Effects of absorption-modifying excipients on jejunal drug absorption in simulated fasted and fed luminal conditions

Abstract

Oral administration of drug products is the preferred administration route. In recent decades there has been an increase in drug candidates with low solubility and/or low permeability. To increase the possibility of oral administration for the poorly permeating drugs, the use of absorption modifying excipients (AMEs) has been proposed. These types of AMEs may also affect the regulatory assessment of a novel drug delivery system if they affect the absorption of a drug from any of the four BCS classes. The effects of AMEs have previously been investigated in various animal models, including the single-pass intestinal perfusion (SPIP) in rats. To further improve the biorelevance and the in vivo predictiveness of the SPIP model, four compounds (atenolol, enalaprilat, ketoprofen, metoprolol) were perfused in fasted or fed state simulated intestinal fluid (FaSSIF or FeSSIF) together with the AMEs N-acetyl-cysteine, caprate, or sodium dodecyl sulfate. For the highly soluble and poorly permeating compounds enalaprilat and atenolol (BCS class III), the flux was increased the most by the addition of SDS in both FaSSIF and FeSSIF. For ketoprofen (BCS class II), the flux decreased in the presence of all AMEs in at least one of the perfusion media. The flux of metoprolol (BCS class I) was not affected by any of the excipients in none of simulated prandial states. The changes in magnitude in the absorption of the compounds were in general smaller in FeSSIF than in FaSSIF. This may be explained by a reduced free concentration AMEs in FeSSIF. Further, the results in FeSSIF were similar to those from intrajejunal bolus administration in rat in a previous study. This suggests that the biorelevance of the SPIP method may be increased when investigating the effects of AMEs, by the addition of intraluminal constituents representative to fasted and/or fed state to the inlet perfusate.

Introduction

Active pharmaceutical ingredients (APIs) with a low molecular mass (approximately smaller than 500 Da) are preferably given orally to patients for convenience and compliance [1]. However, there are an increasing numbers of APIs with less than optimal biopharmaceutical properties, extending beyond Lipinskis “rule-of-5” [2], [3], [4]. A large proportion of these molecules have low solubility and/or low permeability and belong to classes II-IV of the biopharmaceutics classification system (BCS) [5], [6]. There are few possible biopharmaceutical strategies for drugs with low and variable intestinal permeation. The most common approach includes changing the molecular structure, for instance by masking polar properties, which may also change the pharmacokinetics and the pharmacological potency of the drug. Clearly, it would be more desirable to increase the potential for intestinal permeation of the API. Intestinal permeability may be increased by using pharmaceutical excipients that increase the transport over the enterocytes [7]. These excipients are usually labelled absorption modifying excipients (AMEs) or permeation enhancers (PEs) [7], [8]. The two most common mechanisms of action for AMEs are either altering the fluidity/integrity of the apical membrane of the enterocytes, which facilitates passive transcellular diffusion, or increasing the space of the tight junctions, thus increasing the paracellular diffusion [9]. The main concerns with AMEs are safety, because a reduced barrier integrity may facilitate bacterial and/or viral translocation, and potentially increase the unwanted absorption of xenobiotics, pathogens and dietary antigens [9], [10]. A dysregulation of the intestinal epithelial cell barrier may also contribute to pathogenesis, such as inflammatory bowel disease [11].

We have previously investigated the effect of AMEs on jejunal absorption of a set of APIs (Table 1) with different physicochemical properties in buffer using the rat single-pass intestinal perfusion (SPIP) model [12], [13], [14]. The SPIP model has been shown to predict human intestinal permeability and fraction dose absorbed with a high accuracy [15], [16], [17]. The AMEs and their concentrations in the perfusion media were selected based on clinical and pre-clinical experiences 9, 14, 18. The appearance drug flux (J_app) in plasma for the poorly permeating compounds investigated was substantially increased by the AMEs’ effects on the mucosal barrier, with a concentration-dependent enhancement of the transcellular routes. It was also found that the SPIP model was less predictive of the in vivo effect of AMEs from bolus dosing into the proximal small intestine in both rat and dog [12]. These findings emphasize the need to improve the knowledge of the gastrointestinal (GI) factors influencing the effect of the AMEs on the intestinal absorption in vivo [12].

To further improve in vivo predictions from non-clinical models, the use of simulated intestinal fluids (SIF) has been proposed [19]. Fasted or fed state simulated intestinal fluids (FaSSIF and FeSSIF), containing bile acids and lecithin which form colloidal structures (CS), and have an osmolality and pH similar to that seen in human aspirated intestinal fluids, have been extensively investigated in dissolution testing [20], [21], [22]. These systems have been little used in in vivo relevant non-clinical intestinal absorption models, such as SPIP model, due to their complexity and concerns with viability of the intestinal membrane barrier [23], [24]. However, it has recently been shown that the rat jejunum tolerated FaSSIF and FeSSIF without an increase in the blood-to-lumen clearance of the intestinal barrier integrity marker ⁵¹Cr-EDTA [25]. In both FaSSIF and FeSSIF the luminal presence of CS alter the apparent solubility and free concentration of APIs [26]. The presence of CS may increase the rate and extent of intestinal absorption, for low solubility API monomers that are partitioning into and/or interacting with CS, with a subsequent increased transport across the aqueous boundary layer (ABL) [27], [28]. However, for high solubility API monomers partitioning into and/or interacting with CS is high, this may result in a decreased absorption, and/or absorption rate [29], [30].

The primary objective of this study was to investigate the effect of three AMEs (caprate, N-acetyl cysteine, and sodium dodecyl sulfate) on the intestinal absorption rate of four selected model drugs (atenolol, metoprolol, ketoprofen, and enalaprilat) using FaSSIF and FeSSIF and the single-pass jejunal perfusion in the rat. The secondary objective was to determine the effect on the clearance of an intestinal integrity marker (⁵¹Cr-EDTA) from the blood to the intestinal lumen in the two prandial states.