Quantification of central substance P receptor occupancy by aprepitant using small animal positron emission tomography

Abstract

Background: Central substance P receptors, termed NK-1 receptors, have been considered as therapeutic targets in the development of drugs against diverse conditions, including emesis, overactive bladder, and depression.

Methods: Here, we applied small animal positron emission tomography (PET) and a radioligand for NK-1 receptors ([(18)F]FE-SPA-RQ) for measuring occupancies of these receptors by a selective antagonist (aprepitant) in order to examine the validity of this in vivo imaging system for preclinical characterization of candidate agents acting on NK-1 receptors, and as a tool for predicting optimal doses in humans.

Results: PET in gerbils depicted high uptake in the striatum and dose-dependent displacement with increasing doses of aprepitant. Occupancies increased as a function of aprepitant plasma concentrations according to a one-site competition model, which agrees with reported occupancy-concentration relationships in clinical studies after correction for species differences in plasma protein-unbound aprepitant fractions. These occupancy data were further supported by ex vivo autoradiography of brain samples from aprepitant-treated gerbils. In a pilot study of a marmoset, we obtained more accurate determinations of NK-1 receptor occupancy, less affected by spillover of signals from extracranial tissues than in gerbil experiments.

Conclusions: These findings support the utility of small animals and quantitative PET in the development of drugs targeting NK-1 receptors.

Keywords: NK-1 receptor; receptor occupancy; small-animal PET; substance P.

Figure 1.

Representative coronal PET images showing [¹⁸F]FE-SPA-RQ distribution in gerbil brains at baseline and after oral administration (per os; p. o.) of aprepitant. PET data were generated by summation of dynamic data at 0–6h after intravenous radioligand injection, and were merged onto the MRI template. ROIs (outlined by dots) were placed on the striatum (upper row) and cerebellum (lower row).

Figure 2.

Time–radioactivity curves for [¹⁸F]FE-SPA-RQ in the gerbil striatum (closed squares) and cerebellum (open rhomboids) at baseline (A) and after pretreatments with aprepitant at doses of 0.03mg/kg (B), 0.1mg/kg (C), 0.3mg/kg (D), 3mg/kg (E), and 30mg/kg (F). Data were generated by defining ROIs on the PET images displayed in Figure 1. Radioligand uptake into each region was expressed as a percentage of injected dose per unit tissue volume (%dose/mL). Bars indicate standard errors of mean (n = 6 in each treatment group).

Figure 3.

PET quantification of radioligand binding in the gerbil striatum. Time stability of [¹⁸F]FE-SPA-RQ binding potential values obtained by SRTM (BP_ND; A) and on the basis of the target-to-reference ratio of radioactivity (B). Baseline data without pretreatments were used for calculations. In panel B, regional radioactivity was quantified by averaging 15min dynamic data, and binding potential was determined as (striatum-to-cerebellum ratio) – 1. (C) Scatterplot demonstrating the relationship between BP_ND values determined by SRTM analysis of data at 0–120min and the target-to-reference ratio of radioactivity at 120 (112.5–127.5) min. Parameters were calculated using data at baseline and after pretreatment with aprepitant at doses of 0.03, 0.1, 0.3, 3, and 30mg/kg. The solid line indicates regression (y = 0.99x).

Figure 4.

Relationship between plasma aprepitant concentration and NK-1 receptor occupancy by apretitant in the gerbil striatum calculated using [¹⁸F]FE-SPA-RQ-PET data. The regression curve was generated using the following equation: receptor occupancy = maximal receptor occupancy × C_p/[C_p + EC₅₀]. The dot-dashed line represents the maximal receptor occupancy (85%), and the horizontal and vertical dashed lines denote the half-maximal occupancy (42.5%) and EC₅₀ (5.5ng/mL). respectively.

Figure 5.

Ev vivo analysis of NK-1 receptor occupancy by aprepitant in the gerbil striatum at different time points after oral administration. (A) Representative autoradiograms showing binding of [¹⁸F]FE-SPA-RQ to coronal sections of gerbil brains containing the striatum at 0.5, 1, 2, 4, 8, and 24h (left to right) after aprepitant administration at doses of 0.3, 3, and 30mg/kg (top to bottom). Radiolabeling of sections from untreated gerbils without (control) and with 10 µM aprepitant (non-specific binding) are displayed in the far right panel. (B) Striatal NK-1 receptor occupancy by aprepitant plotted against plasma concentrations of aprepitant in gerbils. The regression curve was generated as in the PET assays, and outlier data (indicated by circles), presumably in a non-equilibrium state, were excluded from the model fit.

Figure 6.

In vivo distribution of [¹⁸F]FE-SPA-RQ in the marmoset brain. (A) Coronal PET images containing the caudate/putamen (left) and cerebellum (right) at baseline (top) and after oral aprepitant administration at a dose of 20mg/kg (bottom). PET images were generated by averaging dynamic data from 0–180min after intravenous radioligand injection and overlaid on the MRI template. ROIs were placed on the caudate (arrows), putamen (arrowheads), and cerebellum (cyan lines). (B) Time–activity curves for [¹⁸F]FE-SPA-RQ in the caudate/putamen (red symbols) and cerebellum (black symbols) of the marmoset at baseline (closed symbols) and after aprepitant administration (open symbols). Radioligand uptake into each region was expressed as a percentage of the injected dose per unit tissue volume (%dose/mL).