(T1230-03-18) Development of a UV Imaging Method for Characterization of Intra-Tumoral Injectables Using cAMP as Model Substance

Purpose: Cyclic dinucleotides (CDNs) are stimulator of interferon genes (STING) agonists that lead to an increased inflammatory response against pathogens and tumors [1]. CDNs have recently attracted attention due to the implication of STING in tumors. Intra-tumoral administration of CDNs is necessary to achieve a high and effective concentration. As of today, there are no compendial in vitro release methods for conventional- or controlled-release parenteral formulations [2]. Thus, there is a need for an in vitro method capable of monitoring drug release from parenteral CDN formulations. In the present study, cyclic AMP (cAMP) was used as a model compound for CDNs. The aim of the study was to develop a UV imaging in vitro release method suitable for screening nucleotide formulations. Accordingly, the release of cAMP from various parenteral formulations was determined and the effect of the depot geometry on the release was investigated using a hydrogel-based UV imaging release testing platform.

Methods: The release of cAMP from different parenteral formulations using the SDi2 (UV imaging instrument, Pion Inc., Bilerica, MA, USA) was compared and the effect of formulation geometry was also investigated in the 0.5% (w/v) agarose gel release matrix. Formulations were injected into a premade cavity of the hydrogel matrix. The following formulations containing cAMP were tested, an aqueous solution (Aq. Sol), 20% (w/w) Pluronic F127, 40% (w/w) PLGA (lactide:glycolide molar ratio of 75:25, MW 5-20 kDa) in N-methyl-2-pyrrolidone (NMP), and 33% (w/w) PLGA (lactide:glycolide molar ratio of 50:50, MW 24-38 kDa) in NMP.

Results: A prototype UV imaging release testing setup for the SDi2 imaging system was developed. Image analysis was conducted using in-house developed MATLAB scripts and provided absorbance maps with contour lines facilitating visualization (Fig. 1) and continuous data analysis capability. A sustained release was obtained from the 33% (w/w) PLGA (50:50) formulation, while the other formulations exhibited relatively similar initial release rates (Fig. 2). The slower release rate from the 33% (w/w) PLGA (50:50) formulation was attributed to its high viscosity and depot formation (Fig. 1). The prototype setup was effective in visualizing the effects of formulation geometry, but the release profiles did not change significantly (Fig. 3). The cylindrical cavity was the preferred default configuration, as it ensured symmetric diffusion for early time-points. The geometry could be optimized for selected formulations or injection sites. Relatively thin elongated shapes may be of particular relevance for intra-muscular injection in order to mimic formulation spread between muscle fibers [3].

Conclusion: A prototype UV imaging release setup for parenteral formulations was developed. MATLAB data analysis facilitated visualization of formulation behavior and continuous data analysis. The release of cAMP from different formulations was assessed using UV imaging. The aqueous solution, the in situ forming Pluronic F127 hydrogel, and the in situ forming PLGA (75:25) implant exhibited relatively similar initial cAMP release rates. More sustained cAMP release was obtained from the in situ forming PLGA (50:50) implant. Furthermore, the ability to control depot geometry was shown. The UV imaging release setup constitutes a significant technical development in terms of visualizing and characterizing parenteral formulation behavior upon introduction into a biomimetic matrix.

References: 1 Shae, D. et al. Nat Nanotechnol (2019) 14: 269-278.
2. Larsen, SW. et al. J Drug Deliv Sci Technol (2013) 23: 315-324.
3. Kalicharan, R.W. et al. Int. J. Pharm. 2016, 505, 52–60.
4. Bock, F. et al. IJMS. 2022, 23(7):3599.

Acknowledgments: The project is funded by Bristol Myers Squibb Company.

Fig. 1. Representative absorbance images showing the release of cAMP into an agarose gel (280 nm) or PLGA depot formation (520 nm) at selected time-points and 37 °C. The investigated formulations included an aqueous solution, an in situ forming Pluronic F127 hydrogel, a PLGA (75:25) implant, and a PLGA (50:50) implant. Contour lines at 280 nm represent absorbance values of 0.25, 0.5, 0.75, and 1.



Fig. 2. Apparent release profiles for the aqueous solution (blue), the in situ forming Pluronic F127 hydrogel (black), the PLGA (75:25) implant (green), and the PLGA (50:50) implant (red) at 37 °C. Average of 3 replicates are in bold, and error bars (1 SD) are represented by the dotted lines [4].



Fig. 3. Representative absorbance images showing the release of cAMP (280 nm) into an agarose gel and PLGA (75:25) depot formation (520 nm) at selected time-points and 37 °C. The investigated geometries included a cylinder (3.1 × 4 mm^2 (r × H)), a square cuboid (5.5 × 5.5 × 4 mm^3 (L × W × H)), and a rectangular cuboid (12 × 2.5 × 4 mm^3 (L × W × H)). Contour lines at 280 nm represent absorbance values of 0.25, 0.5, 0.75, and 1.