Abstract 3579: Assessment of target-mediated biodistribution of an 89Zr labeled PD-L1/4-1BB bispecific Mabcalin protein

Abstract

Abstract Background: PRS-344/S095012 is a novel 4-1BB (CD137) and programmed death-ligand 1 (PD-L1) bispecific antibody-Anticalin® fusion protein (MabcalinTM protein) designed to cluster 4-1BB on activated T cells exclusively in the presence of PD-L1 expressing cells. We aimed to study PRS-344/S095012 in vivo biodistribution and pharmacokinetics with 89Zr-positron emission tomography (PET) at a dose with antitumoral activity in mice and evaluate the contribution of each targeting arm. Methods: PRS-344/S095012 lacks cross-reactivity to murine 4-1BB and PD-L1. To explore the PRS-344/S095012 biodistribution in a humanized 4-1BB knock-in mouse model, we synthesized the surrogate 89Zr-Atezo-J10 with the same 4-1BB building block and cross-reactivity to murine PD-L1. Humanized 4-1BB knock-in C57BL/6J (h4-1BB KI B6) and C57BL/6J (B6) mice (n=4-6 per group) were subcutaneously engrafted with murine wildtype MC38 colon adenocarcinoma cells. Tumors were grown to a minimum of ≥50 mm3 (average 163 mm3) before tracer injection. Mice received intravenously 30 µg (2.5 MBq) of 89Zr-PRS-344/S095012 or 89Zr-Atezo-J10 supplemented with PRS-344/S095012 or Atezo-J10 up to 10 mg/kg. Four mice groups were formed to distinguish between bispecific (Atezo-J10 in h4-1BB KI B6), monospecific PD-L1 (Atezo-J10 in B6), monospecific 4-1BB (PRS-344/S095012 in h4-1BB KI B6), and isotype (PRS-344/S095012 in B6) binding up to 4 days post-injection (pi). In addition, a fifth group (5 MBq 89Zr-Atezo-J10) was studied to visualize the bispecific biodistribution up to 7 days pi. At days 1, 2, 4, or 2, 4, 7 pi, mice underwent serial PET imaging to obtain mean and maximum standardized uptake (SUVmean/max) and retro-orbital blood sampling, followed by ex vivo biodistribution. Results: PET imaging showed 89Zr-Atezo-J10 specific tumor accumulation with higher tumor-to-blood ratios of respectively 2.2-, 2.6-, and 2.4-fold (p<0.01) at 4 days pi compared to monospecific binding of PD-L1, 4-1BB, and isotype. The ex vivo biodistribution demonstrated the same trend with respectively 4.2-, 5.5-, and 6.8-fold (p<0.01) increase in tumor-to-blood uptake for 89Zr-Atezo-J10 versus monospecific binding of PD-L1, 4-1BB, and isotype 89Zr-Atezo-J10 spleen uptake was comparable (ns) with monospecific binding of PD-L1 but elevated (p<0.01) compared to 4-1BB or isotype distribution. The uptake in lymph nodes (axillary, cervical, tumor-draining, and mesenteric) did not differ between the groups. Conclusion: 89Zr-Atezo-J10 specific accumulation in PD-L1 expressing tumors is due to both PD-L1 and 4-1BB binding and is higher than with PD-L1 and 4-1BB mono-targeting. This preclinical study supports the clinical evaluation of 89Zr-PRS-344/S095012’s whole-body distribution and the development of tumor-specific 4-1BB targeting bispecifics. Citation Format: Claudia A. van Winkel, Xiaoyu Fan, Danique Giesen, Glenn Gauderat, Lucia Pattarini, Thomas Jaquin, Anissa Barakat, Anne-Marie De La Bigne, Marleen Richter, Nicole Andersen, Julie Legrand, Helene Lelièvre, Elisabeth G. de Vries, Aizea Morales-Kastresana, Marjolijn N. Lub- de Hooge. Assessment of target-mediated biodistribution of an 89Zr labeled PD-L1/4-1BB bispecific Mabcalin protein. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 3579.