(M0930-09-56) Excipient Screening of Long-Acting In Situ Forming Gels during Preclinical Formulation Development

Purpose: In situ forming gels are a very popular approach for administration of long-acting injectables (LAIs). Formulated as solutions containing an active pharmaceutical ingredient (API), these formulations form a semi-solid gel state once injected into the body to initiate the prolonged release behavior in situ.^1-2 Although a common method for LAI drug delivery, there is a lack of techniques developed to test these formulations before reaching the clinic. Animal models provide a less than perfect human alternative, and current in vitro technologies (e.g. USP dissolution) are optimized for bulk solution-based testing and fail to represent a biorelevant environment for in situ gel formulation testing. The SCISSOR (Subcutaneous Injection Site Simulator) instrument is an entirely in vitro tool that aids in mimicking the major physiological properties of the human subcutaneous tissue for more efficient formulation design, comparison, and risk assessment.^3-4 In this study we investigated different in situ forming gel formulations of the same API for rank order comparison of drug release correlated with visual analysis of gelling behavior within the SCISSOR platform.

Methods: Five blinded in situ forming gel formulations (Sample A, B, C, D and E) of the same API were tested using the SCISSOR system (Pion inc.) (Fig 1). The SCISSOR comprises a ‘cartridge’ filled with an artificial extracellular matrix (ECM) which simulates the subcutaneous environment ^3-4 The in situ forming gel was injected into the artificial ECM via standard hypodermic injection before swelling in the environment, which allowed API release through customized dialysis membranes on the cartridge. After liberation through the membranes, the API entered a larger outer chamber containing biorelevant bicarbonate buffer + 0.02% sodium azide, simulating uptake to systemic circulation. The assay timeframe was 1 week, and formulation comparison was investigated by analyzing these release curves, along with the images collected during experimentation. The concentration of API that reached the receiving chamber for the study was measured offline via HPLC analysis (Acuity Arc, Waters). Calibration curves were created as serial dilutions of 1mg/mL API in methanol over a concentration range of 0 – 327.33 µg/mL. Concentration values were then converted to %Release for comparative ranking. Data from SCISSOR turbidity sensors and cameras were also recorded to monitor injectate behavior throughout the experiments, however, only imaging data is displayed in this report.

Results: We tested 5 in situ forming gel solutions of the same small molecule API in the SCISSOR platform over 1 week. These formulations were long-acting injectables (LAIs) and were expected to gel and exhibit ‘swelling’ behavior upon injection and hydration. When injected, they were shown to display different release and stability characteristics (Figures 2 and 3) per formulation. Formulation B displayed the most representative ‘swelling’ behavior post-injection. In contrast, formulations A and E appeared to have separate solution components, as most of the bolus (with respect to apparent volume) diffused through the matrix, while some persisted at the injection site for the duration of the assay. Formulation C migrated initially before precipitating at the bottom of the cartridge, whereas formulation D precipitated immediately at the injection site (Figure 2). The respective release profiles generated for each sample complemented the information being shown via the imaging analysis with respect to the sample stability

Conclusion: The SCISSOR platform effectively differentiates in vitro performance of in situ forming gel systems, allowing for valuable early-stage insights into how excipient variations affect gelation and rate of API release. Using the release data alone, developers may select one of the ‘sustained’ release candidates, which have a persistent release of payload over the assay duration. However, with the addition of the imaging data collected from SCISSOR, we can hypothesize about mechanistic reasons for each response. For example, a curve may be slower due to the presence of solid precipitation upon or after injection, which could in turn increase the immunogenicity risk, making that candidate less suitable for further development despite favorable %Release data. These data can be used by formulation teams to make more informed and rational decisions earlier in the development process for in situ forming gels and other LAI formulations.

References: 1. Advancing Drug Delivery across Diverse Therapeutic Areas, Pharmaceuticals, 17 (4), 493 (2024).
2. Pandya, A. K., et al. in Situ Forming Depots for Long-Acting Drug Delivery Systems. Advanced Drug Delivery Reviews, 200, 115003 (2023).
3. Bown, H. K., et al. In Vitro Model for Predicting Bioavailability of Subcutaneously Injected Monoclonal Antibodies. Journal of Controlled Release, 273, 13–20 (2018).
4. Gomes, C., et al. Hydrogel Formats to Model Potential Drug Interactions Occurring at the Subcutaneous Injection Site. European Journal of Pharmaceutics and Biopharmaceutics, 199, 114308 (2024).

Figure 1. SCISSOR instrument in the top right-hand corner with a zoomed in Schematic of one chamber assembly showing the injection, cartridge and chamber.

Figure 2. Photographic data showing formulations A, B, C, D and E immediately-, 30 minutes-, 24 hours- and 1 week-post injection, and after the experiment was complete.

Figure 3. %Release data over 1 week for all five formulations.