(T1330-03-21) Quantitating the Contributions of the Particle Drift Effect to In Vitro FLUX and the Application of Particle Drift Flux to Predictive Modelling

Purpose: The particle drift effect (PDE) has been previously cited¹ as an explanation for the apparent in-vivo dose dependency for some solubility/permeability limited APIs which exhibit a predominant unstirred water layer (UWL) limitation to their absorption. Celecox, a formulated product of the API celecoxib (CEL), is assayed for flux in this study to investigate the influence of in-vitro dosage on the observed flux when the donor compartment is loaded beyond the solubility limit of the API. The applicability of flux values for assessing the in-vivo influence of the particle drift effect in predictive modeling is also explored.

Methods: Flux assays were carried out using the MicroFLUX^TM two-compartment dissolution/permeability testing apparatus (Pion Inc., Billerica, MA, USA). Quantitation of the API of each formulation was performed by a Rainbow R6 FO spectrometer. Predictions of in-vivo Fa% were generated using the fraction absorbed calculation of the Pion Predictor^TM software package, using the ‘PDE modeling’ function, and the flux data as input. Flux assays were performed at least in triplicate using pH 6.5 FaSSIF as the donor medium, at a variety of API donor loadings above the API solubility. Donor loadings were scaled to match the dose/volume ratio of the comparison set of in-vivo CEL bioavailability data, presuming a human intestinal volume of 150 mL.

Results: The PDE postulates that the apparent permeability of an API can increase when solid particles of the API diffuse and dissolve into the UWL between bulk media and a membrane barrier, increasing the API concentration in the UWL and yielding greater permeation. This UWL concentration increase results in a reduction of the concentration gradient² between the bulk of solution and the UWL, which plateaus when the API solubility is reached in the UWL. The results obtained from the flux assays exhibited increasing flux beyond the solubility limitation of CEL, and showed a dose dependency on the flux values, where flux increased proportionally to the load of undissolved material. Figure 1 shows a linear plot of the measured flux values against the undissolved CEL API concentration. The slope of the best fit line of the flux vs undissolved dose concentration data was applied in the PDE modeling function of the Predictor program, to upscale the observed in-vitro particle drift behavior and estimate the proportion of the in-vivo flux arising due to particle drifting for a given amount of undissolved material in the intestinal lumen. The results of the Predictor model are shown in figure 2 for Celecox at a 50 mg dose. Without correcting for the influence of particle drift on the in-vivo flux, the overall absolute fraction absorbed is underestimated by approximately 1.7-fold. The results of the Predictor fraction absorbed model for all tested dose levels are presented in figure 3, including the Predictor Fa% output with and without correction for the particle drift effect, and in-vivo human Fa% data³ for Celecox. A notably improved correlation to the in-vivo data is observed for the fraction absorbed predictions when the particle drift effect is accounted for, especially at higher loadings where the effect is more significant.

Conclusion: The data presented here illustrates that for solubility/permeability limited APIs, with UWL limited absorption, predictions of in-vivo absorption based on in-vitro flux results are susceptible to underestimation in situations where intestinal absorption is driven in part by small particles, causing an effective increase in the UWL permeability of the drug. In-vitro flux measurements are a capable tool in the assessment of the influence of particle drifting on measured flux values, and in-vitro observations of particle drifting have applicability in the correction of predicted in-vivo behavior for this phenomenon.

References: 1. K. Sugano, Possible reduction of effective thickness of intestinal unstirred water layer by particle drifting effect, International Journal of Pharmaceutics 387 (2010) 103–109.
2. Siddhi S. Hate, Laura I. Mosquera-Giraldo, Lynne S. Taylor, A Mechanistic Study of Drug Mass Transport from Supersaturated Solutions Across PAMPA Membranes, Journal of Pharmaceutical Sciences 111 (2022) 102−115.
3. Sugano K. Biopharmaceutics Modelling and Simulations p266-321 (Aug 2012)

Figure 1. Plot of the measured flux values for CEL formulation at different dose levels, vs the concentration of the undissolved dose. A linear increase in flux is observed and the undissolved dose increases.

Figure 2. Predictor v1.0 absolute fraction absorbed vs time predictions for Celecox at a 50 mg in-vivo dose, with and without particle drift effect correction (PDE) applied.

Figure 3. Comparison of Predictor v1.0 fraction absorbed prediction results with and without particle drift correction, and the human in-vivo fraction absorbed for Celecox at different dose levels.