Pharmacokinetic Analyses
All subjects who received both study treatments, completed scheduled posttreatment PK evaluations, and were protocol compliant were included in the PK analysis set. Subjects who vomited within twice the median Tmax were excluded from the primary analysis.
Paclitaxel plasma concentrations were normalized to 615mg/m2 for oPac+E and 80 mg/m2 for IV paclitaxel. Pharmacokinetic and statistical analyses were based on normalized plasma concentrations. Plasma concentrations for paclitaxel were analysed to determine the following PK parameters by noncompartmental analysis using plasma concentration time data for oral and IV paclitaxel: AUC0-∞ (primary endpoint) as well as Cmax, AUC0-t, Tmax, and t½ (secondary endpoints).
The equivalence of the extent of exposure was determined by comparing the AUC0-∞ of the selected dose of oral paclitaxel (as oPac+E) (administered over 3 consecutive days) to the AUC0-∞ of IV paclitaxel.
The primary PK parameters were compared between IV paclitaxel (reference) and oral paclitaxel (test) formulations. Analysis of variance (ANOVA) was performed (α=0.05) on the untransformed and log10 transformed PK parameters Cmax, AUC0-t, and AUC0-∞ for paclitaxel. The ANOVA model included sequence, subjects nested within the sequence, period, and formulation as factors. The significance of the sequence effect was tested using the subjects nested within the sequence as the error term. Two-sided 90% CIs for the log transformed ratio of test/reference of the least squares means obtained from the ANOVA for Cmax, AUC0-t, and AUC0-∞ were estimated. Equivalence was to be concluded if the 90% CI of the ratio (oral paclitaxel 615mg/m2 over 3 consecutive days [oPac+E] / 80 mg/m2 IV paclitaxel) of the least square means from the ANOVA of the log-transformed AUC0-∞ was within the 80% to 125% acceptance range.