González-Álvarez I, Fernández-Teruel C, Navarro-Fontestad MC, Ruiz-Garcia A., Bermejo M, Casabó VG.
Dpto Farmacia y Tecnologia Farmaceutica. Universidad de Valencia
Purpose: A method of deconvolution by curve fitting is presented and applied to the estimation of oral absolute bioavailability of a ciprofloxacin derivative (CNV97101). The aim of the study was to explain the low oral bioavailabity of CNV97101. In the study three different extrabasal routes has been used: oral, intraduodenal and intraperitoneal. Furthermore an in vitro study with the stomach content of fasted rats for 12/24 h was also developed.
Method: The concentration versus time plasma levels were obtained by administration of CNV97101 by four different routes: intravenous and the extrabasal routes described before. CNV97101 was dosed at 30mg/Kg, with two additional doses (oral and intravenous) at 15 mg/Kg. The same oral doses were also used to do the in vitro studies. Fitting procedures were performed using a non-linear mixed effect model in NONMEM assuming exponential models for intra and interindividual variability. The absorption phase was modeled considering a passive diffusion with an initial fraction of dose precipitated which was re-dissolved by a zero order process, and an absorption window which limited the absorption time. The fraction precipitated was fixed using the experimental results obtained from the in vitro experiment.
Results: The in vitro results showed the non precipitated fractions were 25% for 30 mg/Kg and 45% for 15 mg/Kg. However values for oral bioavailability were 49% and 58% respectively, furthermore a high fraction is re-dissolved and absorbed. In the case of intraduodenal administration, NONMEM estimates around 40% of the dose precipitates, but its final bioavailability is 93%.
Conclusion: The in vivo studies were developed with rats fasted for 12h, and as the in vitro study demonstrates this is the cause of the low oral bioavailabilty. However, in vitro studies with 24 h fasted rats showed a precipitation lower than 10%. So, the food interacts with CNV97101 making more difficult its absorption. As the proposed model indicates, CNV97101 is re-dissolved and absorbed for 1 hour along the small intestine.
Reference: PAGE 14 (2005) Abstr 803 [www.page-meeting.org/?abstract=803]
Poster: poster