Acute Myeloid Leukemia eradication targeting cathepsin g with a dual HLA-restricted TCR

Abstract

Abstract Background Hematopoietic stem cell transplantation has unveiled the sensitivity of acute myeloid leukemia (AML) to immunotherapy and paved the way for the development of adoptive cell therapies. In this context, T cell receptor (TCR)-redirected T cells offer new therapeutic opportunities due to their ability to target both intracellular and extracellular antigens with great sensitivity and specificity. Despite the heterogeneity and the high clonal instability of AML, the broad repertoire of antigens recognized by TCRs offers the possibility to select a target shared across different AML clones and subtypes, absent in healthy tissues and essential for blast survival, a feature that reduces the risk of antigen escape. In this study, we isolated and functionally characterized a unique TCR specific for Cathepsin G (CTSG), a serine protease overexpressed by AML blasts and leukemic stem cells. While normally confined inside azurophilic granules of neutrophils, CTSG has been reported to be aberrantly localized in the cytoplasm of leukemic cells, thus increasing its processing and presentation on human leukocyte antigen (HLA) class I molecules, where it can be detected by T cells. Methods To isolate anti-tumor lymphocytes, peripheral blood mononuclear cells from two healthy donors were repeatedly stimulated with autologous antigen-presenting cells loaded with overlapping 15-mer peptides. The expansion of CTSG-specific T cells was assessed by evaluating cytokine release and degranulation marker expression upon co-culture with HLA-matched target cells. To identify and pair tumor-specific αβ clonotypes, bulk TCR sequencing was performed. Lentiviral vectors encoding the identified TCRs were then generated and used to transduce healthy donor T cells. Prior to transduction, endogenous TCRs were knocked out using CRISPR/Cas9 technology to prevent mispairing with transgenic α and β chains. A second transduction with a lentiviral vector encoding CD8αβ co-receptor was performed to obtain a fully functional T cell army. Results We identified three CTSG-specific TCRs with different HLA restrictions. Among them, P56-TCR stood out for its ability to recognize two CTSG peptides derived from the same 15-mer epitope, presented on either HLA-A*24:02 or HLA-C*07:02—two alleles collectively expressed by approximately 40% of the Caucasian population. The natural processing and presentation of these peptides were confirmed by the ability of P56-TCR-redirected T cells to eliminate primary AML blasts expressing either HLA-A*24:02orHLA-C*07:02 in both in vitro and in vivo models. Notably, in vivo experiments showed a statistically significant reduction in leukemia burden (p < 0.0001) in treated mice compared to untreated controls, with complete disease eradication in the bone marrow, highlighting the therapeutic potential of this approach. In vitro functional re-challenge and single-cell analysis were performed on T cells retrieved from mice at the end of the experiment. Results elucidated the features of P56-TCR T cells, composed of a CD8+ T cell cytotoxic population with a potent antitumoral activity and a double positive CD4+CD8+ T cell subset characterized by a strong helper function. Lastly, safety assessments with alanine scanning and colony-forming unit assays detected no evidence of on-target/off-tumor or off-target/off-tumor toxicities, supporting the favorable safety profile of P56-TCR-engineered T cells. Conclusion The P56-TCR emerges as a promising option for AML treatment across a wide patient population, demonstrating robust antitumor activity and a favorable safety profile. These findings pave the way for its potential clinical translation.