Generation of donor-derived Wilms tumor antigen 1–specific cytotoxic T lymphocytes with potent anti-leukemia activity for somatic cell therapy in children given haploidentical stem cell transplantation: a feasibility pre-clinical study

Abstract

Background

The Wilms tumor antigen 1 (WT1) is over-expressed in a vast majority of adult and childhood acute leukemia and myelodysplastic syndromes, being lowly or transiently expressed in normal tissues and hematopoietic stem cells (HSCs). A number of HLA-restricted WT1 epitopes are immunogenic, allowing the in vitro induction of WT1-specific cytotoxic T lymphocytes (CTLs) from patients and healthy donors.

Aim

The aim of the study was to investigate the feasibility of producing WT1-specific CTLs suitable for somatic cell therapy to prevent or treat relapse in children with acute myeloid or lymphoblastic leukemia given haploidentical HSC transplantation (haplo-HSCT).

Methods

For WT1-specific CTL production, donor-derived either peripheral blood mononuclear cells (PBMCs) or CD8+ lymphocytes were stimulated with WT1 peptide-loaded donor dendritic cells in the presence of interleukin (IL)-7 and IL-12. Effector cells were re-stimulated once with irradiated donor PBMCs pulsed with WT1-peptides, and then expanded in an antigen-independent way.

Results

WT1-specific CTLs, displaying high-level cytotoxicity against patients’ leukemia blasts and negligible activity against patients’ non-malignant cells, were obtained from both PBMCs and CD8+ lymphocytes. WT1-specific CTLs obtained from PBMCs showed a better expansion capacity and better anti-leukemia activity than those obtained from CD8+ lymphocytes, even though the difference was not statistically significant. In CTLs derived from PBMCs, both CD8+ and CD4+ subpopulations displayed strong anti-leukemia cytotoxic activity.

Discussion

Results of this pre-clinical study pave the way to a somatic cell therapy approach aimed at preventing or treating relapse in children given haplo-HSCT for WT1-positive leukemia.

Introduction

The prognosis for children affected by acute leukemia and who underwent transplantation in an advanced disease phase, in the presence of measurable minimal residual disease (MRD) or with unfavorable cytogenetic or molecular abnormalities is still poor. Based on retrospective data, the probability of relapse for high-risk patients often exceeds 50%. When relapse occurs after allogeneic hematopoietic stem cell transplantation (HSCT), in the majority of cases, only palliative therapy is possible [1].

In the last few years, good results have been obtained in the control of relapsed/refractory acute lymphoblastic leukemia (ALL) with cellular therapy based on the infusion of T lymphocytes genetically modified to express chimeric antigen receptors (CARs) targeting B-cell–associated antigens [2], [3]. The increase in CAR-T cell clinical efficacy, however, has been paralleled with the risk of severe adverse events, such as cytokine release syndrome and on-target off-tumor toxicities [2], [4]. In addition, despite data available for the treatment of patients with ALL, major challenges remain to be overcome to safely apply CAR-T cell therapy to patients with acute myeloid leukemia (AML) [5], [6].

Beside CAR-T cell therapy, also somatic cell therapy may offer a new tool to prevent or treat relapse. Clinical trials based on infusion of cytokine-induced killer (CIK) cells in the setting of allogeneic HSCT documented low graft-versus-host disease (GVHD) toxicity and anti-leukemia effect slightly superior to the unmanipulated donor leukocyte infusion (DLI) [7]. In addition, ex vivo–induced cytotoxic T lymphocytes (CTLs) endowed with cytotoxic activity against leukemia cells may represent an effective way to prevent or treat leukemia relapse after transplantation.

In the setting of allogeneic HSCT, we have previously demonstrated the feasibility of in vitro priming and expanding donor-derived naïve T lymphocytes with potent anti-leukemia activity by using apoptotic patient's leukemia blasts (LBs) as the priming stimulus [8]. Anti-leukemia CTLs produced in compliance with Good Manufacturing Practice (GMP) requirements were used to prevent or cure leukemia relapse after allogeneic HSCT in children, obtaining encouraging results [9]. However, the methodology described above has the key limitation of being applicable only when a sufficient number of patient's LBs, cryopreserved at time of diagnosis or relapse, is available.

In recent years, different proteins expressed by leukemic cells have emerged as possible targets of leukemia-specific CTLs, including non-polymorphic proteins that are over- or aberrantly expressed by leukemic cells. An alternative strategy to induce tumor-specific T-lymphocyte responses is the use of synthetic peptides derived from normal proteins preferentially expressed by tumor cells [10]. Wilms tumor antigen 1 (WT1) is a zinc finger-transcription factor that can be considered a model antigen for adoptive T-cell therapy [11], [12]. WT1 is over-expressed in up to 70% of adult and childhood hematologic malignancies such as AML, ALL, chronic myeloid leukemia (CML) and myelodysplastic syndromes [13], [14], [15] as well as in several solid tumors [16]. In contrast, low WT1 expression has been described in normal tissues such as kidneys and ovaries and a limited as well as transient expression in normal hematopoietic stem cells [17], [18].

The in vivo immunogenicity of WT1 has been documented by demonstration that WT1-specific CD8 T cells naturally occur during treatment in patients with AML or after allogeneic HSCT in patients with ALL [19], [20]. Furthermore, several studies have shown that peptides derived from WT1 protein are immunogenic in humans [21], [22] and that it is possible to induce in vitro WT1-pepide–specific cytotoxic T cells endowed with anti-leukemia activity. In particular, the possibility to generate, from patients or healthy donors, WT1-specific CTLs directed against tumor cell lines or leukemia cells has been described, suggesting that WT1-specific CTLs could be used for in vivo anti-tumor immunotherapy approaches [23], [24], [25], [26].

The aim of this study was to verify the possibility to use WT1 peptides for eliciting an in vitro donor-derived anti-leukemia immune response. We compared functional and phenotypic features of WT1-specific CTLs induced starting from peripheral blood mononuclear cells (PBMCs) or CD8+ cells to define the best in vitro strategy to generate donor-derived WT1-specific CTLs in a HLA-A2 donor/recipient pair.