Non-invasive screening method for simultaneous evaluation of in vivo growth factor release profiles from multiple ectopic bone tissue engineering implants

Abstract

The purpose of this study was to develop and validate a screening method based on scintillation probes for the simultaneous evaluation of in vivo growth factor release profiles of multiple implants in the same animal. First, we characterized the scintillation probes in a series of in vitro experiments to optimize the accuracy of the measurement setup. The scintillation probes were found to have a strong geometric dependence and experience saturation effects at high activities. In vitro simulation of 4 subcutaneous limb implants in a rat showed minimal interference of surrounding implants on local measurements at close to parallel positioning of the probes. These characteristics were taken into consideration for the design of the probe setup and in vivo experiment. The measurement setup was then validated in a rat subcutaneous implantation model using 4 different sustained release carriers loaded with (125)I-BMP-2 per animal. The implants were removed after 42 or 84 days of implantation, for comparison of the non-invasive method to ex vivo radioisotope counting. The non-invasive method demonstrated a good correlation with the ex vivo counting method at both time-points of all 4 carriers. Overall, this study showed that scintillation probes could be successfully used for paired measurement of 4 release profiles with minimal interference of the surrounding implants, and may find use as non-invasive screening tools for various drug delivery applications.

Figure 1

Schematic diagram of the non-invasive measurement method in the rat. The active portion of the measurement setup consisted of scintillation probes (A) containing a sodium iodide scintillator (B) were collimated with a hollow tube (C) and wrapped in leaded tape to determine the activity of subcutaneous implanted release vehicles (D) over time. The implant-toprobe distance (1), the detector linearity (2) and interference of multiple sources at different probe angles (3) were characterized prior to the in vivo experiment.

Figure 2

In vitro characterization of important parameters influencing the accuracy of the probe setup. The investigated parameters were (A) the relation between probe-to-source distance and geometric efficiency, (B) the counting rate linearity of the individual probes, and (C) the interference of a surrounding source at different angular deviations from parallel probe positioning. The counting rate linearity and the interference of a surrounding source were determined at a distance of 3 cm between the source and the detector.

Figure 3

¹²⁵I-BMP-2 release profiles from four different implants obtained by the non-invasive measurement method. The implants consisted of (A) BMP-2 loaded gelatin hydrogels, (B) BMP-2 loaded microspheres in a gelatin hydrogel, (C) BMP-2 loaded microspheres in a PPF scaffold and (D) BMP-2 loaded microspheres in a PPF scaffold surrounded by a gelatin hydrogel.

Figure 4

Average ¹²⁵I-BMP-2 retention in the 4 implants after 6 and 12 weeks of implantation measured using the non-invasive and ex-vivo method (A) and the correlation between the two methods (B). The implants consisted of (A) BMP-2 loaded gelatin hydrogels, (B) BMP-2 loaded microspheres in a gelatin hydrogel, (C) BMP-2 loaded microspheres in a PPF scaffold and (D) BMP-2 loaded microspheres in a PPF scaffold surrounded by a gelatin hydrogel.

Figure 5

¹²⁵I counts in the blood (A) and in 24 hour urine and feces excretions (B) after ¹²⁵IBMP- 2 release from the 4 subcutaneously placed implants.