ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2018

Digital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Pharmacy 257

Biopharmaceutical aspects ofintestinal drug absorption

Regional permeability and absorption-modifyingexcipients

DAVID DAHLGREN

ISSN 1651-6192ISBN 978-91-513-0442-7urn:nbn:se:uu:diva-358467

Dissertation presented at Uppsala University to be publicly examined in B41, BMC,Husargatan 3, Uppsala, Friday, 2 November 2018 at 09:15 for the degree of Doctor ofPhilosophy (Faculty of Pharmacy). The examination will be conducted in English. Facultyexaminer: Professor Steffansen Bente (LEO Pharma).

AbstractDahlgren, D. 2018. Biopharmaceutical aspects of intestinal drug absorption. Regionalpermeability and absorption-modifying excipients. Digital Comprehensive Summaries ofUppsala Dissertations from the Faculty of Pharmacy 257. 68 pp. Uppsala: Acta UniversitatisUpsaliensis. ISBN 978-91-513-0442-7.

Before an orally administered drug reaches the systemic circulation, it has to dissolve in theintestinal fluids, permeate across the intestinal epithelial cell barrier, and pass through the liver.The permeation rate of drug compounds can be low and show regional differences.

The thesis had two general aims. The first of these was, to determine and compare regionalintestinal permeability values of model compounds in human and dog. The second was tounderstand the possible effects of absorption-modifying pharmaceutical excipients (AMEs) onthe intestinal permeability of the model compounds. The usefulness of several preclinical animalmodels for predicting the impact of regional intestinal permeability and AMEs in human wasalso investigated.

There was a good correlation between human and dog permeability values in the smallintestines for the model compounds. The colon in dog was substantially more permeable thanthe human colon to the low permeability drug, atenolol. This difference in colonic permeabilitymay have implications for the use of dog as a model species for prediction of human intestinaldrug absorption.

There were no effects of AMEs on the intestinal permeability of any of the high permeabilitycompounds, in any animal model. In the rat single-pass intestinal perfusion model, therewas a substantial increase in permeability of all low permeability drugs, induced by twoAMEs, chitosan and SDS. This AME-induced increase was substantially lower in the more invivo relevant rat and dog intraintestinal bolus models. A shorter AME exposure-time in the ratsingle-pass intestinal perfusion model (15 vs. 75 min) could, however, predict the result from thebolus studies in rat and dog. This illustrates the impact of intestinal transit and mucosal exposuretime on AME effects in vivo. The intestinal luminal conditions and enteric neural activity alsohad an impact on determinations of drug permeability in the rat single-pass intestinal perfusionmodel, which can have implications for its in vivo relevance.

In summary, this thesis used multiple in vivo models to evaluate the impact of severalbiopharmaceutical processes on intestinal drug absorption. This has led to an increasedunderstanding of these absorption mechanisms.

Keywords: intestinal permeability, absorption-modifying excipients

David Dahlgren, Department of Pharmacy, Box 580, Uppsala University, SE-75123 Uppsala,Sweden.

© David Dahlgren 2018

ISSN 1651-6192ISBN 978-91-513-0442-7urn:nbn:se:uu:diva-358467 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-358467)

Till Ilse, Tilda och Elmer

Luff jullie

List of Papers

This thesis is based on the following papers, which are referred to in the text by their numbers.

1. Dahlgren, D., Roos, C., Lundqvist, A., Abrahamsson, B., Tan-

nergren, C., Hellström, P. M., Sjögren, E., Lennernäs, H. Re-gional intestinal permeability of three model drugs in human. Mol. Pharm. 2016, 13(9):3013–3021.

2. Dahlgren, D., Roos, C., Johansson, P., Lundqvist, A., Tanner-gren, C., Abrahamsson, B., Sjögren, E., Lennernäs, H. Regional intestinal permeability in dogs: biopharmaceutical aspects for de-velopment of oral modified-release dosage forms. Mol. Pharm. 2016, 13(9):3022−3033.

3. Dahlgren, D., Roos, C., Lundqvist, A., Tannergren, C., Langguth, P., Sjöblom, M., Sjögren, E., Lennernäs, H. Preclini-cal Effect of Absorption Modifying Excipients on Rat Intestinal Transport of Model Compounds and the Mucosal Barrier Marker 51Cr-EDTA. Mol. Pharm. 2017, 14(12):4243–4251.

4. Dahlgren, D., Roos, C., Johansson, P., Tannergren, C., Langguth, P., Lundqvist, A., Sjöblom, M., Sjögren, E., Len-nernäs, H. The effects of three absorption-modifying critical ex-cipients on the in vivo intestinal absorption of six model com-pounds in rats and dogs. Int. J. Pharm. 2018, 547(1–2):158-168

5. Dahlgren, D., Roos, C., Lundqvist, A., Tannergren, C., Sjöblom, M., Sjögren, E., Lennernäs, H. Effect of absorption-modifying excipients, hypotonicity, and enteric neural activity in an in vivo model for small intestinal transport. Int. J. Pharm. 2018, 459(1–2):239–248.

6. Dahlgren, D., Roos, C., Lundqvist, A., Tannergren, C., Sjöblom, M., Sjögren, E., Lennernäs, H. Time-dependent effects of ab-sorption-modifying excipients on small intestinal transport. Eur. J. Pharm. Biopharm. 2018, 132:19-28.

All reprints of articles and figures were made with permission from the re-spective publishers.

Additional papers not included in this thesis

a. Dahlgren, D., Roos, C., Sjögren, E., Lennernäs, H. Direct In vivo Human Intestinal Permeability (Peff) Determined with Dif-ferent Clinical Perfusion and Intubation Methods. J. Pharm. Sci. 2014, 104(9):2702–2726.

b. Sjögren, E., Dahlgren, D., Roos, C., Lennernäs, H. Human in vivo regional intestinal permeability: quantitation using site-specific drug absorption data. Mol. Pharm. 2015, 12(6):2026-2039.

c. Forner, K., Roos, C., Dahlgren, D., Kesisoglou, F., Konerding, MA., Mazur, J., Lennernäs, H., Langguth, P. Optimization of the Ussing chamber setup with excised rat intestinal segments for dissolution/permeation experiments of poorly soluble drugs. Drug. Dev. Ind. Pharm. 2017, 43(2):338-346.

d. Roos, C., Dahlgren, D., Sjögren, E., Tannergren, C., Abra-hamsson, B., Lennernäs, H. Regional intestinal permeability in rats: a comparison of methods. Mol. Pharm. 2017, 14(12):4252-4261.

e. Roos, C., Dahlgren, D., Berg, S., Westergren, J., Abrahamsson, B., Tannergren, C., Sjögren, E., Lennernäs, H. In vivo Mecha-nisms of Intestinal Drug Absorption from Aprepitant Nanofor-mulations. Mol. Pharm. 2017, 14(12):4233-4242.

Contents

Introduction ................................................................................................... 11 Background .............................................................................................. 11 Physiology and function of the gastrointestinal tract ............................... 14 Intestinal drug absorption ......................................................................... 16 

Basic pharmacokinetic processes ........................................................ 16 Definition of intestinal absorption and bioavailability ........................ 17 Mucosal transport mechanisms and permeability ................................ 19 

Strategies for increasing intestinal drug absorption ................................. 21 Low solubility drugs ............................................................................ 21 Low permeability drugs ....................................................................... 22 

Methods for studying intestinal drug absorption ...................................... 23 In silico models .................................................................................... 24 In vitro models ..................................................................................... 24 In situ and in vivo models .................................................................... 25 

Aims of the thesis.......................................................................................... 27 

Methods ........................................................................................................ 28 Study chemicals ........................................................................................ 28 

Model compounds ............................................................................... 28 Absorption-modifying excipients ........................................................ 29 

Description of study subjects and animals ............................................... 29 Clinical Study ...................................................................................... 30 Dog Studies .......................................................................................... 30 Rat Studies ........................................................................................... 30 

Drug compound interactions in the rat Ussing chamber .......................... 30 Regional intestinal permeability studies ................................................... 31 

Clinical study ....................................................................................... 31 Dog study ............................................................................................. 32 

Absorption-modifying excipient studies .................................................. 32 SPIP studies in rat ................................................................................ 32 Intraintestinal bolus studies in rat and dog .......................................... 35 

Bioanalytical method ................................................................................ 36 Data analysis ............................................................................................ 37 

Apparent permeability (Papp) in the Ussing chamber ........................... 37 Intestinal effective permeability (Peff) and absorptive flux (Jabs) ......... 37 

Intestinal 51Cr-EDTA clearance (CLCr) ................................................ 38 Statistical analysis ................................................................................ 38 

Results and discussion .................................................................................. 39 Transport and metabolic interactions of the model drugs ........................ 39 Regional intestinal permeability ............................................................... 40 

Human clinical study ........................................................................... 40 Dog study ............................................................................................. 42 

Absorption-modifying excipients ............................................................. 43 Effects on the transport of high permeability drugs ............................ 43 Correlation between Jabs and CLCr in the rat SPIP model ..................... 44 Effects in the rat SPIP model at isotonic conditions ............................ 44 Effects of nerve activity and hypotonicity in the rat SPIP model ........ 46 Effects in the rat and dog intraintestinal bolus models ........................ 48 Time-dependent effects in the rat SPIP model .................................... 49 

Conclusions ................................................................................................... 52 

Populärvetenskaplig sammanfattning ........................................................... 54 

Acknowledgements ....................................................................................... 56 

References ..................................................................................................... 58 

Abbreviations

a/b y-intercept of distribution (a) and elimination (b) phases α/β slope of distribution (α) and elimination (β) AME absorption-modifying excipient BCS biopharmaceutics classification system C concentration COX cyclooxygenase enzyme CL clearance CLCr blood-to-lumen 51Cr-EDTA clearance CM carrier-mediated EG gut-wall extraction EH hepatic extraction fa fraction absorbed GI gastrointestinal HBA/HBD hydrogen bond acceptor/donor iv intravenous Jabs lumen-to-blood absorptive flux Km michaelis constant Log P n-octanol−water coefficient Log D7.4/6.5 n-octanol−water partition coefficient at pH 7.4/6.5 MM molar mass NAC n-acetylcysteine Papp apparent permeability Peff effective permeability PK pharmacokinetic(s) pKa dissociation constant PSA polar surface area Q flow rate SDS sodium dodecyl sulfate SPIP single-pass intestinal perfusion V apparent volume of distribution Vmax maximum rate

11

Introduction

Background In terms of units sold and total revenues, oral drug treatment is the most com-mon administration route for systemically acting drugs.1 This is partly because there is a long tradition of dosing drugs orally, but also because its cost-effi-cient manufacturing and non-invasiveness generate a high acceptance among patients. Taken together, this indicates that the oral route of drug administra-tion will prevail also in the future. However, the nature of the gastrointestinal (GI) tract is to prevent absorption and translocation of potentially harmful lu-minal constituents into the central circulation, while still allowing the absorp-tion of nutrients and water.2,3 There are consequently several obstacles asso-ciated with the oral administration route that need to be overcome for success-ful systemic drug treatment. These barriers and processes are described sche-matically in Figure 1 and will be discussed in detail.

