Research ReportInfluence of time to achieve substrate distribution equilibrium between brain tissue and blood on quantitation of the blood-brain barrier P-glycoprotein effect

Active efflux transport processes at the blood-brain barrier (BBB), such as P-glycoprotein (P-gp)-mediated efflux, can limit brain uptake of therapeutics. Accurate determination of the consequent impact on brain uptake is assumed to require sampling post-attainment of brain-to-blood distribution equilibrium. Because this approach is not always feasible, understanding the relationship between apparent degree of efflux (e.g., calculated BBB P-gp effect) and the fraction of time remaining until distribution equilibrium is achieved (FTDE) would be advantageous. This study employed simulation strategies to explore this relationship in the simplest relevant system (absence of protein binding, saturable uptake, or metabolism at the BBB). Concentration-time profiles were simulated with a 4-compartment system (blood, peripheral tissues, BBB endothelium and brain parenchyma). A unidirectional endothelium-to-blood rate constant, PSe, represented P-gp-mediated efflux. A parameter space was selected to simulate an 18-fold P-gp effect, (Kp,brain at distribution equilibrium in the absence [Kp,brain = 82] vs. presence [Kp,brain = 4.5] of P-gp-mediated flux), as observed for paclitaxel in P-gp-deficient vs. P-gp-competent mice. Hypothetical compounds with different P-gp effects, peripheral compartment distribution kinetics, or times to achieve distribution equilibrium were simulated by perturbing the values of relevant model parameters. P-gp effects calculated prior to attainment of distribution equilibrium may be substantially erroneous. However, reasonably accurate estimates can be obtained relatively early in the net distributional phase (under 20% error at FTDE > 0.36 or 0.11 for bolus or infusion administration, respectively). Potential errors associated with non-equilibrium calculations are dependent on both P-gp-mediated and P-gp-independent components of flux across the BBB.

► Kinetics of brain partitioning were modeled with a simple system (blood, brain, a barrier, and a peripheral compartment). ► The calculated P-gp effect (and associated error) change most rapidly at early times after substrate administration. ► P-gp effect differs between bolus and infusion input. Guidelines on experimental design are developed for each. ► Error in calculating the P-gp effect is determined by fractional approach to blood-brain distribution equilibrium. ► The kinetics of brain partitioning are influenced by substrate distribution into a peripheral (non-brain) compartment.