Intestinal Lymphatic Drug Transport

For the majority of drug candidates, the intestinal lymphatics are unlikely to play a significant role in the transport of drug from the intestine to the systemic circulation. However, for some highly lipophilic compounds, intestinal lymphatic transport may play a role in their absorption. From a development perspective, early information as to the possible role of lymphatic transport and its formulation dependence is important for several reasons. Firstly, drug transport to the systemic circulation via the lymphatics avoids first pass hepatic metabolism and therefore for drugs where first pass hepatic metabolism is high, formulations or prodrug approaches which enhance intestinal lymphatic transport may lead to significant improvements in oral bioavailability. An excellent example of such a prodrug approach is the use of the highly lipophilic undecanoate ester of testosterone to promote intestinal lymphatic transport and to afford delivery of testosterone to the systemic circulation (Shackleford et al., 2003). In this example, whilst the proportion of the testosterone undecanoate dose delivered systemically via the intestinal lymph is low (2-3%), it has been shown that this accounts for greater than 80% of the systemically available testosterone, illustrating that the relatively modest proportion of the dose transported lymphatically was clinically relevant where first pass metabolism is significant (Shackleford et al., 2003). Secondly, if the proposed site of action of a drug candidate is in the lymph, such as an immune system modulator or an anti-cancer agent, specific delivery to the lymphatics and exposure of the lymphatic capillaries to relative high drug concentration may be advantageous. Thirdly, it has been suggested that the clearance of drugs delivered to the systemic circulation via the lymphatics, and in association with lymph chylomicrons, may be different to that of the same drug absorbed systemically (Hauss et al., 1994; Caliph et al., 2000) which has implications for the design and conduct of safety assessment studies -- particularly if the formulation used during chronic toxicology studies results in a different proportion of lymphatic transport compared with the clinical formulation.

Historically, there have been relatively few studies of drugs transported through the intestinal lymphatics because (i) there have been various animal model limitations associated with the study of lymph transport, and (ii) prior to the utilisation of higher throughout screening and in vitro technologies in drug discovery, many of the emerging drug candidates were not sufficiently lipophilic to be lymphatic transport candidates.

The intestinal lymphatics are a specialised absorption pathway for lipids and lipidic derivatives as well as a number of highly lipophilic xenobiotics and drugs including DDT (Sieber, 1976; Charman et al., 1986), benzo[a]pyrene (Laher et al., 1984), cyclosporin (Ueda et al., 1983), naftifine (Grimus and Schuster, 1984), probucol (Palin and Wilson, 1984), lipid soluble vitamins (Kuksis, 1987), mepitiostane (Ichihashi et al., 1992a,b), testosterone undecanoate (Shackleford et al., 2003) and halofantrine (Porter et al., 1996 a,b; Khoo et al., 2002; Khoo et al., 2003). Whilst these compounds have widely varying structures, it is possible to identify the relevant features that support lymph transport which include (i) high lipophilicity (e.g. log P > 5 and significant solubility in a triglyceride lipid) to support increased partitioning and association with enterocyte-derived lipoproteins [if the compound is a salt of a highly lipophilic free acid or free base, then the profile of the neutral form should also be considered (Khoo et al., 2002; Taillardat-Bertschinger et al., 2003)], (ii) reasonable molecular weight to provide transcellular permeability, (iii) adequate metabolic stability within the gastrointestinal tract and the enterocyte. In terms of a prodrug approach, it is possible to advantageously incorporate structural features of natural lipids into the prodrug structure to provide a potential mechanism to exploit endogenous absorption and biosynthetic pathway for lipids such as triglycerides and phospholipids (Shackleford et al., 2004).

Biosynthesis of Enterocyte-derived Lipoproteins

The absorption of lipid digestion products by the enterocyte is preceded by transport through the unstirred water layer to the brush-border surface. Hence, the efficiency of transport to the enterocyte depends on lipid diffusion in association with the colloidal structures in which they are solubilised in the gastrointestinal tract (since their intrinsic aqueous solubility is too low to support sufficient mass transport). The uptake mechanism(s) of lipid digestion products across the apical membrane of the enterocyte is not fully understood, although it appears to involve both active and passive processes (Stremmel et al., 1985; Stremmel, 1988; Thomson et al., 1993; Poirier et al., 1996).