Figure 1. Schematic presentation of the gastrointestinal and metabolic processes that determine the rate and extent of drug absorption into systemic circulation following oral drug administration of a solid dosage form. CM = carrier-mediated, F = bioa-vailability, fa = fraction absorbed, EG/EH = gut/hepatic extraction, PLD/PPD = pas-sive lipoidal/paracellular diffusion.

12

Figure 1 shows that the drug product/formulation (e.g., capsule, tablet) must initially disintegrate, so that drug particles can dissolve in the GI fluid. The drug solubility in the intestinal GI fluids must be high enough to enable a suf-ficiently fast dissolution rate. The free drug molecules in solution also deter-mine the concentration gradient between the intestinal lumen (i.e., the inside of the intestinal tube) and the blood; this gradient is the driving force for per-meability and absorption, where permeability is the transport across the apical membrane, the rate limiting membrane barrier. A reduction in the luminal free drug concentration can occur if the drug molecule precipitates, is chemically or enzymatically degraded, or forms complexes with the luminal content. The drug molecules are considered absorbed after they have been transported across the outer lipophilic cell membrane, and this membrane can impose sub-stantial resistance to large and/or hydrophilic drug molecules. Finally, before an absorbed drug molecule is introduced into the central blood circulation, it passes through the intestinal barrier and liver, where it may be metabolized and lose its pharmacological effect, or be excreted with bile back into the in-testines.

In the drug discovery process, candidate drug molecules are selected based on physicochemical properties, the affinity for the pharmacological target, membrane transport properties, chemical and metabolic stability, and their safety/toxicity profile.4 These properties are evaluated by applying various preclinical tools and models with varying degrees of complexity. These range from simple screening assays with thousands of molecules, to complex pre-clinical in vivo animal models that evaluate only a few selected molecules.

The aim of these approaches and models is to select one, or a few, drug candidates suitable for further development in human clinical studies. There are three phases of clinical studies (I–III) with an ascending number of sub-jects. Phase I establishes pharmacokinetics and safety in healthy volunteers at different dose levels, phase II assesses efficacy and side effects at different doses in patients, and phase III assesses effectiveness and safety in patients. The clinical part of drug development is estimated to account for about half of the total costs, partly because of the number of subjects involved and the sur-rounding organization required.5 The high cost is also associated with the high attrition rate in drug development; only about 1 in 10 clinically tested drug candidates reaches the market.6 The high cost and attrition rate at the later stage of the drug-development process emphasizes the importance of selecting the candidate drugs most likely to succeed; this requires robust and accurate preclinical tools. The pharmaceutical industry has recently recognized that many of the preclinical in vitro and in vivo models for prediction of human intestinal drug absorption generate suboptimal predictions or are inadequately validated.7-9 This is exemplified by the fact that 16% of all drug compounds that fail in the early phase of clinical development do so because of undesira-ble pharmacokinetic properties.10

13

In addition to the drug discovery process, already approved drug com-pounds are frequently reformulated into new drug formulations and applica-tions.11 Generic drug products are an example of this; these are new formula-tions of an existing drug product with an expired patent. A generic drug prod-uct must be bioequivalent to the original product. Bioequivalence is attained if the two drug products give rise to the same plasma exposure of the drug compound following oral administration.12 This can be proved in a clinical study with healthy volunteers, or in vitro for drugs with a high permeability and solubility that are formulated into immediate-release products in accord-ance with the biopharmaceutics classification system (BCS).13,14 Drug refor-mulation is also used to improve the effectiveness of a drug treatment, for instance by extending/delaying the release of drug in the intestines, and con-sequently giving more stable plasma concentrations.15,16 Such a formulation may enable, for instance, once per day drug administration, a more steady ef-fect of the drug during the day, as well as a reduction of side effects. The reformulation from immediate to modified release does not always require a complete preclinical evaluation, as the regulatory application can refer to the original product.

Ideally, drug molecules with unfavorable PK properties due to poor intes-tinal absorption should be identified in preclinical evaluations so that their development is discontinued. Nonetheless, further development of formula-tions containing drugs with low intestinal solubility and/or low intestinal per-meability, is sometimes warranted, if the drug product can be designed to mit-igate the impact of unfavorable PK properties. For instance, the drug product can be designed to increase the drug dissolution rate, and thereby temporarily increase the dissolved amount of a low solubility drug. The drug product can also contain absorption-modifying excipients (AMEs) that increase the per-meation rate of dissolved drug molecules over the intestinal membrane, and thereby increase the absorption of low permeability drugs.

Biopharmaceutics is the field that investigates how physiology and drug product influence the rate and extent of absorption and bioavailability of an active pharmaceutical ingredient. Biopharmaceutical features such as intesti-nal drug product performance and intestinal drug absorption are thoroughly evaluated in the drug development, as well as in the reformulation of drug products. In both these cases, the aim is to predict the intestinal in vivo drug absorption in humans. This thesis describes and discusses preclinical tools for the determination of intestinal drug absorption, with an emphasis on in vivo models, regional intestinal drug absorption, and the effect of AMEs on intes-tinal drug absorption.

14

Physiology and function of the gastrointestinal tract The GI tract is a continuous, muscular and hollow tube that stretches from the mouth to the anus.17 It is divided into regions that each have their own mor-phology, physiology, and function (Figure 2). First are the pharynx and esoph-agus, followed by the stomach, small intestine, and large intestine. The small intestine is in turn subdivided into the duodenum, jejunum, and ileum, and the large intestine into the colon and rectum. The total length of the human intes-tines during normal muscle tonus is about 5 meters, but it can double in length post mortem.17

Figure 2. Gross anatomy of the intestines. For the national cancer institute © 2018 Terese Winslow LLC, U.S. Govt. has certain rights.

The characteristics and morphology of the intestinal barrier vary between re-gions, but it has a common histology (Figure 3). Between the lumen and the outside of the intestines, the mucosal wall is divided into four distinct lay-ers/subdivisions: the mucosa (composed of epithelium, lamina propria, and muscularis mucosae); the submucosa; the muscle layer (composed of circular muscle, myenteric nerve plexus, and longitudinal muscle); and the serosa.17 The primary barrier between lumen and blood is the mucosal epithelium, which is comprised of columnar epithelial cells.18 These intestinal epithelial cells form a protective barrier as they are tightly connected by intercellular tight junctions and covered by a protective mucus layer.2,19-21 The underlying mucosal layer, the lamina propria, contains blood vessels, nerve fibers, lym-phatic tissue, immune cells, and smooth muscle that regulates blood flow and villi movement.22 The submucosa contains connective tissue with major blood and lymphatic vessels; the circular and longitudinal muscles in the muscle layer control GI movement.17 The serosa is mainly composed of connective tissue that supports the GI tract in the abdominal cavity. The 400-600 million neurons and their nerve fibers in the GI system are jointly called the enteric

15

nervous system, which is partly autonomous from the central nervous sys-tem.17,23 The enteric nervous system constitutes a variety of sensory neurons, interneurons, motor neurons, and secretory neurons, all of which are involved in regulation of peristalsis, secretion, and absorption.

Figure 3. General structure of the intestinal barrier. Used with permission OpenStax College, Rice University.

The primary function of GI tract is to create a selective barrier which enables absorption of nutrients, water, and electrolytes, while restricting transport of larger, harmful luminal contents, such as bacteria, viruses, and proteins.2,3,17,24 Nutrient absorption starts with mechanical digestion of food by chewing in the mouth, and by contractions in the stomach and intestines. Food is also chemically digested by enzymes secreted in the mouth, stomach, and small intestine. The small intestine is also the primary absorptive organ for digested food (nutrients, vitamins and electrolytes) and water. Unabsorbed water in the small intestines is absorbed in the colon, which leads to concentration of the feces that contains undigested material and commensal bacteria.25 The peri-staltic movement in the intestines continuously transports ingested food dis-tally, from the stomach to the end of the rectum, which also reduces the spread and overgrowth of bacteria in the intestines.17

The various regional intestinal differences related to the luminal contents and to the mucosal barrier have potential implications for drug absorption. Regional intestinal differences in luminal water content and pH, length and area of the intestinal epithelium, and intestinal transit times, are summarized in Table 1. The small intestine epithelial surface area is increased compared to a smooth tube by a factor of 1.6 because of circular folds, and by a factor

16

of 6 because of the finger-like protrusions called villi.26,27 In addition, the ep-ithelial cells in both the small and large intestines are covered by microvilli that further increase the surface area by a factor of about 10.

The free intestinal water content in the fasted state is low, and resides in distinct pockets. Water content has been measured using MRI, and it should be mentioned that total water content (bound and free) may be higher, as only the free water content was determined.28,29 The pH in the stomach is about 1.9 in the fasted state, and it varies between pH 6.0 and 7.4 in the small and large intestine.30,31 The gastric emptying half-life and intestinal transit times differ depending on prandial state. Gastric emptying is also affected by what is being transported, where the emptying half-life is shorter for liquids than for solid dosage forms, such as capsules.30,32,33

Table 1. Regional intestinal anatomical and physiological (fasted state) differences that can have implications for drug absorption.26-33

Segment Subsegment Water content

pH Length Mucosal

surface area Transit

Liquid Capsule

Stomach - 50 mL 1.9 0.5 m2 t½ 15 min 0-4 h

Small intestine

Duodenum 50-100 mL

6.3 30 cm 30 m2

Jejunum 6.8 150 cm 3-5 h

Ileum 7.4 150 cm

Large intestine

Cecum 13 mL

6.0 150 cm 1.9 m2

Colon 7.0 8-28 h

Rectum 7.3

In addition to the parameters mentioned in Table 1, the thickness of the intes-tinal mucus layer varies between segments of the intestine. It is thicker and more firmly attached in the stomach and large intestine, while it is thinner and loosely attached in the small intestine.34 Finally, abundance of transporters in-volved in absorption and efflux of molecules (e.g. nutrients, drugs) differ be-tween intestinal segments. There are conflicting data about the regional abun-dance of intestinal transporters, but absorptive transporters are found primar-ily in the small intestine, while efflux transporters may be distributed in all parts of the intestinal tract.35-37

Intestinal drug absorption Basic pharmacokinetic processes Pharmacokinetics (PK) describes what happens to a xenobiotic (e.g., a drug) within the body, while pharmacodynamics describes the relationship between the concentration of the drug at its site of action and the magnitude of its phar-macological response.38 The PK is determined by four fundamental processes

17

that determine the fate of a drug in the body: absorption, distribution, metab-olism, and excretion (ADME).

