Abstract
176
Objectives: CAR-modified T cells show impressive results in clinical trials. However, cytokine release syndrome and “on-target, off-tumor” reactions represent most concerning side effects. To avoid such side effects a switchable UniCAR platform was established consisting of two components: UniCAR-modified T cells and specific targeting modules (TMs). For targeting of EGFR positive tumor cells a mono- (α-EGFR-) and a novel bivalent (α-EGFR-EGFR-) TM were constructed, expressed, purified, and compared with respect to their binding and killing capability, and their biodistribution was analyzed by PET.
Methods: The UniCAR contains a dual CD28/CD3ζ signaling domain (UniCAR 28/ζ). The epidermoid carcinoma cell line A431 was transfected with the firefly luciferase. Cloning of the α-EGFR TM into the lentiviral vector p6NST50, transduction of CHO wt cells, Ab expression, purification and characterization by SDS-PAGE/immunoblotting was described previously (1). Both TMs, including the novel α-EGFR-EGFR TM are based on the camelid Nb clone 7C12. Both TMs were compared by HPLC, flow cytometry, cytotoxicity and cytokine-release assays. Receptor expression density was estimated for and binding of the TMs was verified with A431, FaDu, PC3-PSCA and MDA-MB-435S. The TMs were conjugated with p-SCN-Bn-NODAGA and radiolabeled with 64Cu according standard methodologies. The PET kinetics and biodistribution was studied in A431-Luc+ tumor bearing female NMRI-Foxn1nu/Foxn1nu mice. For optical imaging of anti-tumor effects, 1.5x106 A431 cells were investigated alone or together with 1.5x106 human UniCAR 28/ζ-armed T cells in the presence or absence of 600 pmol of the α-EGFR- or the α-EGFR-EGFR TM.
Results: Both TMs were successfully established, purified, conjugated with NODAGA and radiolabeled with specific activities larger than 24 GBq/µmol TM. The α-EGFR-EGFR TM showed an improved redirection of UniCAR T cells against EGFR+ carcinoma cells and was able to bind to cell lines expressing high to low levels of EGFR. In contrast, the α-EGFR TM did only bind to cell lines with high EGFR expression. The UniCAR epitope present in both TMs was accessible for interactions with an Ab domain directed against the UniCAR epitope even after binding of the TM to EGFR. For the α‑EGFR-EGFR TM we estimated a three times higher Kd value (24 nM) compared to the α-EGFR TM (Kd = 77 nM). The improved in vitro killing efficacy of the α-EGFR-EGFR TM turns also into an improved killing capability in vivo. The biodistribution of the α-EGFR- and α-EGFR-EGFR TMs showed 2 h after injection specific tumor accumulation (SUV) 1.05±0.39, 0.94±0.05, blood concentration of 0.28±0.10, 1.55±0.19 and tumor to muscle ratios of 17±4, 12±5, respectively. The PET studies (Fig. 1) after 24 h demonstrated higher tumor to background ratio for the α-EGFR-EGFR TM.
Conclusions: The study shows, that the avidity of the bivalent TM is higher than of its monovalent counterpart. Most of interest the killing efficacy of the α-EGFR-EGFR TM is also improved and is by far better than that of the monovalent TM. According to biodistribution and PET analysis, the increased molecular weight of the α-EGFR-EGFR TM delays its elimination and thereby improves the enrichment at the tumor site, the increased contrast after sufficient time and consequently its use for PET imaging. References: 1. Albert S, Arndt C, Feldmann A, et al. A novel nanobody-based target module for retargeting of T lymphocytes to EGFR-expressing cancer cells via the modular UniCAR platform. Oncoimmunology. 2017;6:e1287246.Figure 1. Maximum intensity projections of PET images after injection of the TMs (left) and the corresponding activity concentration in the tumors.