¹⁸F-BMS-986192, an adnectin-based human programmed cell death ligand 1 (PD-L1) tracer, was developed to non-invasively determine whole-body PD-L1 expression by positron emission tomography (PET). We evaluated usability of ¹⁸F-BMS-986192 PET to detect different PD-L1 expression levels and therapy-induced changes of PD-L1 expression in tumors. Methods: In vitro binding assays with ¹⁸F-BMS-986192 were performed in human tumor cell lines with different total cellular and membrane PD-L1 protein expression levels. Subsequently, PET imaging was executed in immunodeficient mice xenografted with these cell lines. Mice were treated with interferon gamma (IFNγ) intraperitoneally for 3 days or with the mitogen-activated protein kinase kinase (MEK1/2) inhibitor selumetinib by oral gavage for 24 hours. Thereafter ¹⁸F-BMS-986192 was administered intravenously, followed by a 60-minute dynamic PET scan. Tracer uptake was expressed as percentage injected dose per gram tissue (%ID/g). Tissues were collected to evaluate ex vivo tracer biodistribution and to perform flow cytometric, Western blot, and immunohistochemical tumor analyses. Results: ¹⁸F-BMS-986192 uptake reflected PD-L1 membrane levels in tumor cell lines, and tumor tracer uptake in mice was associated with PD-L1 expression measured immunohistochemically. In vitro IFNγ treatment increased PD-L1 expression in the tumor cell lines and caused up to 12-fold increase in tracer binding. In vivo, IFNγ did neither affect PD-L1 tumor expression measured immunohistochemically nor ¹⁸F-BMS-986192 tumor uptake. In vitro, selumetinib downregulated cellular and membrane levels of PD-L1 of tumor cells by 50% as measured by Western blotting and flow cytometry. In mice, selumetinib lowered cellular, but not membrane PD-L1 levels of tumors and consequently no treatment-induced change in ¹⁸F-BMS-986192 tumor uptake was observed. Conclusion: ¹⁸F-BMS-986192 PET imaging allows detection of membrane-expressed PD-L1, as soon as 60 minutes after tracer injection. The tracer can discriminate a range of tumor cell PD-L1 membrane expression levels.

Background: To better predict response to immune checkpoint therapy and toxicity in healthy tissues, insight in the in vivo behavior of immune checkpoint targeting monoclonal antibodies is essential. Therefore, we aimed to study in vivo pharmacokinetics and whole-body distribution of zirconium-89 (⁸⁹Zr) labeled programmed cell death protein-1 (PD-1) targeting pembrolizumab with positron-emission tomography (PET) in humanized mice. Methods: Humanized (huNOG) and non-humanized NOG mice were xenografted with human A375M melanoma cells. PET imaging was performed on day 7 post ⁸⁹Zr-pembrolizumab (10 µg, 2.5 MBq) administration, followed by ex vivo biodistribution studies. Other huNOG mice bearing A375M tumors received a co-injection of excess (90 µg) unlabeled pembrolizumab or ⁸⁹Zr-IgG₄ control (10 µg, 2.5 MBq). Tumor and spleen tissue were studied with autoradiography and immunohistochemically including PD-1. Results: PET imaging and biodistribution studies showed high ⁸⁹Zr-pembrolizumab uptake in tissues containing human immune cells, including spleen, lymph nodes and bone marrow. Tumor uptake of ⁸⁹Zr-pembrolizumab was lower than uptake in lymphoid tissues, but higher than uptake in other organs. High uptake in lymphoid tissues could be reduced by excess unlabeled pembrolizumab. Tracer activity in blood pool was increased by addition of unlabeled pembrolizumab, but tumor uptake was not affected. Autoradiography supported PET findings and immunohistochemical staining on spleen and lymph node tissue showed PD-1 positive cells, whereas tumor tissue was PD-1 negative. Conclusion: ⁸⁹Zr-pembrolizumab whole-body biodistribution showed high PD-1-mediated uptake in lymphoid tissues, such as spleen, lymph nodes and bone marrow, and modest tumor uptake. Our data may enable evaluation of ⁸⁹Zr-pembrolizumab whole-body distribution in patients.

Purpose: Probody™ therapeutic CX-072 is a protease-activatable antibody that is cross-reactive with murine and human programmed death-ligand 1 (PD-L1). CX-072 can be activated in vivo by proteases present in the tumor microenvironment, thereby potentially reducing peripheral, anti-PD-L1-mediated toxicities. To study its targeting of PD-L1-expressing tissues, we radiolabeled CX-072 with the positron emission tomography (PET) isotope zirconium-89 (⁸⁹Zr). Experimental Design: ⁸⁹Zr-labeled CX-072, non-specific Probody control molecule (PbCtrl) and CX-072 parental antibody (CX-075) were injected in BALB/c nude mice bearing human MDA-MB-231 tumors or C57BL/6J mice bearing syngeneic MC38 tumors. Mice underwent serial PET imaging 1, 3 and 6 days post intravenous injection (pi), followed by ex vivo biodistribution. Intratumoral ⁸⁹Zr-CX-072 distribution was studied by autoradiography on tumor tissue sections, which were subsequently stained for PD-L1 by immunohistochemistry. Activated CX-072 species in tissue lysates were detected by Western capillary electrophoresis. Results: PET imaging revealed ⁸⁹Zr-CX-072 accumulation in MDA-MB-231 tumors with 2.1-fold higher tumor-to-blood ratios at 6 days pi compared to ⁸⁹Zr-PbCtrl. Tumor tissue autoradiography showed high ⁸⁹Zr-CX-072 uptake in high PD-L1-expressing regions. Activated CX-072 species were detected in these tumors, with 5.3-fold lower levels found in the spleen. Furthermore, ⁸⁹Zr-CX-072 uptake by lymphoid tissues of immune-competent mice bearing MC38 tumors was low compared to ⁸⁹Zr-CX-075, which lacks the Probody design. Conclusions: ⁸⁹Zr-CX-072 accumulates specifically in PD-L1-expressing tumors with limited uptake in murine peripheral lymphoid tissues. Our data may enable clinical evaluation of ⁸⁹Zr-CX-072 whole-body distribution as a tool to support CX-072 drug development (NCT03013491).