Absorption after oral drug administration is the process of drug transport into the epithelial cells lining the intestinal lumen.

Distribution describes the relative proportion of drug in the different body tissues, such as the blood, brain, and lungs, and the movement of drug to and from the blood to these tissues. Distribution is often described as volume of distribution (V), which is the apparent volume in the body that contains drug, i.e., the relation between the amount of drug in the body and the concentration in the blood or plasma.

Metabolism is the biotransformation of the absorbed drug and is performed by more or less specialized enzymes that convert the drug to a more hydro-philic metabolite. This conversion is divided into two phases: Phase 1 (oxida-tion/reduction/hydrolysis), and Phase 2 (conjugation).39 Phase 1 enzymes add a new functional group to the drug molecule. This new compound can in some cases be more potent and/or toxic than the original drug molecule. The Phase 2 conjugating enzymes attach a larger molecule to the compound, which de-toxifies it and often makes it more hydrophilic. These processes are mainly performed intracellularly and the primary metabolic organ is the liver, but other organs such as the intestines, lungs, and kidneys can be involved.

Excretion is the elimination of the unchanged drug from the body, and is primarily performed by the liver (biliary) and kidneys (renal). Biliary excreted material enters the intestine with the bile duct into the proximal small intestine and the material is considered excreted once it leaves the body with the feces, but not if it is reabsorbed in the intestine. Renally excreted material leaves the body with the urine.

Elimination describes the irreversible loss of drug from the body, which can be exerted both by metabolism and excretion. Elimination is often ex-pressed as clearance (CL), which is the volume of blood (or plasma) that is being cleared of drug per min (mL/min). Consequently, the plasma elimina-tion half-life (t1/2) of a drug compound decreases with an increased CL, and increases with an increased V. Disposition describes both the distribution and elimination of a drug from the central circulation.

Definition of intestinal absorption and bioavailability Bioavailability (F) is the most important PK parameter for characterizing the fraction of an orally administered dose of a drug that reaches the systemic circulation in its unchanged form; F also describes the rate of this process in a regulatory context.40,41 Different drug compounds, or drug products contain-ing the same compound, can be easily compared by investigating their F val-ues following oral administration. F describes three serial processes: the frac-tion dose absorbed (fa) in the intestines; and the first pass extraction ratio of

18

the drug in the gut wall (EG) and liver (EH), as an absorbed drug must pass through these two organs before reaching the systemic circulation (Eq. 1).

1 1 (1) where fa describes the intestinal absorption of a drug compound; this is reg-

ulatory and scientifically defined as the total mass (M) of a dose that is being transported from the intestinal lumen over the intestinal epithelial apical cell membrane at any time (Eq. 2).14,42

(2)

where A is the area of the intestinal mucosa, Peff is the effective permeabil-ity over the intestinal membrane, and Clumen is the free luminal drug concen-tration at the intestinal site of absorption. From Eq. 2 it can then be derived that the driving forces for absorption are Peff and Clumen. The Peff and Clumen of a drug compound are, in turn, affected by three factors: physicochemical, pharmaceutical, and physiological.

Physicochemical properties of a drug molecule, such as its structure, sol-ubility, and lipophilicity, determine its solubility in water as well as its parti-tioning into the enterocyte apical cell membrane (i.e. Peff).

Pharmaceutical factors, such as, the formulation type, choice of excipi-ents, and crystalline form, may also affect Clumen and Peff. A drug product can be designed to either increase or delay/prolong drug dissolution, to control the absorption behavior of the drug compound along the GI tract. The drug prod-uct may also contain excipients that reduce the mucosal barrier integrity, and thereby increase the Peff of the drug. The intestinal transit time can also be reduced by the drug formulation if it affect the GI motility, something that has been observed for osmotically active excipients.43,44

Physiological factors, such as individual and regional intestinal variations in luminal content (e.g., bile, pH, and fluid volume), may affect Clumen and Peff. Chemical and/or metabolic degradation in the intestinal lumen reduces Clumen, as well as the total drug amount available for absorption. Variations in luminal pH affect the ionization of weak acids and bases, which influences both Clumen and Peff; an ionized molecule typically has a higher solubility and lower per-meability. The drug compound may also partitions or bind to luminal food contents, potentially reducing both Clumen and Peff. Food content may also af-fect the release of drug from a formulation.

The extraction ratio (E) of an organ is assessed by measuring the drug con-centration in the blood (or plasma) entering (Cin) and leaving (Cout) the intes-tinal segment (Eq. 3).

(3)

The extraction in the gut wall is primarily attributed to metabolism, while extraction in the liver can also include biliary secretion of the mother com-pound. As not all blood that passes the intestines does so in proximity to the metabolizing enterocytes, Cin and Cout do not resemble the EG of a drug being

19

transported from the intestinal lumen to the blood. Therefore, EG can be de-termined by comparing the fraction of a drug escaping first-past extraction following portal vein and intestinal (if corrected for fa) administration.

As a drug passes through the intestinal barrier and liver before entering the systemic circulation, extraction in these organs reduces F. The extraction in these organs is dependent on enzyme expression/activity (i.e. intrinsic CL), which can deviate between intestinal sites for EG, and between different indi-viduals for both EG and EH.45,46 Such variations in metabolism are especially important to consider when investigating the F of drugs with a high extraction ratio.47 The F of metoprolol can, for instance, vary up to six-fold between in-dividuals as a result of polymorphism of the CYP2D6 metabolic liver en-zyme.48,49

Mucosal transport mechanisms and permeability The mechanisms of mucosal drug transport and the concept of intestinal per-meability are discussed in this section, as this thesis primarily revolves around the factors that determine the transport rate of a dissolved drug compound over the intestinal barrier.

Transport mechanisms The mass transfer of luminally dissolved drug molecules across the apical in-testinal epithelial cell barrier includes one, or several, of the following transport mechanisms: passive lipoidal and paracellular diffusion, and/or carrier-mediated transport in both the absorptive and secretive (efflux) di-rections (Figure 4).50

Figure 4. Transport mechanisms over the intestinal epithelial cell membrane. fa = fraction absorbed.

20

Passive lipoidal diffusion is the process whereby a compound is transported into the intestinal epithelial cell cytosol by diffusion across the apical mem-brane. This membrane is composed primarily of phospholipids organized as a bilayer with a lipophilic core. Substantial passive transport of solutes over this lipophilic barrier is therefore attributed mainly to small and lipophilic mole-cules.51-53

Passive paracellular diffusion is the process whereby a compound is transported across the mucosal barrier in the water-filled space between the intestinal epithelial cells. The paracellular space is sealed by tight junction proteins, which are partly under physiological regulation.2,19,54 This transport mechanism is mainly associated with small (MM < 300 Da), hydrophilic mol-ecules (logD6.5 < -2), but also large, hydrophilic molecules, such as peptides and proteins.55-58 However, the total area of the paracellular space is very small compared to the total intestinal area, and this transport mechanism is therefore generally regarded to be of low quantitative importance for drug absorption in vivo.1,56

Carrier-mediated (CM) transport is the process whereby a compound is transported into (influx) or out of (efflux) the intestinal epithelial cell cytosol across the lipoidal membrane bilayer using a protein transporter. CM transport can be either active or facilitated.36 Active transport uses energy to create a concentration gradient across membranes, and is classified as primary or sec-ondary. It is primary if the energy comes from ATP hydrolysis (e.g., ABC transporters), and secondary if the energy comes from a previously generated ion gradient (e.g., SLC transporters), such as the high extracellular Na+ con-centration generated by the Na+/K+ ATPase.59 Facilitated transport is an en-ergy-independent process whereby a solute is transported across the cell mem-brane using a protein transporter (e.g., SLC transporters) downstream of the electrochemical gradient.

CM transport is important for absorption of water soluble nutrients, such as glucose, vitamins, and amino acids, but this transport mechanism can also be important for drug compounds. The herpes simplex drug, valacyclovir, for example, is specifically designed to be a substrate for the peptide transporter 1 to increase its bioavailability.60,61 Valacyclovir is a prodrug, meaning that it is transformed into its active form (acyclovir) after absorption.62

Recently, the CM transport route has been proposed to be the universal transport mechanism, with no impact from passive lipoidal diffusion.63,64 However, the experimental evidence for this transporter-only theory is weak, and the opposing view—that there is a coexistence between CM and passive transport processes—is more widely accepted.65

It should be noted that a dissolved drug molecule must diffuse across a water layer with limited convection covering the epithelial cell membrane be-fore it can be absorbed. This layer is called the aqueous boundary layer.66-68 The relevance of this layer for absorption of dissolved drug molecules has been thoroughly investigated, and is generally considered not to be the rate-

21

limiting step in drug absorption in vivo.69,70 However, for drugs with a very high permeability, and for specific drug formulations, such as nanosuspen-sions, it can restrict drug absorption kinetics.71,72

Permeability Permeability (cm/s) is the intrinsic constant of a solute that relates flux (J)—the mass transfer per area per time—to the concentration gradient between the intestinal lumen and the blood, regardless of transport mechanism (Eq. 4).

(4)

where C is the concentration in the lumen, PP is passive permeability (para-cellular and/or lipoidal diffusion), PCM is CM permeability, Vmax is the maxi-mum rate of CM transport, and Km is the Michaelis constant that describes the concentration at Vmax/2. For Eq. 4 to be valid, the concentration in the lumen must be at least 10 times higher than the free blood concentration (sink condi-tions). During in vivo conditions in both human and pig, the concentration in the blood is << 10% of the luminal concentration; this is true for a wide range of drugs, regardless of permeability class and transport mechanism.73,74 This indicates that drugs already present in the blood do not affect the permeation from the intestinal lumen into the blood, because a concentration gradient is established.

Strategies for increasing intestinal drug absorption The chemical structure of a drug molecule influences Clumen and Peff, which are the two primary variables determining intestinal fa (Eq 2). To increase the sol-ubility of a drug molecule, hydrogen bond donors and/or acceptors can be added. Permeability can be increased by increasing the passive transcellular permeability of the molecule, which is typically done by increasing its lipo-philicity or reducing its size, or by modifying it to become a substrate for an influx protein transporter.75 However, changes to the molecular structure to improve PK properties cannot be performed without affecting the pharmacol-ogy. Nonetheless, there are drug formulation strategies that can temporarily increase the drug solubility or the transport rate over the intestinal membrane, which may ultimately result in an increased fa and F.

