Predictive ability of level A in vitro-in vivo correlation for RingCap controlled-release acetaminophen tablets

Purpose. The goal of this study was to establish and validate an in vitro-in vivo correlation (IVIVC) for two sustained-release formulations (i.e., a matrix tablet and a RingCap banded matrix tablet) containing 750 mg of acetaminophen. Methods. The in vitro dissolution and in vivo disposition of these formulations were examined by using a USP type III dissolution apparatus and a single-dose, three-way, crossover study that included an immediate-release acetaminophen dosage form, respectively. An IVIVC was established by using the mean fraction dissolved (FD) and mean fraction absorbed (FA) and used to simulate the plasma concentation-time profile of acetaminophen after administration of the matrix tablet (i.e., internal validation) and RingCap banded matrix tablet (i.e., external validation). Results. A statistically significant relationship (r(2) = 0.997, P < 0.001) existed between the FD and FA for matrix tablets and was best described by the equation (FA) = 0.984 x (FD) + 0.0133. The percent predictions errors in CMAX and AUCL were <10% when predicting the plasma concentration-time profiles for the two formulations, validating the internal and external predictability of the IVIVC. Conclusions. The data (i) show that in vitro dissolution data are a good predictor of in vivo fraction absorbed for acetaminophen, (ii) support the general use of in vitro dissolution data for readily soluble and readily absorbed drugs, (iii) suggest that acetaminophen may serve as a model drug for evaluating novel sustained-release delivery systems, and (iv) provide a tangible example of the limitations of current methods for predicting and validating IVIVC.