In vitro and in vivo evaluations of biodegradable implants for hormone replacement therapy: effect of system design and PK-PD relationship

This investigation evaluated the feasibility of using subdermally implantable devices fabricated by nonconventional 3-dimensional printing technology for controlled delivery of ethinyl estradiol (EE2). In vitro release kinetics of EE2 and in vivo pharmacokinetics/pharmacodynamics in ovariectomized New Zealand White rabbits were carried out to study 3 implant prototypes: implant I (single-channel EE2 distribution in polycaprolactone polymer core), implant II (homogeneous EE2 distribution in polycaprolactone polymer matrix), and implant III (concentration-gradient EE2 distribution in polycaprolactone and poly(dl-lactide-co-glycolide) (50:50 matrix). EE2 was found to be released from all the implants in a nonlinear pattern with an order of implant III > implant II > implant I. The noncompartmental pharmacokinetic analysis of plasma EE2 profiles in rabbits indicated a significant difference (p < .05) in Cmax, tmax, and mean residence time between implant I and implants II and III, but no difference in the area under the plasma concentration time curves calculated by trapezoidal rule (AUC) among the implants. For pharmacodynamic studies, endogenous follicle-stimulating hormone (FSH) and luteinizing hormone (LH) levels were observed to be suppressed following implantation of all implants, which demonstrated that a therapeutically effective dose of EE2 had been delivered. Furthermore, the noncompartmental analysis of plasma FSH and LH profiles in rabbits showed a significant difference (p < .05) in AUC and the mean residence time between implant III and implants I and II. A good in vivo/in vitro relationship was observed between daily amounts of EE2 released and plasma profiles of EE2 for all implants. This relationship suggests that plasma profiles of EE2 could be predicted from in vitro measurement of daily amount of EE2 released. Therefore, performing in vitro drug release studies may aid in the development of an EE2 implant with the desired in vivo release rate.