Low solubility drugs Target-based assays and high-throughput screening models have resulted in a marked increase of poorly water-soluble drug candidates.76 A low drug solu-bility increases the risk of a low intestinal absorption, which can be critical when a high dose is necessary. A multitude of formulations strategies are im-

22

plemented to increase the dissolution rate of these compounds, as it is gener-ally the rate-limiting absorption step for lipophilic drugs.77 The main factors that affect drug dissolution rate (dM/dt) are the effective surface area of the drug particles (A), molecular diffusivity (D), diffusion layer thickness (h), sat-uration concentration (Cs), intestinal bulk concentration (C), and dissolution media volume (Vm), (Eq. 5).78

(5)

The formulation strategies generally aim at increasing either the saturation solubility (Cs) and/or the particle surface area (A).

The saturation solubility in the diffusion layer can be increased by modi-fying the crystal structure of the drug particles, or by using complexation agents and pH modifiers. Examples of crystal modifications include the use of metastable polymorphs of the crystal structure, different salt forms of an acid or base, inclusion of pH modifiers (which change the pH in the diffusion layer), or amorphous solids.79-81 These modifications have the potential to tem-porarily increase the solubility in the diffusion layer, and hence the dissolution rate of a poorly water-soluble drug compound. An example of a complexation agent that can be included in a formulation is cyclodextrin, it is a hydrophilic molecule with a hydrophobic cavity where the drug molecule can bind, thereby increasing its apparent solubility.82

The surface area for the solid state available for dissolution can be in-creased, and the dissolution layer thickness can be reduced, by reducing the particle size.77,83 This can be performed by mechanical grinding that results in a particle diameter down to about 2-3 µm. Particles with a diameter below 1 µm are called nanoparticles and they can be manufactured using, for example, wet-milling with beads or controlled precipitation. In addition to affecting dis-solution rate, nanoparticles are suggested to diffuse across the aqueous bound-ary layer and thereby locally increase the concentration at the intestinal epi-thelial cell wall, which is the driving force for permeation.72 In should be noted that reduction of the particle size generally increases the cohesive forces be-tween the particles, and formulations strategies to reduce particle aggregation is usually needed.84

Low permeability drugs During the last three decades there have been several attempts to develop oral formulations containing pharmaceutical excipients with the purpose of in-creasing the intestinal transport of low-permeability drugs.85-87 These attempts have all been unsuccessful, which is attributed to a low in vivo effect of the permeation enhancers, a high variability in the drug absorption enhancement, and safety issues.88,89 Nevertheless, there is a growing interest in these types of technologies in the last decade, as the cost of developing and manufacturing therapeutic peptides has plummeted. Peptides are generally not administered

23

orally due to their very low intestinal permeability, and their instability in the intestinal lumen. Advocates of intestinal absorption enhancing technologies are hopeful that novel formulation strategies will change this.90

For an absorption-modifying excipient (AME) to effectively increase in-testinal drug absorption, it must render the intestinal barrier more permeable. This can be achieved by altering the fluidity of the intestinal epithelial cell lipid bilayer; and/or by modulating the tight junction proteins to increase the size of the paracellular space; and/or by degrading or reducing the viscosity of the intestinal protective mucus.91 Membrane fluidity is primarily affected by surface active agents when incorporated into the lipid bilayer.92 Tight junc-tion modulators can act either directly on the tight junction structure, or indi-rectly, by affecting the intracellular signaling mechanism(s) and/or cytoskele-tons involved in regulation of tight junctions.91 The mucus barrier can be re-duced by administering a mucolytic agent. In addition, intestinal peptide ab-sorption can be increased by incorporating an agent that reduces the luminal degradation of the peptide.91 This agent can either directly inhibit luminal en-zymes, or change the microclimate pH around the formulation, which affects the activity of luminal enzymes.90,93

The in vivo ability of an AME to increase intestinal drug absorption is, however, not only about its ability to affect the intestinal barrier. The effect on the barrier must be rapid, and large enough, to allow sufficient drug ab-sorption before the drug has been transported from the luminal area with the compromised intestinal barrier.91 At the same time, the effect must be rapidly reversible, and not too potent, so that translocation of harmful luminal con-tents over the mucosa is still restricted.88 The impact of the prandial state (fed/fasted) must also be taken into consideration, as this can have a substan-tial impact on the effect of an AME on the mucosal membrane.91,94

Methods for studying intestinal drug absorption A variety of models are used in mechanistic intestinal absorption studies and in the drug discovery/development process for predicting human intestinal drug absorption. These models can be in silico, in vitro, in situ, or in vivo. The models vary in degree of complexity, and choice of a model system depends on the question to be investigated and the stage of drug development. More simple models with high-throughput capacity, such as quantitative structure–activity relationship models, and cell-monolayer models, are typically used in early drug development, whereas more complex animal and in silico models are used in a later stage of non-clinical, or early clinical phase, drug develop-ment, when more in vivo relevant predictions are needed.

24

In silico models An in silico method to investigate permeability is the quantitative structure–activity relationship model, which relates molecular descriptors and/or physi-cochemical properties (e.g., lipophilicity, pKa, MM) of the drug molecule to experimentally in vivo determined permeability values.95,96 This is a computa-tional approach to predict membrane permeability and is ideal in the early high-throughput drug discovery phase, because it requires only small amounts of new experimental data. However, the accuracy of the model is dependent on the statistical approach, choice of molecular descriptors, and the quality of the experimental permeability data. The quantitative structure–activity rela-tionship approach is consequently of limited used in the drug development process, and is therefore primarily used for excluding molecules with obvious permeability limitations, or when there is no alternative.97 However, due to the increase in computer power, studies of drug permeation can be performed using complex molecular simulations. These models can simulate the interac-tion between a molecule and a biological membrane, and thereby increase the mechanistic understanding of membrane transport.98,99

More complex in silico simulations can be used to estimate intestinal ab-sorption following oral administration of a drug or its formulation. These sim-ulations depend on mathematical equations that describe drug compound (e.g., solubility, log D) and drug product parameters (e.g., disintegration and disso-lution rate), physiological parameters (e.g., intestinal pH, transit times, and morphology), and the drug disposition in vivo.100 Computer simulations can be performed solely on the chemical structure, means of administration, and understanding of the human body and its mechanisms, but simulations should ideally integrate experimental in vitro and in vivo data to increase the accuracy of the simulations.101 There are several, physiologically-based pharmacoki-netic software programs commercially available that aim to predict fa follow-ing oral drug administration.100 Currently, the accuracy of these models for predicting fa based on well-characterized physicochemical and biopharmaceu-tical factors is too low to compete with conventional in vitro and in vivo stud-ies in drug development.102 However, a validated in silico model can be useful for evaluating, for instance, the impact of changes in drug formulation and drug–drug interactions, which can help guide the design of both preclinical and clinical studies.103

In vitro models The most common in vitro model for studying membrane permeability is the investigation of flux across a barrier that separates two chambers. The barrier can consist of an artificial membrane (e.g., parallel artificial membrane per-meability assay), a single layer of grown cells (e.g., Caco-2), or an excised intestinal tissue sample (Ussing chamber).96,104,105 An apparent permeability (Papp) of a molecule can be calculated by relating the mass appearing in the

25

receiver chamber at multiple time points (dM/dt) to the area of the barrier (A), and concentration in the donor chamber (Cdonor), using Eq. 6.

(6)

Papp is the intrinsic constant of a molecule that relates flux and concentra-tion gradient, and can therefore be used to predict the transport over any type of biological cell barrier by adjusting for, for instance, area, hydrodynamics, and solution media pH. In addition, the controlled conditions in a cell-based in vitro system make it ideal for mechanistic transport investigations. For in-stance, carrier-mediated transport can be investigated by comparing the transport rate between the two chambers—in both directions—by determining the Papp at different substrate concentrations, or before and after addition of a transport inhibitor.106,107 The more complex Ussing chamber system enables investigation of, for instance, gut-wall metabolism and regional intestinal per-meability.104 Limitations associated with these models are the high inter- and intra-laboratory variability, and sensitivity of the cell/tissue to preparation setup and chamber media.108,109 For permeability investigations in drug dis-covery, it is therefore recommended that relative Papp values compared to ref-erence standards be used, instead of absolute Papp values.108,110 The previously mentioned BCS system can also be used to predict in vivo drug absorption based on in vitro drug product dissolution data.14

In situ and in vivo models In situ permeability models can be designed in many ways, but they are gen-erally based on the drug disappearing from an isolated intestinal segment. This segment can be continuously perfused, as in the single-pass intestinal perfu-sion (SPIP) model, or be a closed-off segment, as in the closed-loop Doluisio model.111 Peff is calculated in different ways depending on the hydrodynamics in the specific model. For instance, the SPIP model calculates the Peff using Eq. 7, as there is an exponential decrease in drug concentration in the perfused segment.112,113

ln (7)

where Qin is the perfusion rate, Cin and Cout are the concentrations entering and leaving the perfused segment corrected for fluid transport, and A is the surface area of the perfused segment. Peff is consequently luminal CL per sur-face area (volume/time/area).

The SPIP model is generally used in a later stage of drug development, when more in vivo relevant data are needed. One major advantage is that SPIP enables mechanistic evaluations of drug absorption and physiological effects, as intestinal morphology and blood flow are kept intact and are combined with controlled luminal conditions.114

26

Classical in vivo PK models in which drug solutions or formulations are dosed orally, or directly into the stomach or intestine of humans or animals, can also be used to investigate permeability.115 The absolute (or relative) F, or fa (or appearance rate) of a drug, is then determined and compared to other drugs/formulations.116 Such models are the most clinically relevant ones be-cause physiological factors, such as gastric emptying time, luminal water con-tent and drug degradation, and post-absorption metabolism, affect the deter-mined parameters. These types of models are hence less useful for mechanistic studies of intestinal absorption, as the relative impact of the different factors can be difficult to determine.

27

Aims of the thesis

The general aims of this thesis were to: i) improve the mechanistic understand-ing of intestinal drug absorption; ii) study the effect of absorption-modifying excipients; and iii) evaluate common preclinical absorption models used in the evaluation of both regional intestinal differences in absorption and absorption-modifying excipients.

The specific aims of the thesis were as follows:

The aims of Paper 1 were to for the first time determine human in

vivo reference values of the regional intestinal permeability in all parts of the intestines and to validate the deconvolution-permeabil-ity model.

The aim of Paper 2 was to evaluate the usefulness of the dog trans-abdominal stoma model for prediction of regional intestinal perme-ability in human.

The aims of Paper 3 were to improve the mechanistic understand-ing of four commonly used AMEs and to evaluate their effect in the rat SPIP model.

The aim of Paper 4 was to evaluate the AMEs from Paper 3 in the more in vivo relevant rat and dog intraintestinal bolus models.

