Physiochemical and physiological mechanisms for the effects of food on drug absorption: the role of lipids and pH

Reduced systemic availability of an antiarrhythmic drug, bidisomide, with meal co-administration: relationship with region-dependent intestinal absorption

PURPOSE

The aim of this research was to determine the mechanism by which a co-administered meal decreases the oral absorption of bidisomide and does not influence the oral absorption of the chemically-related antiarrhythmic agent, disopyramide.

METHODS

Bidisomide plasma levels, following oral administration and intravenous infusion in the fasted state and with various meal treatments, were determined in human subjects. A dialysis technique was employed to examine the potential for drug binding to meal homogenates. Plasma levels, following drug administration through duodenal and jejunal intestinal access ports and following various meal treatments with oral drug co-administration, were compared for bidisomide and disopyramide in a canine model.

RESULTS

Bidisomide plasma AUC was significantly reduced following oral drug co-administration with breakfast compared to fasted-state controls in human subjects and in dogs independent of the composition of the solid cooked breakfast. While intravenous bidisomide infusion in human subjects showed a statistically significant reduction in AUC 15 minutes after oral administration of a high fat breakfast as compared to drug infusion in the fasted state, the reduction (-13%) was substantially smaller than the reduction (from -43% to -63%) observed with oral bidisomide meal co-administration. The percentages of bidisomide and disopyramide lost by binding to homogenates of cooked breakfast were 25.0 +/- 5.7% and 23.7 +/- 7.7%, respectively, as determined by dialysis at 4 hours. In dogs, the extent of absorption of disopyramide was comparable from oral, duodenal and mid-jejunal administration while the extent of bidisomide absorption from mid-jejunal administration was significantly lower than for oral or duodenal administration. Non-viscous liquid meals decreased Cmax but not AUC, while viscous homogenized solid meals decreased both Cmax and AUC for bidisomide with oral drug-meal co-administration. Oral non-caloric hydroxypropyl methylcellulose meals decreased bidisomide to the same extent as homogenized solid meals but did not lower disopyramide AUC.

CONCLUSIONS

The significant reduction in bidisomide plasma levels observed with meal co-administration in human subjects was predominantly mediated through a reduction in drug absorption and was independent of solid meal composition. The difference in meal effect on the absorption of the two drugs in humans did not appear to be a function of drug binding to cooked meal components over typical human upper gastrointestinal residence times. In dogs, the high-viscosity medium generated by oral co-administration of a solid meal reduced the upper intestinal absorption of bidisomide and disopyramide. Bidisomide AUC was decreased since it was well absorbed in the upper but not lower small intestine. Disopyramide AUC was not significantly affected since it was well absorbed from both regions. A similar mechanism may play a role in drug plasma level reductions following oral co-administration with solid meals for drugs showing similar regionally-dependent absorption profiles.

Ionization and solubilization of 4 alkyl benzoic acids and 4 alkyl anilines in sodium taurodeoxycholate solutions

PURPOSE

The aqueous solubility and the extent of solubilization and ionization constant in sodium taurodeoxycholate (NaTDC) solutions of a series of benzoic acid and aniline derivatives were measured as a basis to characterize and thereby help predict the nature of the interaction of drugs with bile aggregates.

METHODS

The aqueous solubility and the solubilization of two series of compounds, 4-alkyl benzoic acids and 4-alkyl anilines, was measured as a function of NaTDC in 0 and 150 mM NaCl. The ionization constants were determined in water and in 50 mM NaTDC at sodium chloride concentration of 0, 75 and 150 mM by spectrophotometric titration. The diffusion coefficients of NaTDC and the solutes were measured by pulsed-field gradient spin echo NMR spectroscopy.

RESULTS

The aqueous solubilities decreased with increasing alkyl chain length in both series, and the aniline derivatives had larger solubilities than the benzoic acid derivatives. The number of moles of solute solubilized per mole of bile salt ranged from 0.17 to 0.31 for the benzoic acid derivatives and from 1.3 to 3.0 for the aniline derivatives. The pKa values of the benzoic acid derivatives in the presence of NaTDC were higher relative to the controls, and the difference in the pKa (delta pKa,obs) increased with increasing chain length. With the aniline derivatives, the pKa values were also shifted to higher values in NaTDC relative to the control but only in the absence of salt. The presence of the solute caused a decrease in the diffusion coefficient of NaTDC, and the diffusion coefficients of the solutes decreased with increasing alkyl chain length. With the hexyl derivative, the diffusion coefficient of the solute was smaller than the diffusion coefficient of the bile salt. The chemical shift of the protons attached to carbon 18 and 19 of the salt were decreased to a greater extent in the presence of the solutes than the protons attached to carbon 26.

CONCLUSION

Both the solubilization and ionization behavior of solutes were affected by the presence of bile salt aggregates. The surface potential and effective polarity of NaTDC aggregates were found to be dependent on the alkyl chain length for these two homologous series of solutes. The solubilization ratio was largely independent of alkyl chain length, but the unitary partition coefficient was dependent on both alkyl chain length as well as ionization state. The derivatives reduced the diffusivity of the micelles suggesting the formation of larger size aggregates and the solutes (hexyl derivatives) appear to favor association with the larger sized aggregates. The phenyl ring of the solutes appears to be oriented parallel to the plane of the steroid frame with preferential positioning near the hydrophobic rings.