Discovery of bispecific T-cell engagers targeting the non-mutated regions of mutant calreticulin for the treatment of myeloproliferative neoplasms

Abstract Introduction Calreticulin (CALR) is a chaperone protein found in the endoplasmic reticulum (ER). A +1 frameshift mutation in exon 9 of the CALR gene generates a novel C-terminal peptide that induces the pathogenesis of primary myelofibrosis (PMF) and essential thrombocythemia (ET). In tumor cells expressing mutant CALR, a complex formation between mutant CALR and the thrombopoietin receptor (c-MPL) has been observed. The mutant CALR/c-MPL complex on the surface of tumor cells activates the JAK2/STAT pathway and its downstream signaling during the development of myeloproliferative neoplasms (MPNs). Therefore, therapeutic antibodies targeting the mutation-specific neoepitope located at the C-terminal region of the mutated CALR have recently been developed to treat PMF and ET caused byCALR mutations. However, the mutant CALR/c-MPL complex is exposed on the cell surface of tumor cells harboring CALR mutations, suggesting that not only the mutation-specific neoepitope at the C-terminal region of mutant CALR, but also the non-mutated regions of mutant CALR, which share amino acid sequences with wild-type CALR, may serve as potential therapeutic targets. We hypothesized that using a bispecific T-cell engager molecule (BTCE) targeting the non-mutated regions of mutant CALR could induce cytotoxicity against CALR-mutated tumor cells without affecting normal cells. This study aimed to generate BTCE targeting the non-mutated regions of mutant CALR and evaluate its function in vitro. Methods and Materials To obtain novel high-affinity monoclonal antibodies against the non-mutated regions of mutant CALR, rabbits were immunized with a recombinant protein consisting of conserved regions from wild-type and mutant CALR. Positive B-cell clones were screened and sorted as single cells. The Fab region of each clone was amplified via PCR and expressed in a cell-free system using “Ecobody technology®.” The binding capacities of the obtained Fab clones to conserved regions of wild-type and mutant CALR were examined using ELISA. Selected Fab clones were incorporated into human IgG to generate chimeric IgG clones. The binding affinity (KD) to UT-7/TPO/Del52 and UT-7/TPO/Ins5 cells expressing mutant CALR was measured by flow cytometry. We then created BTCEs consisting of high-affinity clones targeting mutant CALR and anti-CD3 in IgG1 format. We evaluated their T-cell-dependent cytotoxic activity against UT-7/TPO/Del52 and UT-7/TPO/Ins5 cells. For the in vitro cytotoxicity assay, we used either lymphokine-activated T cells (T-LAKs) or primary T cells as effector cells. Several selected high-affinity clones targeting mutant CALR were humanized and their binding to peripheral blood cells from healthy individuals was examined. Results and Discussion Using single-B-cell screening technology, we obtained many Fab clones with binding affinity to the non-mutated regions of the mutant CALR. Fifteen Fab clones were selected as candidates for BTCE development. The clones were converted into chimeric IgGs, and their ability to bind to UT-7/TPO/Del52 and UT-7/TPO/Ins5 cells was assessed. The KD values for the high-affinity clones were approximately 2×10E-10 M for UT-7/TPO/Del52 and UT-7/TPO/Ins5 cells. They did not bind to UT-7/TPO/vector cells that do not express the mutant CALR. Furthermore, some high-affinity clones were humanized, and it was confirmed that these humanized clones did not bind to peripheral blood cells from healthy individuals. We created BTCEs consisting of a humanized clone and an anti-CD3 antibody. One of these BTCEs exhibited specific cytotoxic activity against UT-7/TPO cells expressing mutant CALR in the presence of T-LAK or primary T cells. The EC50 values for this activity were 119.3 and 475.4 pM in UT-7/TPO/Del52 and 65.6 and 77.4 pM in UT-7/TPO/Ins5, respectively. Notably, this BTCE did not elicit a cytotoxic response to the UT-7/TPO/vector cells. These results suggest that BTCEs targeting the non-mutated regions of mutant CALR can induce cell death in CALR-mutated tumor cells without harming normal peripheral blood cells. Conclusion We successfully developed new BTCEs that target the non-mutated regions of mutant CALR. These BTCEs induce cell death in mutant CALR-expressing tumor cells while leaving normal peripheral blood cells unaffected. Thus, our BTCEs represent a promising therapeutic option for the treatment of CALR-mutated MPNs.