A polycaprolactone nanoparticle formulation of cyclosporin-a improves the prediction of area under the curve using a limited sampling strategy

Therapeutic monitoring of Cyclosporine (CyA) by using area under the curve (AUC) from abbreviated kinetic profiles is of recent trend in clinical prac tice due to the potential improvement in transplant and clinical outcome wi th costs reduction in mind. Several papers describe successful use of the l imited sampling strategy to predict AUCs in different transplant population s when treated with Sandimmun or Sandimmun Neoral. However, the same predic tive potential is achieved for the latter formulation with lesser effort. T he present paper describes the application of the limited sampling strategi es to demonstrate the advantages of using CyA incorporated in polymeric nan oparticles (CyA-NP) as compared to two reference Sandimmun formulations whi ch consisted of an emulsion of the oily solution in milk (SIM-EM) and a mic roemulsion (SIM-Neoral) formerly tried on rats. Two independent data batche s were used: group I which included 36, 31 and 10 animals receiving SIM-EM, CyA-NP and SIM-Neoral, respectively, and group 2 made of nine and eight ra ts treated with SIM-EM and CyA-NP. Several limited sampling equations were derived for each formulation from group I by stepwise multiple linear regre ssion. Statistical analysis disclosed that CyA concentrations 8 and 32 h af ter dose administration vouched for 88 and 69% variability in AUC (0-48 h) for CyA-NP and SIM-EM, respectively When summed up, these two concentration s revealed nearly 97% of AUC (0-48 h) variability. CyA concentrations 8 h p ost-treatment with SIM-Neoral explained 89% variability in AUC (0-48 h). Th is value raised to 98% when a second CyA concentration (24 h) was introduce d. The equations derived from group I were then employed to predict AUCs in group 2. CyA brood levels at 8 h post-treatment confirmed AUC for CyA-NP ( r(2) = 0.98) to be very precise and unbiased (error = 1.46%, interval - 16. 2 to 21.33%), while the results for SIM-EM obtained with the CyA concentrat ion at 32 h were r(2) = 0.93 plus error = 5.71%, interval - 44.33 to 105.94 %. Similar results were obtained when the study period was reduced to 24 h. The use of these limited sampling models manifested the coincidence betwee n CyA-NP and SIM-Neoral as well as the advantages of both formulations over SIM-EM when it comes to CyA monitoring. (C) 1999 Elsevier Science B.V. All rights reserved.