The aim of Paper 5 was to investigate how the physiological regu-lation of mucosal permeability induced by low luminal hypotonic-ity was affected by AMEs with different modes of actions, and by intestinal neural activity.

The aims of Paper 6 were to investigate if a shorter mucosal expo-sure time in the SPIP model could predict the previously observed lower effect of two AMEs in the rat and dog bolus models in Paper 4, and to investigate the recovery time of the intestinal mucosa fol-lowing AME exposure.

28

Methods

Study chemicals Model compounds Six model compounds were investigated in this thesis: acyclovir, atenolol, en-alaprilat, ketoprofen, metoprolol, and phenol red (Table 2). In all of the studies they were dosed as a cassette dose (i.e. a mixture of compounds) including at least three of the compounds. They were selected to be compounds with mo-lecular characteristics representative of approved small drugs, while still hav-ing a range of physicochemical properties (Table 2).117 Another selection cri-terium was that the compounds should be passively absorbed in the intestine by paracellular and/or lipoidal diffusion.

Table 2. The papers in which each of the six model compounds were investigated; their Biopharmaceutics Classification System (BCS) classification;, physicochemi-cal properties; human jejunal effective permeability (Peff) historically determined with the single-pass jejunal perfusion model; fraction excreted unchanged in the kidneys (Fe); and reported metabolic enzyme and transporter affinity.53,56

Model compounds Acyclovir Atenolol Enalaprilat Ketoprofen Metoprolol Phenol Red Papers 3-6 1-6 2-6 1-6 1,2,4-6 3-6 BCS class III III III II I n/a MM (g/mol) 225 266 348 254 267 354 pKa 2.19a/9.25a 9.6b 3.17b/7.84a 3.89a 9.6b 7.9a PSA 117 88.1 102 54 58 92 HBA/HBD 7/3 4/4 6/3 3/1 4/2 5/2 Log P -1.8 0.18 -0.13 3.37 2.07 3.02 Log D7.4 -1.8 -2 -1 0.1 0 n/a Log D6.5 n/a <-2.0 -1 0.8 -0.5 n/a Human Peff no data 0.2 0.2 8.7 1.3 no data Fe >0.9118 >0.9119,120 1.0121 <0.05122,123 <0.09124,125 0.75126 Transporter OAT137 OCT2127 – – – – Metabolism ADH – – Gluc19,20 CYP2D649 Gluc128 aacid; bbase; ADH = alcohol dehydrogenase; CYP = Cytochromes P450; Gluc = glucuronida-

tion; HBA/HBD = hydrogen bond acceptor/donor; Log P/D7.4/6.5 = n-octanol−water partition

coefficient at pH 7.4/6.5; MM = molar mass; OAT1 = organic anion transporter 1; OCT2 =

Organic cation transporter 2; pKa = dissociation constant; PSA = polar surface area.

29

Absorption-modifying excipients The molecular properties and proposed mechanisms of action of the four ab-sorption-modifying excipients (AMEs) evaluated in Papers 3-6 are presented in Table 3, and illustrated in Figure 5.

Table 3. Molecular properties and mechanism of action of the absorption-modifying excipients (AMEs) investigated in Papers 3-6.

AME Properties MM Mechanism(s) of action Paper

Caprate fatty acid 172.3 Da TJ modulation, membrane fluidity129 3–5

Chitosan polysacharide 40-300 kDa TJ modulation130 3–6

NAC cysteine derivate 163.2 Da mucolytic agent131 3

SDS anionic surfactant 288.4 Da membrane fluidity, solubilization92 3–6

NAC = n-acetylcysteine; MM = molar mass; SDS = sodium dodecyl sulfate, TJ = tight junction

The AMEs were selected on the basis of their previously reported absorption enhancing effect in various preclinical in vitro and in vivo absorption models. The aim was to select a set of AMEs with different mechanisms of action to evaluate their impact on the absorption rate of the model compounds.

Figure 5. Mechanism(s) of action of the four absorption-modifying excipients inves-tigated in Papers 3–6. Solid/dashed blue line = Proposed/potential mechanism.

Description of study subjects and animals Studies were performed with humans, dogs, rats, or rat tissue samples, in this doctoral thesis. A human clinical intraintestinal bolus study was performed in Paper 1. Dog in vivo intraintestinal bolus studies were performed in Papers 2 and 4. Rat in vivo intraintestinal bolus and perfusion studies were performed

30

in Papers 3-6. An in vitro Ussing study on rat intestinal tissue samples was performed in Paper 4. Some details of the studies are described below.

Clinical Study Paper 1 included 14 healthy male and female volunteers between 18−55 years of age and with a body mass index between 19−30 kg/m2. The study was ap-proved by the Medical Products Agency (No. 5.1-2015-49712), the regional ethics committee for human research (No. 2015/243), and the radiation pro-tection committee (No. D15/27), Uppsala, Sweden. The study was reported in EudraCT (2015-000256-22). All subjects signed an informed consent form and were judged healthy by a clinical physician before the study. The study was conducted by Clinical Trial Consultants AB at Uppsala University Hos-pital in accordance with the Declaration of Helsinki.

Dog Studies Paper 2 and 4 included three and four male Labrador dogs, respectively. They weighed between 35-39 kg and were between 2-5 years old. The studies were approved by the local ethics committee for animal research (no: 34-2015) in Göteborg, Sweden. The studies were conducted at AstraZeneca R&D, Göte-borg, Sweden.

Rat Studies Papers 3–6 included at least 8 weeks old male Wistar Han rats (strain 273) weighing between 280–483 g. The studies were approved by the local ethics committees for animal research in Uppsala (no: C64/16) and Göteborg (no: 48-2013), Sweden. The rat Ussing study was conducted at AstraZeneca R&D, Göteborg, Sweden, and the rat perfusion and bolus studies at the department of Neuroscience, Uppsala University, Sweden.

Drug compound interactions in the rat Ussing chamber The previously described six model compounds were administered in all stud-ies as a cassette dose containing three to six of the compounds. The cassette dose was therefore evaluated and presented in Paper 4 by comparing the drug compound (i.e., not phenol red) permeabilities when they were administered simultaneously, or individually. The aim was to identify any possible drug-drug interactions in absorption rate.

The Ussing experiment was performed by mounting rat jejunal specimens between two chambers containing Krebs-Ringer buffer (37 °C). The pH was 6.5 on the mucosal side and 7.4 at the serosal side. The viability of the tissue

31

was monitored by measuring the electrical resistance (R) and potential differ-ence (PD) across the specimens. Those with R and PD values below 30 Ohm×cm2 and 4 mV, respectively, were replaced. The specimens were ini-tially allowed to equilibrate for 20 min, and at time zero the mucosal solution was replaced with one of the five model drugs, individually, or as a cassette dose with all five together (atenolol, ketoprofen, metoprolol, and acyclovir or enalaprilat). Samples of 200 µL were thereafter taken for drug quantification from the serosal side at 0, 30, 60, 90, 120, and 150 min, and from the mucosal side at 0 and 150 min. Sampled volumes were compensated for by fresh Krebs-Ringer buffer. Samples were immediately frozen and stored at −20 °C awaiting bioanalysis.

Regional intestinal permeability studies Clinical study The clinical human study in Paper 1 was performed to determine relevant regional intestinal permeability in vivo data for three of the model drugs (atenolol, ketoprofen, and metoprolol), at controlled conditions in the same subjects. The study was an open, crossover trial, where the compounds were dosed to three groups on four occasions in a randomized order. The study schedule is presented in Table 4.

Table 4. Study schedule describing the administration order of three model drugs to 14 volunteers.

Groups Study Occasions

1 2 3 4

1 Intravenous Jejunum Ileum Colon

2 Intravenous Colon Jejunum Ileum

3 Intravenous Ileum Colon Jejunum

The model drugs were cassette dosed as a solution (290 mOsm, pH 6.8), in-travenously (iv), and also intraintestinally directly into the jejunum, ileum, and colon of fasted subjects. The iv administration generated individual reference PK data, which was used in the calculation of intestinal absorption rate (de-scribed in detail in data analysis). The intraintestinal administrations were per-formed using a capsule (Bioperm AB, Lund, Sweden) connected to a nasal tube. The cassette-dose solution (10 mL) was administered once the capsule was at the intended intestinal location, which was verified by length of the tube and by x-ray. Venous blood was sampled immediately before the iv and intraluminal administrations, and at 5, 10, 20, 30, 40, 50, and 60, 120, 240, 360 min (also at 8, and 24 h following the iv administration). The blood sam-ples were put on ice and centrifuged (3000 × g, 10 min at 4 °C) within 20 min.

32

Fifty microliters of the plasma was transferred to a 1 mL 96-well plate (Thermo Scientific), and the remaining plasma was transferred to 2 mL tubes (Sarstedt) for storage. The plasma samples were frozen and stored at −20 °C until analysis.

Dog study The study in Paper 2 was performed to evaluate the dog as a preclinical model for prediction of regional intestinal permeability in human. The study was similar to the human clinical study, with some exceptions. For instance, four model drug compounds were tested (enalaprilat, atenolol, ketoprofen, and metoprolol). The cassette dose was administered intravenously, and intraintes-tinally into the jejunum and colon (not the ileum). The cassette dose was fur-thermore administered twice into each intestinal segment to evaluate intrain-dividual variability in permeability. Each study dog had been previously in-serted with permanent trans-abdominal access ports, one in the jejunum and one in the colon. The cassette dose was thus administered directly through these ports, instead of a capsule connected to a nasal tube. Venous blood was sampled and handled in the same way as in the human study in Paper 1.

Absorption-modifying excipient studies SPIP studies in rat Surgery and experimental setup The AME studies in Papers 3, 5, and 6 all used the same rat SPIP model. The experimental setup is described briefly here. The rats were initially anaesthe-tized by an intraperitoneal injection of a 10% w/v inactin solution (160 mg/kg). Body temperature (37.5 °C) and free airways were maintained, and the systemic arterial blood pressure was continuously recorded by a catheter in the femoral artery to validate the condition of the animal. The abdomen was thereafter opened with a 3-5 cm longitudinal midline incision and the bile duct was cannulated. Thereafter, a jejunal segment of about 10 cm was cannulated with a silicone tube in both ends. The segment was covered in polyethylene wrap to avoid fluid loss, and thereafter placed on the outside of the abdomen of the rat to enable visual monitoring of intestinal motility and perfusate flow.

After the surgery a 51Cr-EDTA solution was administered as an iv bolus (75 µCi) and continued by a continuous infusion (50 µCi/h) during the whole experiment. The jejunal segment was then perfused with phosphate buffered saline (6 mM, pH 6.5) for 30 min, to stabilize 51Cr-EDTA activity in the plasma. The experimental period started after the 30-min stabilization, and the perfusate was changed to the isotonic control solution, which was phosphate buffered saline (6 mM, pH 6.5) containing all six model compounds at 50 µM.

