IV, we identify a TCR reactive to the recurrent FLT3D835Y driver mutation in AML. We show that FLT3 TCR T cells specifically eliminate primary human AML cells harboring the relevant mutation while sparing normal cells expressing wild type FLT3 in vitro and in vivo using two different patient-derived primary leukemia samples engrafted in immune-compromised mice. Our results from Paper III and Paper IV indicates that TCR T cell therapy can serve as a promising immunotherapy for hematopoietic malignancies as our studies show efficient and specific targeting of tumor cells with either shared tissue specific antigens or recurrent mutations. These studies can pave the way for more therapeutic TCRs that can be safely utilized in the clinic.