Full preclinical validation of the 123I-labeled anti-PSMA antibody fragment ScFvD2B for prostate cancer imaging

Abstract

Purpose: In the context of prostate cancer (PCa) imaging, the aim of this study was to optimize (in vitro) the specificity and assess preclinically (in vivo) the tumor targeting properties of the 123I-scFvD2B antibody specific for prostate-specific membrane antigen (PSMA).

Experimental design: The 123I-labeling conditions of the antibody fragment scFvD2B, produced in an eukaryotic system under GMP-compliant conditions, were optimized and assessed for purity and immunoreactivity. The specificity and potency of tumor uptake were tested in three preclinical in vivo models of subcutaneously xenografted human tumors expressing different levels of PSMA (LNCaP, naturally expressing PSMA; PC3-PIP and LS174T-PSMA, transfected with PSMA) or PC3 and LS174T, as negative controls, to assess the clearance, biodistribution and imaging potential of 123I-scFvD2B.

Results: The set conditions of production and radiolabeling yielded a reagent suitable for human delivery thanks to the purity of the formulation and the high immunoreactivity. In all preclinical models 123I-scFvD2B showed specific targeting only to PSMA-positive tumors with the final specific activity ranging up to 1500 MBq/mg. Despite different levels of PSMA expression, biodistribution analyses and SPECT/CT imaging demonstrated similar results and maximal signal-to-background ratios 24 hours after injection.

Conclusions: Due to its in vitro and in vivo properties, 123I-scFvD2B could be a promising tool for the early diagnosis of PCa, and may represent a molecular imaging option to monitor disease progression and assist in the clinical management of PCa patients.

Keywords: 123I-radiolabeled antibody; imaging; prostate cancer; prostate-specific membrane antigen; scFv antibody fragment.

Figure 1

A. Flow cytometric analysis of scFvD2B. scFvD2B (solid line) binding on PSMA-positive cell lines (PC3-PIP, LNCaP and LS174T-PSMA) and PSMA-negative cell lines (PC3, A431, LS174T). The shift in fluorescence was assessed relatively to a negative control (gray histogram, cells incubated with biotinylated Protein L and streptavidin-PE). B. BIAcore analysis: scFvD2B obtained from prokaryotic system (Blue line) and from eukaryotic system (Green line) on PC3-PIP lisate; scFvD2B obtained from prokaryotic system (Red line) and from eukaryotic system (Purple line) on LNCaP lisate.

Figure 1

A. Flow cytometric analysis of scFvD2B. scFvD2B (solid line) binding on PSMA-positive cell lines (PC3-PIP, LNCaP and LS174T-PSMA) and PSMA-negative cell lines (PC3, A431, LS174T). The shift in fluorescence was assessed relatively to a negative control (gray histogram, cells incubated with biotinylated Protein L and streptavidin-PE). B. BIAcore analysis: scFvD2B obtained from prokaryotic system (Blue line) and from eukaryotic system (Green line) on PC3-PIP lisate; scFvD2B obtained from prokaryotic system (Red line) and from eukaryotic system (Purple line) on LNCaP lisate.

Figure 2. Clearance of

¹²³I-scFvD2B in tumor-bearing mice. Blood (★), PSMA-positive tumor (PC3-PIP; ·) PSMA-negative tumor ( PC3;) at different time points after injection of 7.5 MBq (100 µg; specific activity [SA] = 75 MBq/mg) ¹²³I-scFvD2B (mean of 2 experiments). Values are expressed as mean %ID/g ± SD; the number of animals ranged from 3 to 14.

Figure 3. Biodistribution and localization after intravenous administration of

¹²³I-scFvD2B to athymic mice. Uptake and retention were measured in different organs as decay-adjusted percentage of the injected dose per gram of tissue (% ID/g). A. Biodistribution in non-tumor-bearing mice evaluated 24 hours post injection, 7.5 MBq (41 µg; SA = 181 MBq/mg) administered; error bars represent SD from the mean value of 4 mice. B. Biodistribution in LNCaP tumor bearing mice evaluated 24 hours post injection; 7.5 MBq (51 µg; SA = 145 MBq/mg) administered; error bars represent SD from the mean value of 2 mice. C. Biodistribution in PC3-PIP/PC3 tumor bearing mice evaluated at 3, 9, 15 and 24 hours post injection; 7.5 MBq (95 µg; SA = 78 MBq/mg) administered; error bars represent SD from the mean value of 3 to 6 mice.

Figure 4. Representative SPECT/CT images in LS174T-PSMA/LS174T model after intravenous administration of

123I-scFvD2B. PSMA-positive tumor LS174T-PSMA (right flank) and PSMA-negative tumor LS174T (left flank) evaluated at 3, 9 and 24 hours post injection. Left: 12 MBq (8 µg; SA = 1500 MBq/mg) administered. Right: 12 MBq administered plus a 100-fold excess of cold scFvD2B.