BioJect: An in vitro platform to explore release dynamics of peptides in subcutaneous drug delivery

Abstract

Predicting the release performance of subcutaneous (SC) drug formulations is challenging due to the complex interplay between physicochemical properties and the physiological microenvironment, which includes the extracellular matrix (ECM), fluid composition, and fluid availability, factors that collectively influence bioavailability and absorption rates. The ECM often acts as a bandpass filter modulated by local ion and protein content. In this study, we introduce the BioJect cell, a modern release test method based on the compendial flow-through cell, integrating a perfusion system with customizable biomatrix components. We systematically investigated the release mechanisms of four insulin formulations: regular human insulin, insulin aspart, insulin glulisine, and Neutral Protamine Hagedorn (NPH) insulin. A modified simulated subcutaneous interstitial fluid (mSSIF) comprising multiple components of the SC physiological environment was employed. It incorporates important ions and proteins (138.5 mM sodium, 10 mM potassium, 1.8 mM calcium, 0.8 mM magnesium, 111.3 mM chloride, 28 mM bicarbonate, 0.5 mM sulfate, 5 mM acetate, 4.2 mM phosphate, 30 g/L total protein added as bovine serum albumin). Our release test method discriminated the tested formulations under varying biorelevant conditions, demonstrating its biopredictive capabilities. Notably, we discovered a previously undocumented albumin binding affecting the release rate of insulin glulisine, likely occurring in the low-shear environment of SC tissue only. Additionally, the inclusion of biorelevant components like hyaluronic acid and collagen into the biomatrix of the BioJect cell provided profound insights into potential absorption and release mechanisms, supported by two in vitro-in vivo relationships (level C and level A). The BioJect cell represents a significant advancement in simulating the SC environment for drug release testing. Our findings highlight the importance of considering protein binding and ECM components in predicting drug absorption, offering a promising tool for the development and optimization of SC formulations.