Feasibility of in-vitro/in-vivo correlation in the case of a new sustained-release theophylline pellet formulation.

This contribution attempts to relate the in-vitro drug release properties of theophylline from Euphylong pellets to serum theophylline concentration/time curves. Comparisons are made according to methods presented in recent publications (e.g. using mean residence time) and by juxtaposition of analogous values. For this, the mutual relationship of equal release times (absorption) and % amounts released or absorbed at fixed times are evaluated. The in-vitro data required for the computational evaluations were obtained using a paddle apparatus (according to USP XXI) at 100 rpm, buffer pH 7.4. The feasibility of converting the in-vitro to the in-vivo curve by means of a chronological transformation is examined, as well as in-vitro/in-vivo dependence on numerically fitted model parameters. Finally, in an attempt to imitate the in-vivo curve profile, the apparatus parameters for the in-vitro tests were subjected to extreme variation. The effect of the presence or absence of sink conditions is discussed and evaluated. In the concluding overview, the feasibility of predicting the in vivo behaviour of Euphylong pellets on the basis of current so-called "in-vitro/in-vivo correlation" procedures is assessed.