Once within the enterocyte, 2-monoglycerides and fatty acids are rapidly and efficiently incorporated into a number of lipid processing pathways (Figure 1). As a rule of thumb, medium chain fatty acids (e.g. C12 and below) are transported directly into portal blood, whereas longer chain length fatty acids and 2-mono-glygerides (e.g C18 and greater) are re-synthesised within the enterocyte to triglyceride primarily via the mono-acyl glycerol pathway involving direct and sequential acylation of 2-monoglyceride by CoA-activated fatty acids. However, the mono-acyl glycerol is not the sole source of TG as the glycero-3-phosphate (G-3-P) pathway produces de novo TG under conditions of low lipid load, and indeed this pathway is inhibited under conditions where MG is readily available such as after post prandial administration (Nordskog et al., 2001).

Irrespective of the biosynthetic pathway from which they are derived, triglycerides are progressively processed through various intracellular organelles where the surface of the developing colloid (lipoprotein) is stabilised by the ordered addition of phospholipids (which are absorbed or synthesised de novo in a series of specific enzymatic processes) and various apoproteins. Under conditions of low lipid load (i.e. fasting conditions), the primary lipoproteins produced by the small intestine incorporating TG synthesised by the G-3-P pathway are very low density lipoproteins (VLDL), whereas under conditions of high lipid load (e.g. after a meal) the predominant lipoproteins produced which incorporate TG synthesised by the mono-acyl glycerol pathway are chylomicrons (CM) (Nordskog et al., 2001). Following their assembly, VLDL and CM fuse with the basolateral membrane of the intestinal cell and are released into the lamina propria where they are preferentially absorbed via the open capillaries of the mesenteric lymphatics rather than into intestinal blood vessels. The exclusive movement of VLDL and CM into mesenteric lymphatic capillaries rather than into blood vessels is due to the fact that lymphatic capillaries lack a basement membrane therefore being "permeable" to the large colloids, whereas blood vessels possess tight inter-endothelial junctions and a continuous basal lamina precluding facile access of colloidal lipoproteins (Swartz, 2001).

Assessment of Intestinal Lymphatic Drug Transport

An early indication of whether lymphatic transport is likely to play a role in drug disposition is important. A number of animal models have been described and reviewed recently (Hauss et al., 1998; Caliph et al., 2000; Edwards et al., 2001; Boyd et al., 2004) for estimating intestinal lymphatic drug transport, with the majority of studies having been conducted in either rats or dogs. As discussed, drugs transported to the systemic circulation via the intestinal lymphatics are typically highly lipophilic and have very low aqueous solubility. Therefore, to assess the likelihood of lymphatic transport, conditions and models should be sought which maximise the quantity of drug that is absorbed into the enterocyte, and enhance the proportion of the absorbed dose transferred into the lymphatics. The co-administration of food can, in many cases, represent a simple method by which the otherwise limiting luminal solubility and dissolution of poorly water soluble drugs can be diminished. Therefore, post-prandial administration of a candidate drug to a lymphatically cannulated dog is an excellent proof-of-concept study as the post-prandial administration can enhance drug absorption while providing a ready supply of lipids to enable chylomicron formation and intestinal lymphatic drug transport. Details of the triple cannulated dog model have been described (Edwards et al., 2001; Khoo et al., 2001) as well as recent drug-related studies (Khoo et al., 2001; Khoo et al., 2003; Shackleford et al., 2003; Shackleford et al., 2004). In this model, if the quantity of lipid (fat) administered in food is known, then it is possible to monitor the efficiency of lipid absorption and lymphatic transport by determining lymph triglyceride levels. This allows indicative transport data to be obtained in a small number of dogs, and in our experience, duplicate experiments are typically sufficient to provide a clear yes/no indication of the role of lymphatic transport for the administered drug. A practical consideration is the relatively high cost per animal associated with the complex surgery and aftercare. Therefore, screening studies examining the impact of formulation approaches on the extent of lymphatic transport may be more readily conducted in smaller species such as rats. There are varied methodologies/protocols used with rat studies encompassing different sites of cannulation and lymph fistulation (Noguchi et al., 1985), extent of hydration, fasting/fed state of the rat after lymph duct cannulation (Charman et al., 1986), whether the experiment is performed in a conscious or anaesthetized animals (Porter et al., 1996a,b) and the site of drug/lipid administration (Porter et al., 1996a). For example, we have previously examined the impact of anaesthesia, degree of formulation dispersion and administration route on the lymphatic transport of halofantrine in rats (Porter et al., 1996a,b). In anaesthetised rats, there was a significant effect of the degree of lipid dispersion in the formulation after intraduodenal administration with the lymphatic transport of halofantrine increasing from 3.9% when administered as a simple lipid solution, to 11.8% after administration as an emulsion to 17.7% when administered as a highly dispersed micellar formulation (Porter et al., 1996b). In all cases the type (2:1 molar mixture of oleic acid:glycerylmonooleate) and the volume of administered lipid (50 |L) was the same. In contrast, when the same formulations were administered orally to conscious rats there was no effect of formulation dispersion on lymphatic transport of halofantrine which was similar for the lipid solution and micellar solution (19.1% and 20.0%, respectively) (Porter et al., 1996a). These data suggest that the inherent reduction in intestinal processing in an anaesthetised rat and the intraduodenal administration led to the differing degree of formulation dispersion impacting the relative extent of drug absorption. In contrast, in the conscious rats that received oral formulations, the improved gastrointestinal processing of the lipid solution led to efficient in situ dispersion thereby removing the likely consequence of differences in the extent of initial formulation dispersion.

