Sorption of Neuropsychopharmaca in Microfluidic Materials for In Vitro Studies

Sorption (i.e., adsorption and absorption) of small-molecule compounds to polydimethylsiloxane (PDMS) is a widely acknowledged phenomenon. However, studies to date have largely been conducted under atypical conditions for microfluidic applications (lack of perfusion, lack of biological fluids, etc.), especially considering biological studies such as organs-on-chips where small-molecule sorption poses the largest concern. Here, we present an in-depth study of small-molecule sorption under relevant conditions for microphysiological systems, focusing on a standard geometry for biological barrier studies that find application in pharmacokinetics. We specifically assess the sorption of a broad compound panel including 15 neuropsychopharmaca at in vivo concentration levels. We consider devices constructed from PDMS as well as two material alternatives (off-stoichiometry thiol-ene-epoxy, or tape/polycarbonate laminates). Moreover, we study the much neglected impact of peristaltic pump tubing, an essential component of the recirculating systems required to achieve in vivo-like perfusion shear stresses. We find that the choice of the device material does not have a significant impact on the sorption behavior in our barrier-on-chip-type system. Our PDMS observations in particular suggest that excessive compound sorption observed in prior studies is not sufficiently described by compound hydrophobicity or other suggested predictors. Critically, we show that sorption by peristaltic tubing, including the commonly utilized PharMed BPT, dominates over device sorption even on an area-normalized basis, let alone at the typically much larger tubing surface areas. Our findings highlight the importance of validating compound dosages in organ-on-chip studies, as well as the need for considering tubing materials with equal or higher care than device materials.