Evaluation of [18F]TZ4877, a radioligand targeting the Sphingosine 1 Phosphate Receptor 1, for imaging neuroimmune response in Nonhuman Primates

Introduction: The sphingosine-1-phosphate receptor-1 (S1PR₁) is involved in regulating microglia and astrocyte responses to neuroimmune stimuli, and generally considered to be neuroprotective or anti-inflammatory. The S1PR₁-specific radioligand [¹¹C]CS1P1 is a promising candidate for human imaging, supported by whole body dosimetry studies¹ and pilot studies in MS patients², but it exhibits protracted kinetic properties. Therefore, a need remains for a S1PR­­­₁ radioligand with faster kinetic properties. The goal of this work was to evaluate a ¹⁸F-labeled radiotracer, [¹⁸F]TZ4877, including comparison with [¹¹C]CS1P1, in nonhuman primates.

Methods: [¹⁸F]TZ4877 was prepared via nucleophilic substitution of a tosylate precursor with K[¹⁸F]/F^- followed by deprotection, while [¹¹C]CS1P1 was prepared by methylation of t-Bu-protected precursor followed by deprotection. PET data were acquired with a Focus-220 scanner in three rhesus monkeys (3 M; 6.5-12.5 years). Bolus injection of radiotracer was followed by up to 240 min ([¹⁸F]TZ4877) or 120 min ([¹¹C]CS1P1) of PET acquisition, with arterial plasma assay and metabolite analysis to measure the parent input function and plasma free fraction (f_P). Baseline [¹⁸F]TZ4877 and [¹¹C]CS1P1 scans were acquired in two animals each. Challenge [¹⁸F]TZ4877 scans were acquired 167-196 min after 1 ng/kg endotoxin administration (n=2). Blocking studies were acquired 33 min after administration of 0.4-0.8 mg/kg TZ82112, which is highly specific for S1PR₁. Data were analyzed with one- and two- tissue compartment models (1TCM & 2TCM) and multilinear analysis (MA1) to estimate volumes of distribution (V_T).

Results: [¹⁸F]TZ4877 and [¹¹C]CS1P1 were prepared in high molar activity (45-418 MBq/nmol at time of injection) and >96% radiochemical purity. At baseline, [¹⁸F]TZ4877 compared with [¹¹C]CS1P1 exhibited lower parent fractions (22-30% vs 55-80% at 60 min) and comparable f_P values (0.6-0.9% vs 0.5-0.9%). Tissue kinetics were faster for [¹⁸F]TZ4877, and data were best modeled by 2TCM with regional V_T values ranging from 2.2-5.1 mL/cm³. [¹¹C]CS1P1 SUVs continued to increase at 120 min post-injection. 2TCM V_T estimates were unstable (s.e.>45%) while 1TCM yielded poor fits. MA1 (t*=20 min) yielded [¹¹C]CS1P1 V_T estimates ranging from 5.2-18.2 mL/cm³. Endotoxin increased [¹⁸F]TZ4877 V_T in all brain regions for both animals, with a robust mean V_T increase of 36% (range 9-59%) in one animal and modest mean increase of 6% (range 1-13%) in the other. Blocking with TZ82112 increased [¹⁸F]TZ4877 f_P to 1.6-2.0%, resulting in dose-dependent reductions in [¹⁸F]TZ4877 V_T/f_P throughout the brain.

Conclusions: [¹⁸F]TZ4877 exhibits suitable properties for imaging S1PR₁ in nonhuman primates, with faster kinetic properties than [¹¹C]CS1P1.

Funding: This work was supported by P30DA046345, P41EB025815, NS103988, and NS075527

References:

1) Brier MR, et al. Phase 1 evaluation of 11C-CS1P1 to assess safety and dosimetry in human participants. J Nucl Med. 2022.

2) Brier MR, et al. Measuring sphingosine-1-phosphate receptor-1 in multiple sclerosis using a novel positron emission tomography radioligand. ACTRIMS Forum 2022-Poster Presentations. 2022