Partitioning of halofantrine hydrochloride between water, micellar solutions, and soybean oil: Effects on its apparent ionization constant

Abstract Recent studies in a conscious dog model demonstrated intestinal lymphatic transport to be a significant contributor to the bioavailability of the highly lipid‐soluble free‐base of halofantrine (Hf), and surprisingly, also the poorly lipid‐soluble hydrochloride salt (Hf · HCl). Partial conversion of solubilized Hf · HCl to Hf base within the intestinal lumen prior to the lymphatic uptake seemed to be the most likely explanation for these results. This hypothesis was supported by studies exploring the partitioning behavior of Hf · HCl between soybean oil (SBO) and aqueous micellar solutions containing different ionic and nonionic surfactants. Mixed micelles prepared from sodium taurodeoxycholate (NaTC) and lecithin (PC) were chosen to represent fed‐state intestinal fluids. The apparent ionization constants derived from the partitioning versus pH profiles showed marked shifts when compared with the likely aqueous p K a value. In the present paper, the apparent p K a values of Hf in aqueous micellar phases, without a coexisting oil phase, were investigated to further probe the mechanisms underlying the effect of micellar media on the apparent ionization equilibrium, and subsequently, on its partitioning behavior in the triphasic systems. Another aim of this study was to further evaluate the aqueous p K a value of Hf. The results indicate that the aqueous p K a of Hf is most probably in the range ∼8–9, and that the ionization equilibrium is highly dependent on the solution environment. For example, marked p K a shifts of several units were observed for Hf in the presence of different micellar species and SBO. The apparent ionization equilibrium depends not only on interaction of Hf with the micelles, but also on its partitioning into the oil phase. © 2003 Wiley‐Liss, Inc. and the American Pharmacists Association J Pharm Sci 92:2217–2228, 2003