33

Perfusate flow rate was 0.2 mL/min throughout the stabilization and experi-mental periods. During the perfusions, blood (femoral artery) and outgoing perfusate were sampled at 15-min intervals for up to 135–180 min after start of the experimental period. The blood samples were centrifuged, and the blood plasma and perfusate were stored at -20 °C until analysis.

In Papers 3, 5 and 6, the experimental period was divided into two 75-min parts: i) perfusion of the segment with an isotonic control solution (containing model compounds but no AME), directly followed by ii) perfusion of the seg-ment with one of the 25 test formulation described in Figure 6 and Table 5.

Paper 5 also contained two evaluations of the effects of the AME exposure time on the mucosal barrier integrity. An exposure time of 15-min was se-lected to account for the intestinal exposure time in vivo, because the small intestinal transit of fluid is about 3 h, and divided into distinct motility phases.28,29,132,133 The 60-min exposure time was selected to account for an ex-posure scenario in which GI motility is compromised, for example during gas-troparesis.134 Long exposure times may also be a consequence of formulation strategies, such as mucoadhesion, or because of enteral administration of the drugs, which is common for patients with dysphagia.135-137 The 15- and 60-min evaluations were divided into three parts as follows: i) an initial perfusion with the isotonic control solution for 60 min; ii) a 15- or 60-min perfusion of one of the AME-containing solutions (Table 5); and iii) subsequent perfusion with the isotonic control solution for 60 min. Each perfusion period was started with a rapid filling of the whole segment with the perfusion solution.

Finally, to generate reference PK data, the model compounds were cassette dosed as an iv solution in Paper 3. These reference PK data were used in the calculation of rat intestinal absorption rate in Papers 3-6 (see Data Analysis).

Study formulations In Papers 3, 5 and 6, a total of 25 formulations were tested. These formula-tions are described in Table 5 and Figure 6.

Nine test solutions (I–IX) and a control solution, all at isotonic conditions (290 mOsm), were evaluated in Paper 3. The test solutions contained one of the four AMEs (caprate, NAC, chitosan, or SDS) described in Table 3. The effect of the combination of chitosan and NAC was also investigated (VII). Each AME was evaluated at two clinically relevant luminal concentrations, between 0.4 and 5 mg/mL, where 5 mg/mL is a dose of 1 g dissolved in 250 mL water.14

In Paper 6, the control solution, as well as the 5 mg/mL dose of SDS and chitosan, were evaluated after addition of the nicotinic receptor antagonist, mecamylamine (0.1 mM), to the perfusate (X, XI, XII).138 This tested if local enteric nerve activity had any effect on the basal intestinal permeability, or on the effect of the two AMEs. The time-dependent effects of 1 and 5 mg/mL chitosan and SDS were also evaluated after both 15-min (V, VI, VIII, IX) and 60-min (VI, IX) of mucosal exposure.

34

Table 5. The control solution and 25 experimental test formulations investigated in the evaluation of absorption-modifying excipients (AMEs) in the rat SPIP model in Paper 3, 5, and 6. All solutions contained a cassette dose of acyclovir, atenolol, en-alaprilat, ketoprofen, metoprolol, and phenol red, at 50 µM, in phosphate buffer (8 mM, pH 6.8).

Conditions AME concentration

(mg/mL) Denotation

Osmolarity (mOsm)

Paper

– – Control 290 3, 5, 6

Caprate 0.4 I 290 3

Caprate 2 II 290 3

NAC 1 III 290 3

NAC 5 IV 290 3

Chitosan 1 V 290 3, 6

Chitosan 5 VI 290 3, 6

NAC+Chitosan 5+5 VII 290 3

SDS 1 VIII 290 3, 6

SDS 5 IX 290 3, 6

Mec – X 290 6

Chitosan+Mec 5 XI 290 6

SDS+Mec 5 XII 290 6

Caprate^ 1 XIII 170 5

Caprate 2 XIV 170 5

Caprate^ 5 XV 170 5

Caprate+Mec 2 XVI 170 5

Caprate+Par iv 2 XVII 170 5

Chitosan 5 XVIII 170 5

Chitosan+Mec 5 XIX 170 5

SDS 5 XX 170 5

SDS+Mec 5 XXI 170 5

– – XXII 170 5

Par iv – XXIII 170 5

– – XXIV 100 5

Mec – XXV 100 5

^pH 7.4; Mec = luminal mecamylamine; NAC = n-acetylcysteine; Par iv = Pretreatment with

intravenous parecoxib; SDS = sodium dodecyl sulfate.

In Paper 5, the effect of luminal hypotonicities (100 and 170 mOsm) on in-testinal permeability were evaluated. The hypotonic test solutions were eval-uated individually (XXII, XXIV). The 170 mOsm solution was also evaluated in combination with luminal mecamylamine (XXV) that blocks enteric nerve activity, and intravenous parecoxib (XXIII) that induce post-surgery enteric nerve activity.139 The combined effect of hypotonicity and caprate, chitosan,

35

and SDS, at different concentrations (1–5 mg/mL) was also evaluated (XIII, XIV, XV, XVIII, XX), as well as the effect of mecamylamine on the combined effect (XVI, XIX, XXI). Finally, the effect of intravenous parecoxib in com-bination with caprate and hypotonicity was evaluated (XVII).

Only the key findings from the investigations of the 25 different test for-mulations in Table 5 and Figure 6 are discussed in this thesis.

Figure 6. The control solution and 25 experimental test formulations investigated in the evaluation of absorption-modifying excipients (AMEs) in the rat SPIP model in Paper 3, 5, and 6. All solutions contained a cassette dose of acyclovir, atenolol, en-alaprilat, ketoprofen, metoprolol, and phenol red, at 50 µM, in phosphate buffer (8 mM, pH 6.8).

Intraintestinal bolus studies in rat and dog Surgery and experimental setup Paper 4 investigated the effects of AMEs in the rat and dog intraintestinal bolus models. The bolus model is a more clinically relevant in vivo model than the SPIP model, as luminal dilution of AME, intestinal transit, and luminal colloidal structures, affect the mucosal barrier integrity.

In the rat bolus experiments in Paper 4, the rats were prepared in the same way as in the perfusion studies, with the exception that there was no cannula-tion of the bile duct and jejunum, and there was no iv administration of 51Cr-EDTA. Once the abdomen was opened, the stomach was punctured, and a silicone tube was inserted into the duodenum where the formulation was in-jected as a 0.5 mL bolus. The effect of the rat surgery on the intestinal transit time was evaluated prior to the rat bolus study. A dye, methylene blue, was injected with the control solution in nine rats. At 20, 45, and 120 min, three rats were euthanized, and the front of the dye in the intestine was determined.

36

The dog experimental and administration procedure in Paper 4 was the same as in the regional intestinal permeability study in Paper 2, with the dif-ferences that four dogs (one additional) were investigated, formulations were only administered to the proximal small intestine, and the administered vol-ume was 20 mL.

Blood was sampled immediately before injection of the intraintestinal bo-lus, and during 2 h for rat, and during 6 h for dog (at 5, 10, 20, 30, 45, 60, 90, 120, 240, 360 min). The blood samples were centrifuged, and the blood plasma and perfusates were stored at -20 °C until analysis.

Study formulations In Paper 4, one isotonic control solution and seven isotonic test solutions were investigated at pH 6.5 (Table 6). Each AME was investigated at two clinically relevant concentrations (2.3 and 11.4 mg/mL) in rats, representing doses of 0.2 and 1.0 g administered to a human of 70 kg. The 11.4 mg/mL dose caprate was also investigated at pH 7.4, where caprate was fully dissolved. In dog, the control solution, and the 11.4 mg/mL dose of SDS and caprate (pH 7.4), were investigated. Each investigated formulation contained the six model com-pounds previously described.

Table 6. Description of the eight formulations administered as an intraintestinal bo-lus to rat (0.5 mL) and dog (20 mL) in Paper 4. All formulations contained a cas-sette dose of acyclovir, atenolol, enalaprilat, ketoprofen, metoprolol, and phenol red, and different concentrations of absorption-modifying excipients (AMEs).

AME AME concentrations (mg/mL)

Rat doses (mg) Dog doses (mg) pH

Control 0 0 0 6.5

Caprate 2.3 1.15 - 6.5

11.4^ 5.7 - 6.5

11.4 5.7 228 7.4

Chitosan 2.3 1.15 - 6.5

11.4 5.7 - 6.5

SDS 2.3 1.15 - 6.5

11.4 5.7 228 6.5

^ = suspension; SDS = sodium dodecyl sulfate

Bioanalytical method All bioanalyses for the determination of model compound concentrations in plasma in Papers 1–6 were performed with the same LC-MS/MS method. The limit of quantification (LOQ) was 0.2 nM for all six model compounds, and the standard deviation for the lower limit of quantification was <15%. Analy-sis of model compounds in the intestinal perfusates in the rat SPIP studies in

37

Papers 3, 5, and 6 was not possible due to interaction between the AMEs and the MS and LC equipment.

Data analysis Apparent permeability (Papp) in the Ussing chamber The apparent permeability (Papp) of the six model compound in Paper 4 was calculated using Eq. 6. The Papp was determined with a mucosal area of 1.14 cm2, and when the compound concentration increased linearly in the receiver chamber.

Intestinal effective permeability (Peff) and absorptive flux (Jabs) The deconvolution model was used to calculate: i) the model compound ef-fective permeability (Peff) in the human, rat, and dog intraintestinal bolus stud-ies in Papers 1, 2, and 4; and, ii) the model compound absorption flux (Jabs) in the rat AME SPIP studies in Papers 3, 5, and 6.115

First-order two-compartment disposition PK parameters (a, b, α, β) were determined in WinNonlin software version 6.3, following the iv administra-tions of the model compounds to human, rat, and dog rat, in Papers 1–3. These parameters were used in the determination of absorption rate following the intraintestinal administrations of the same model compounds.

An input rate (amount/time) of drug into plasma was determined by decon-volution of the plasma concentration–time curves following the different in-testinal bolus administrations and perfusions. The disposition PK parameters (a, b, α, β) used as weighting functions in the deconvolution were from the same subject for human and dog, and from the same strain for rat. An absorp-tion rate was then calculated by compensating for the first-pass extraction, which was based on the reported data on fraction metabolized of each com-pound, and plasma CL values from the iv administration in each subject (or rat strain).

The deconvoluted Peff was calculated by relating the absorption rate to the remaining amount of model compound in the intestinal lumen (i.e. the yet un-absorbed amount following the dose) following the bolus administration, and to the intestinal luminal radius of each species in the relevant intestinal seg-ment (Eq. 8).