Fig1 illustrates how conscious rat data can be used to broadly screen for formulation-related changes in lymphatic transport, and that the rank order data are comparable to data obtained in higher species such as dogs. In this example, the lymphatic transport of halofantrine after oral administration of a simple lipid solution formulation to conscious rats (Fig 1A) was increased after administration as a medium chain lipid solution compared with a lipid free suspension formulation (Caliph et al., 2000). However, significantly more lymphatic transport was observed after administration of halofantrine formulated in a long chain (C18) lipid solution vehicle. These data presumably reflect the increased propensity of long chain lipids for resynthesis back to triglyceride in the enterocyte and assembly into lymph lipoprotein precursors and lymph lipoproteins, thereby providing a lymphatic lipid sink into which the drug may partition. In contrast, medium chain lipids are typically transported through the enterocyte without re-esterification and are absorbed into the portal blood, thereby limiting the possible extent of lymphatic transport. A key issue associated with assessment of prospective formulations to rats is the self-evident realisation that prototypical humans formulations (e.g. tablets, hard or soft gelatin capsules) cannot physically be administered to small animals. However, the data in Fig 1B suggest that the relative patterns of lymphatic transport in conscious rats may reflect the rank

Panel A: Fasted Dogs

4 6 8 Time (hours)

Panel B: Fasted Rats

Time (hours)

Figure 1. Lymphatic transport of halofantrine (Hf) after administration to fasted dogs or fasted in formulations comprising lipids with different fatty acid chain lengths. Panel A. Extent of lymphatic transport of Hf (% dose, mean ± SD, n = 3/4) in greyhound thoracic lymph after oral administration of 50 mg Hf in self emulsifying formulations containing cremophor EL and either long chain lipids (•) or medium chain lipids (▲). Figure adapted from Khoo et al., 2003. Panel B. Extent of lymphatic transport of Hf (% dose, mean ± SD, n = 3/4) in rat mesenteric lymph after oral administration of 2.5 mg Hf dissolved in long chain triglyceride lipid (•), medium chain triglyceride lipid (▲) or as a lipid-free aqueous suspension (H). Figure adapted from Caliph et al., 2000.

order effect after administration of single unit dose formulations (in this case soft gelatin capsule-based, self emulsifying formulations based on medium and long chain lipids) to larger species such as the dog (Khoo et al., 2003).

Recently, our laboratory examined whether the small quantities of lipid present in unit dose microemulsion formulations comprising medium (C8-10) or long chain (C18) glyceride lipids was able to stimulate the intestinal lymphatic transport of halofantrine when administered to thoracic lymph duct cannulated fasted dogs (Khoo et al., 2003). Drug was formulated as a single soft gelatin capsule containing approximately 1 g of a microemulsion preconcentrate based on either medium or long chain glycerides and thoracic lymph was collected and systemic plasma samples taken over 10 hr post dose. The extent of lymphatic transport of Hf after administration of the long chain lipid formulation was high (28.3% of dose), and significantly higher than that seen after administration of the medium chain formulation (5.0% of dose). These data are the first to demonstrate that the small amounts of lipid present within a single lipid-based dose form can support intestinal lymphatic transport in the fasted state, with long chain glycerides appearing to be more effective with respect to lymphatic transport than the equivalent medium chain formulation.

Lower Your Cholesterol In Just 33 Days

Lower Your Cholesterol In Just 33 Days

Discover secrets, myths, truths, lies and strategies for dealing effectively with cholesterol, now and forever! Uncover techniques, remedies and alternative for lowering your cholesterol quickly and significantly in just ONE MONTH! Find insights into the screenings, meanings and numbers involved in lowering cholesterol and the implications, consideration it has for your lifestyle and future!

Get My Free Ebook

Post a comment