Purpose: CX-072, a PD-L1–targeting Probody therapeutic, is engineered to be activated by tumor proteases that remove a masking peptide. To study effects on biodistribution and pharmacokinetics, we performed 89Zr-CX-072 positron emission tomography (PET) imaging. Experimental Design: Patients received ∼1 mg, 37 MBq 89Zr-CX-072 plus 0, 4, or 9 mg unlabeled CX-072 and PET scans at days 2, 4, and 7. After that, treatment comprised 10 mg/kg CX-072 q2 weeks (n = 7) + 3 mg/kg ipilimumab q3w 4× (n = 1). Normal organ tracer uptake was expressed as standardized uptake value (SUV)mean and tumor uptake as SUVmax. PD-L1 expression was measured immunohistochemically in archival tumor tissue. Results: Three of the eight patients included received 10-mg protein dose resulting in a blood pool mean SUVmean ± SD of 4.27 ± 0.45 on day 4, indicating sufficient available tracer. Tumor uptake was highest at day 7, with a geometric mean SUVmax 5.89 (n = 113) and present in all patients. The median follow-up was 12 weeks (4–76+). One patient experienced stable disease and two patients a partial response. PD-L1 tumor expression was 90% in one patient and ≤1% in the other patients. Mean SUVmean ± SD day 4 at 10 mg in the spleen was 8.56 ± 1.04, bone marrow 2.21 ± 0.46, and liver 4.97 ± 0.97. Four patients out of seven showed uptake in normal lymph nodes and Waldeyer's ring. The tracer was intact in the serum or plasma. Conclusions: 89Zr-CX-072 showed tumor uptake, even in lesions with ≤1% PD-L1 expression, and modest uptake in normal lymphoid organs, with no unexpected uptake in other healthy tissues.

The human epidermal growth factor receptor 3 (HER3) is an interesting target for antitumor therapy. For optimal HER3 signaling inhibition, a biparatopic Nanobody construct (MSB0010853) was developed that binds 2 different HER3 epitopes. In addition, MSB0010853 contains a third HER3 epitope that binds albumin to extend its circulation time. MSB0010853 is cross-reactive with HER3 and albumin of mouse origin. We aimed to gain insight into MSB0010853 biodistribution and tumor uptake by radiolabeling the Nanobody construct with ⁸⁹Zr. Methods: MSB0010853 was radiolabeled with ⁸⁹Zr. Dose- and time-dependent tumor uptake was studied in nude BALB/c mice bearing a subcutaneous HER3 overexpressing H441 non–small cell lung cancer xenograft. Dose-dependent biodistribution of ⁸⁹Zr-MSB0010853 was assessed ex vivo at 24 h after intravenous injection. Protein doses of 5, 10, 25, 100, and 1,000 μg were used. Time-dependent biodistribution of MSB0010853 was analyzed ex vivo at 3, 6, 24, and 96 h after intravenous administration of 25 μg of ⁸⁹Zr-MSB0010853. PET imaging and biodistribution were performed 24 h after administration of 25 μg of ⁸⁹Zr-MSB0010853 to mice bearing human H441, FaDu (high HER3 expression), or Calu-1 (no HER3 expression) tumor xenografts. Results: Radiolabeling of MSB0010853 with ⁸⁹Zr was performed with a radiochemical purity of greater than 95%. Ex vivo biodistribution showed protein dose– and time-dependent distribution of ⁸⁹Zr-MSB0010853 in all organs. Uptake of ⁸⁹Zr-MSB0010853 in H441 tumors was only time-dependent. Tumor could be visualized up to at least 96 h after injection. The highest mean SUV of 0.6 ± 0.2 was observed at 24 h after injection of 25 μg of ⁸⁹Zr-MSB0010853. ⁸⁹Zr-MSB0010853 tumor uptake correlated with HER3 expression and was highest in H441 (6.2 ± 1.1 percentage injected dose per gram [%ID/g]) and lowest in Calu-1 (2.3 ± 0.3 %ID/g) xenografts. Conclusion: ⁸⁹Zr-MSB0010853 organ distribution and tumor uptake in mice are time-dependent, and tumor uptake correlates with HER3 expression. In contrast to tumor uptake except for kidney uptake, organ distribution of ⁸⁹Zr-MSB0010853 is protein dose–dependent for the tested doses. ⁸⁹Zr-MSB0010853 PET imaging gives insight into the in vivo behavior of MSB0010853.