(8)

The mean Peff of the model compounds during the first 30 min following bolus administration was taken to be representative in each subject.

38

The deconvoluted Jabs from the AME rat SPIP studies was calculated by relating the absorption rate to the intestinal luminal area (A) in the perfused segment (Eq. 9).

(9)

After an initial 15-min equilibration period, the mean Jabs of the model com-pounds during each of the two perfusion periods (75+75 min) was taken to be representative in each rat (Papers 3, 5, and 6). The Jabs from 0 min, to 150 or 180 min, was evaluated in the exposure time experiments in Paper 6.

Intestinal 51Cr-EDTA clearance (CLCr) The blood-to-lumen CLCr was determined in all AME rat SPIP studies in Pa-pers 3, 5, and 6. This was done by analyzing the 51Cr activity (cpm) in all outgoing luminal perfusates, and in the first and last plasma sample in each experiment. The CLCr (mL/min/100 g wet intestinal tissue weight) was calcu-lated using Eq. 10.140

100 (10)

where Cperfusate and Cplasma is the activity in the perfusate and plasma (cpm/mL), and Qin is the perfusate flow rate (mL/min). The mean CLCr during the two perfusion periods (75+75 min) was taken to be representative in each rat in Papers 3, 5, and 6, after an initial 15-min equilibration period. CLCr from 0 min, to 150 or 180 min was evaluated in the exposure time experiments in Paper 6.

Statistical analysis Peff, Papp, CLCr, and Jabs values, are expressed as mean ± standard deviation (SD) or standard error of the mean (SEM), or as median (range). The mean (±SEM) CLCr and Jabs ratio between the 60 min control and 60 min test period in the rat perfusion studies in Papers 3, 5, 6 are also presented (Eq. 11).

(11)

The mean of two groups was compared using the student’s t-test (paired or unpaired) with the Benjamini–Hochberg multiple t-test correction. Multiple comparisons between groups were performed using a one-way ANOVA with a post-hoc Tukey’s or Dunnett’s multiple comparison test. Log transformation of values was performed when the original measured data were heteroscedas-tic and not normally distributed; this was investigated using the Bartlett test and the quantile-quantile plot. Comparison between CLCr and model com-pound Jabs ratios was performed using linear regression.

39

Results and discussion

Transport and metabolic interactions of the model drugs There were no significant differences in Papp for any of the five model drugs when evaluated individually or as a cassette dose (atenolol, ketoprofen, metoprolol, and acyclovir or enalaprilat) in the rat Ussing model in Paper 4 (Table 7).

Table 7. Mean (±SD) apparent rat jejunal permeability (Papp) of five model drugs (10 µM) when determined individually, or as a cassette dose of atenolol, metoprolol, ketoprofen, and acyclovir or enalaprilat, in the Ussing chamber (n ≥ 4).

Model drugs Model drug Papp (cm/s ×10-6)

Individually Cassette dose

Acyclovir 2.0 ± 1.3 3.5 ± 1.3

Atenolol 2.3 ± 1.1 2.9 ± 0.96

Enalaprilat 1.6 ± 0.66 2 ± 0.37

Ketoprofen 24 ± 6.9 26 ± 6.9

Metoprolol 6.8 ± 2.8 8 ± 4.3

The model compounds in this study were initially selected on the basis of their passive intestinal transport mechanisms (paracellular and/or lipoidal transcel-lular diffusion). This was to avoid any concentration dependent transport, or drug-drug transport interactions. The passive absorption is illustrated by the (clinical) dose-linear absorption rates in humans for atenolol (25-200 mg), acyclovir (200-600 mg), metoprolol (50-200 mg), and ketoprofen (12.5-100).141-144 The contribution of CM transport is negligible for atenolol, metoprolol, and ketoprofen in Caco-2 cell monolayers; furthermore, the intes-tinal permeability of atenolol, enalaprilat, metoprolol, and ketoprofen is con-centration independent in both the rat SPIP (0.01–10 mM) and Ussing models (0.1–5 mM).107,145-148 Atenolol is the only one of the model compounds with a proposed affinity for an intestinal transporter (OATP1A2), as the plasma ex-posure of atenolol decreases during coadministration with apple or orange juice.21,22 However, juice studies are non-mechanistic PK investigations for which GI transit, osmolarity, and pH effects cannot be disregarded. In conclu-sion, reported in vitro transport data, reported dose proportional plasma PK

40

data, and the Ussing data, indicate that there is no impact from intestinal trans-porters on the absorption rate of any of the model compounds at the concen-trations used in our studies.

Affinity for excretion transporters and/or metabolic enzymes have been re-ported for several of the model compounds (Table 2).37,49,122,123,125,127,128,149 There are reports of some CM renal secretion acyclovir (OAT1) and atenolol (OCT2/MATE1) in human and rat; extensive CYP2D6 metabolism of metoprolol in rat, human, and dog; and some hepatic glucuronidation of keto-profen and phenol red in human and rat. We consider that the risks for any elimination interactions between acyclovir, atenolol, metoprolol, ketoprofen, and phenol red are minimal, since different transporters or metabolic enzymes are involved for these compounds. Competitive inhibition in the hepatic glu-curonidation of ketoprofen and phenol red is, however, possible, but the CL (and Ffirst-pass) of these drugs are low in both rat (10 mL/h) and dog (4.4 L/h). Any potential metabolic drug–drug interactions will consequently have lim-ited effect on the deconvoluted absorption rate values.

Finally, the model compounds were administered as a cassette dose on all occasions in each study in this doctoral thesis. Any potential interacting effects (transport or metabolic) would consequently be similar between study groups, and would therefore have a limited impact on the conclusions made from the studies.

Regional intestinal permeability Human clinical study The small intestine is the primary intestinal region for drug absorption. How-ever, colonic drug absorption can be substantial for drugs formulated into modified-release dosage forms, and potentially also for low permeability drugs (i.e. BCS class III and IV) incompletely absorbed in the small intes-tine.141,150-152 Currently, most in vivo determinations of human intestinal Peff using the SPIP model have focused on the small intestine, as colonic intuba-tion is impractical.56,73,115,153 An alternative to the SPIP model is to determine intestinal Peff values from plasma appearance data following intraintestinal drug administrations, using the deconvolution-Peff model.115 However, this model has not been properly validated in the same subjects under controlled experimental conditions.

The objectives in Paper 1 were to determine the regional intestinal perme-ability of three model drugs, and to validate the deconvolution-Peff model.

The determined human regional intestinal Peff values are presented in Table 8. The Peff was substantially higher (> 10-fold) in the small intestinal regions than in the colon for the low permeability drug, atenolol. In contrast, the Peff was high (> 3.4 ×10-4 cm/s) in all intestinal regions for the high permeability

41

drug, ketoprofen, and medium-to-high (0.75 to 1.7 ×10-4 cm/s) for the high permeability drug, metoprolol. This confirms the conclusion from PK studies evaluating plasma exposure, that a compound with a medium-to-high small intestinal permeability is suitable for formulation into a modified-release dos-age form, as its permeability is often adequate in all intestinal regions.150,154 However, for the first time this has been shown for regional intestinal perme-ability; in PK studies one cannot readily identify whether the difference in exposure comes from differences in, for example, permeability, transit, or sol-ubility.

Table 8. Median (range) human (n=14), regional intestinal effective permeability (Peff) of three model drugs.

Model compounds

Human regional intestinal Peff (×10-4 cm/s)

Jejunum Ileum Colon

Atenolol 0.45 (0.1-1.5)* 0.15 (0.1-0.4)* 0.01 (0.01-0.1)

Ketoprofen 8.8 (0.5-16.1)* 6.5 (1.5–10.4) 3.4 (1.0-9.1)

Metoprolol 1.7 (0.02-3.4) 0.72 (0.1-7.5) 1.3 (0.5-7.7)

*Statistically significant difference in Peff compared to the colon. The doses were: atenolol 10

mg, ketoprofen 5 mg, and metoprolol 25 mg.

A comparison of the jejunal Peff values derived using the deconvolution-Peff model in Paper 1, with those previously determined with the SPIP model, showed a close association for all three model compounds (Figure 7). These results strongly suggest that the deconvolution-Peff model is suitable for deter-minations of Peff values in all parts of the intestinal tract.

Figure 7. Comparison of human jejunal effective permeability (Peff) of atenolol, metoprolol and ketoprofen determined using the deconvolution-Peff model (Table 8) with those historically determined (Table 2) using the SPIP model.

42

Dog study The dog absorption model is commonly used in the pharmaceutical industry as dogs have an intestinal anatomy and physiology similar to humans.155,156 The size of dogs also enables oral administration of full-dose, clinical formu-lations, which makes the model especially suitable for evaluation of modified-release formulations that release drug in all parts of the intestinal tract.116,149,157 However, based on PK data from oral and intraintestinal drug administrations, the dog has an intestinal mucosa that is more permeable to drug molecules than human, but there are no direct determinations of regional intestinal dog Peff.158-160

The aims of Paper 2 were to determine the regional intestinal permeability of four drug molecules, and to compare them to the human regional intestinal Peff values from Paper 1 at the same doses (mg/kg).

The determined dog regional intestinal Peff values are presented in Table 8, which also includes the dog jejunal Peff values of acyclovir and phenol red from Paper 4. There were no significant differences in regional intestinal per-meability for any of the investigated drugs. Nonetheless, there was a trend for a higher (>5-fold) permeability in the jejunum than in the colon for the low permeability drugs atenolol and enalaprilat. There were no regional differ-ences for the high permeability drugs metoprolol and ketoprofen.

Table 9. Mean (±SD) dog (n=3–4), regional intestinal effective permeability (Peff) of six model compounds.

Model compounds Dog regional intestinal Peff (×10-4 cm/s)

Jejunum Colon

Acyclovir 0.93 ± 0.3 –

Atenolol 0.62 ± 0.77 0.13 ± 0.08

Enalaprilat 0.14 ± 0.19 0.02 ± 0.01

Ketoprofen 3.7 ± 1.4 2.2 ± 1.3

Metoprolol 1.1 ± 1.26 1.0 ± 0.73

Phenol Red 0.09 ± 0.02 –

The doses in dog were: acyclovir 5 mg, atenolol 5 mg, enalaprilat 20 mg, ketoprofen 2.5 mg,

metoprolol 12.5 mg, and phenol red 5 mg. There were no statistical differences in Peff.

