T cell receptors targeting mutant NPM1 for adoptive cell therapy in acute myeloid leukemia Free

Background:

Acute myeloid leukemia (AML) is a rare form of blood cancer affecting myeloid stem and progenitor cells, associated with a poor prognosis and a 5-year survival rate of ~33%. Current treatments, such as intensive chemotherapy and stem cell transplantation, are not suitable for all patients and can cause significant toxicities. Therefore, there is an unmet need for safer, more effective forms of treatment.

One potential approach for AML is to target oncogene-derived epitopes using adoptively transferred T cells transduced with specific T cell receptors (TCRs), as shown successfully in solid tumors (Kim et al., 2022; Leidner et al., 2022). Promising targets in AML include nucleophosmin 1 (NPM1) mutations, which occur in ~30% of cases. Most result in a common frameshifted sequence that generates peptides presented by several class I human leukocyte antigens (HLAs) (van der Lee et al., 2021; van der Lee et al., 2019). Despite this, only an HLA-A*02:01-restricted TCR is being tested in NPM1-mutated patients (NCT06424340), highlighting the need to identify clinically useful TCRs restricted to other HLAs.

While most therapeutic TCR discovery efforts have relied on isolating reactive T cells from patients or HLA-matched healthy donors, an alternative approach is to vaccinate HLA-transgenic mice with peptides of interest, enabling the isolation of highly avid T cells targeting select antigens (Ade et al., 2023; Wang et al., 2016). In clinical trials, patient-derived T cells transduced with murine TCRs have been shown to induce objective responses in cancer, without eliciting host immune responses that compromise their effectiveness (Davis et al., 2010; Johnson et al., 2009; Morgan et al., 2013). Here, we present findings from a mutation-specific TCR search conducted in HLA-A*11:01-transgenic mice.

Methods:

HLA-A*11:01-transgenic mice were vaccinated with a mutant NPM1 (mtNPM1) nonamer peptide (AVEEVSLRK), previously shown to bind to this HLA. Lymphocytes from the spleens and lymph nodes of vaccinated mice were harvested and stimulated in vitro with the same peptide. Stimulated cells were then co-cultured with mtNPM1-pulsed K562-HLA-A*11:01 cells, followed by measurement of their reactivity using IFN-γ ELISA. Reactive cultures were stained with HLA-A*11:01-mtNPM1 tetramer and sorted for single-cell TCR sequencing. Murine TCR sequences were reconstructed and retrovirally transduced into human T cells for subsequent safety and efficacy testing against engineered cell lines and primary human samples.

Results:

Upon transduction into human T cells, six unique TCRs isolated from HLA-A*11:01-mtNPM1-reactive murine lymphocytes demonstrated recognition of mtNPM1 peptide presented by K562 cells expressing HLA-A*11:01. Of these, five mediated complete and specific lysis of a leukemia cell line (OCI-AML3) bearing mtNPM1 and transduced with HLA-A*11:01. Two also mediated robust lysis of primary AML cells co-expressing HLA-A*11:01 and mtNPM1 but not the cells expressing only one of the two, suggesting strong target specificity. These two TCRs were selected for further characterization and safety testing. Their recognition motifs, determined by alanine substitution assay, had no sequence matches in the human proteome. Additionally, neither TCR recognized structurally similar peptides, derived via a physicochemical grouping strategy and predicted to bind to HLA A*11:01. Both TCRs also failed to recognize non-tumor samples, including HLA-A*11:01-positive peripheral blood mononuclear cells and Epstein-Barr virus-immortalized B cells expressing diverse HLAs. Together, these data suggest that the two TCRs are highly efficacious in vitro and carry a potentially low risk for broad cross-reactivity and alloreactivity. In vivo efficacy studies are ongoing in a patient-derived xenograft mouse model.

Conclusion:

Vaccinating HLA-A*11:01-transgenic mice with mtNPM1 peptide generated two TCRs capable of specifically lysing HLA-A*11:01- and mtNPM1-postive primary AML samples, with additional in vitro data suggesting absence of cross- or alloreactivity. Should their efficacy be validated in vivo, both receptors would represent strong candidates for clinical development in adoptive T cell therapy for AML. These results also support the use of HLA-transgenic mice as a platform for generating high-specificity mtNPM1-reactive TCRs restricted to alleles other than HLA-A*11:01.

This research was supported by the Intramural Research Program of the NIH.