A comparison between the human and dog jejunal and colonic Peff values from Papers 1 and 2 is presented in Figure 8. In the jejunum, the differences in Peff between the human and dog were small for all three model drugs, but the hu-man tended to a have a higher permeability (1.7 to 3.4-fold). The differences between human and dog Peff were also small (<1.5-fold) in the colon for the medium-to-high permeability drugs, metoprolol and ketoprofen. The low per-meability drug atenolol, on the other hand, was 10 times more permeable in the dog than in the human colon. Drugs have higher intestinal absorption and bioavailability in dogs than humans. This has been attributed primarily to the

43

colon, because the jejunum is more similar in the two species.160 This assump-tion is supported by the data from this dog study.

Figure 8. Comparison of effective jejunal and colonic Peff values determined using the deconvolution-Peff model in human and dog (Tables 8 and 9).

In summary, the results from Papers 1 and 2 suggest that the dog intestinal absorption model is well suited for evaluation of drug products containing medium-to-high permeability drugs without solubility problems (BCS class I and II). However, care should be taken when interpreting absorption data in dog of low permeability drugs (BCS class III), as the absorption can be sub-stantially higher than in human. This is especially important to consider if in-vestigating novel formulations intended for colonic drug release.

Absorption-modifying excipients A critical AME can potentially increase intestinal drug absorption by affecting the mucosal barrier properties. An AME-containing drug product may conse-quently affect the bioequivalence of a novel generic drug delivery system, or the absorption of co-administered drug compounds.90,131,161,162 In addition, AMEs can be used in enabling formulations, with the objective of increasing the intestinal absorption of low permeability drugs, such as BCS class III and IV drugs, peptides, and nucleotides.85,86,91 The objectives of Papers 3–6 were to systematically investigate the mechanism of action, effect, and safety, of AMEs, as well as the accuracy of three common preclinical absorption models for predicting in vivo effects of AMEs.

Effects on the transport of high permeability drugs The transport of the model drugs with a medium-to-high intestinal permeabil-ity, metoprolol and ketoprofen, were unaffected regardless of transport model

44

(SPIP, bolus), species (rat, dog), AME investigated, or AME concentra-tion/dose. This was in agreement with the hypothesis that the intestinal epi-thelium is less of an absorption barrier for high permeability com-pounds.68,69,163 The results from Papers 3–6 suggest that addition of an AME in a novel drug formulation has limited effect on the absorption of BCS I and II drugs. However, it is still possible that an AME can affect the absorption rate also of compounds with higher permeability, even if its fa is unaffected.

Correlation between Jabs and CLCr in the rat SPIP model In the rat SPIP model, AME-induced increase in blood-to-lumen CLCr corre-lated closely to an increase in lumen-to-blood Jabs of the low permeability model compounds. This correlation was observed at both isotonic (Paper 3) and hypotonic conditions (Paper 5), illustrated in Figure 9. CLCr and Jabs are hence regulated largely in the same way. We therefore suggest that CLCr ratio in the rat SPIP model may be used as a surrogate variable when investigating the potential effect of an AME on drug absorption.

Figure 9. Relationship between the lumen-to-blood Jabs ratio and the blood-to-lumen CLCr ratio. Data are for caprate, chitosan, and SDS during isotonic (●, Paper 3) and hypotonic (■, Paper 5) conditions in the rat SPIP model.

Effects in the rat SPIP model at isotonic conditions The objectives of Paper 3 were to evaluate the effect potential and to increase the mechanistic understanding of four AMEs (Table 6) in the rat SPIP model. The effects of nine AME-containing test solutions (I–IX), on CLCr and Jabs ratio, relative to the control solution, are presented in Figure 10A–D.

There was an increase in Jabs and CLCr, induced by the chitosan and SDS solutions (V, VI, VIII, XI). The increase was concentration–dependent for SDS. This is consistent with the proposed effect of surfactants (e.g., SDS) on transcellular permeability; an increased surfactant concentration increases the membrane fluidity and the solubilization of membrane phospholipids.92,164

45

The positively charged ammonium groups of the chitosan molecule are pro-posed to interact directly with the cell membrane, and/or paracellular proteins, thereby increasing the paracellular permeability.130,135,165 This interaction seems to have a maximal epithelial response, as the increase in Jabs and CLCr was not dependent on concentration for chitosan, confirming the results re-ported from Caco-2 in vitro investigations.165 The maximal effect of chitosan may be restricted by intestinal mucus that limit the transport of chitosan to the epithelial membrane, as the effect of chitosan is shown to be reduced in cell monolayers covered with mucus.166 This mechanism was not supported in the more in vivo relevant SPIP model. In that model, the coadministration of NAC (mucolytic agent) with chitosan (VII) did not increase either Jabs or CLCr.

Figure 10. The relative increase in (A) CLCr ratio and (B-D) Jabs ratio during perfu-sion of nine AME-containing test solutions (NAC, SDS, caprate, chitosan, chi-tosan+NAC (ChNAC)), compared to a control solution. The dashed line illustrates no difference in CLCr or Jabs (ratio = 1). All doses are in mg/mL.

Caprate is reported in many in vitro and in vivo studies to have an intestinal absorption-enhancing effect on model compounds from about 2 mg/mL.87,129,167,168 The effect is assumed to be primarily related to its ability to modulate the tight junction proteins by a mechanism involving intracellular

46

Ca2+ that increases the paracellular permeability. We did not observe any ef-fect on Jabs or CLCr by caprate (II) at 2 mg/mL in Paper 3. A higher dose than 2 mg/mL was no possible without causing precipitation of the test formula-tion, as this is the solubility of caprate at pH 6.5. An effect could potentially have been observed if the pH was increased to 7.4, where the solubility of caprate is higher. Such effects were observed for the caprate-induced increase in rectal absorption of phenoxymethylpenicillin, when the pH was increased from 6.5 to 7.4.168

In conclusions, the proposed effects of chitosan and SDS, illustrated in Fig-ure 5, were supported by the results in this rat SPIP study. There was, however, no effect of either caprate or NAC at the concentrations and conditions eval-uated.

Effects of nerve activity and hypotonicity in the rat SPIP model Luminal conditions can affect the physiological regulation of, for instance, motility, CLCr, and luminal osmolarity by affecting the transport of osmo-lytes.169,170 These effects are primarily mediated by enteric acetylcholine and serotonin receptors, as they can be reduced by coadministration of mecamyl-amine, a nicotinic acetylcholine receptors antagonist, and ondansetron, a ser-otonin inhibitor.169,171 However, mecamylamine did not affect the basal CLCr value of the control solution, or the increase in CLCr induced by either chitosan or SDS, in Paper 3 (Figure 11). This suggests that the basal permeability in our SPIP model was not regulated by nicotinergic receptors, and neither was the permeation-enhancing effect of chitosan and SDS.

Figure 11. The relative increase in CLCr ratio compared to a control solution, (A) during perfusion of six test solutions (control, and 5 mg/mL SDS and chitosan), with (black) and without (gray), 0.1 mM mecamylamine (M), a luminal nicotinic enteric receptor antagonist, or (B) during perfusion of four test solutions (two at 170 mOsm and two at 100 mOsm). Parecoxib (Par, a COX-2 inhibitor) was administered intra-venously to one group, and mecamylamine (Mec) luminally to one. The dashed line illustrates no difference in CLCr or Jabs (ratio = 1).

47

Abdominal surgery reduces enteric neural activity, called intestinal ileus, which manifests as a reduction in motility. This also results in, for instance, a reduction in the hypotonicity-induced increase in intestinal permeabil-ity.139,170,171. The reduction of neural enteric activity must be partly mediated by cyclooxygenase-2 (COX-2)-derived intestinal prostacyclins, because it is known that parecoxib, a COX-2 inhibitor, reduces the effect of surgery on mucosal functions.172 The impact of experimental conditions on mucosal neu-ral activity, and its implications for determinations of mucosal drug transport in the SPIP model, were investigated in Paper 5.

The effects on CLCr ratio of different hypotonic luminal conditions are pre-sented in Figure 11. A luminal hypotonicity of 100 mOsm (XXIV), alone, induced a 4.8-fold increase in CLCr. This increase was reduced by the nicotinic receptor antagonist, mecamylamine (XXV).139 The lack of effect on CLCr of 170 mOsm luminal hypotonicity, alone (XXII), was also increased by iv pre-treatment of the COX-2 inhibitor, parecoxib (XXIII). These results confirm the influence of enteric nicotinic activity on the regulation of mucosal perme-ability induced by luminal hypotonicity. The effects of luminal hypotonicity, intestinal surgery, mecamylamine, and COX-2 inhibition, as proposed in this thesis, are schematically presented in Figure 12.

Figure 12. The effects on the intestinal mucosa of intestinal surgery, luminal hypo-tonicity, parecoxib (a COX-2 inhibitor), and mecamylamine (a nicotinic enteric re-ceptor antagonist). Red arrows, pharmacological effects. Green arrows, physiologi-cal effects/modulation.

48

In summary, the effects of mecamylamine and parecoxib illustrate that the experimental and luminal conditions can have a profound effect on determi-nations of intestinal permeability in the SPIP model, especially in cases when the physiological regulation of mucosal transport is investigated.

Effects in the rat and dog intraintestinal bolus models The intraintestinal bolus model takes into account several biopharmaceutical aspects not investigated in the SPIP model, such as intestinal spreading and dilution of the AMEs, potential inclusion of AME into colloidal structures, as well as intestinal transit of the drug and AME. These factors affect the free AME concentration at the mucosal membrane, and the local mucosal exposure time of both the drug and AME. The objective of Paper 4 was to investigate if the AME-induced increase in CLCr, and Jabs in the rat SPIP model in Paper 3, also occurred in the more in vivo relevant rat and dog intraintestinal bolus models.

The basal jejunal rat and dog Peff values of the six model compounds, from deconvolution of plasma data following administration of the control solution, are presented in Table 10. The basal Peff values of the six model compounds were 3- to 10-fold lower in rat than in dog, which is in agreement with the reports that dog intestinal mucosa is generally more permeable than for human and rat.58,160,173 This may also explain why the effects of caprate and SDS were higher in rat than in dog (1.5- and 3-fold, respectively); the dog intestinal mu-cosa was less affected by AMEs, as its intestinal epithelium is less of an ab-sorption barrier. This should be taken into consideration when the dog is used as a model for investigation of AME effects.

Table 10. Mean (±SD) dog (n=3–4) and rat (n=6) jejunal basal effective permeabil-ity (Peff) of six model compounds.

Model compounds Jejunal Peff (×10-4 cm/s)

Dog Rat

Acyclovir 0.93 ± 0.30 0.07 ± 0.05

Atenolol 0.75 ± 0.39 0.08 ± 0.03

Enalaprilat 0.14 ± 0.19 0.04 ± 0.03

Ketoprofen 3.4 ± 1.0 2.0 ± 0.5

Metoprolol 2.0 ± 1.8 0.30 ± 0.18

Phenol Red 0.09 ± 0.02 0.03 ± 0.01

The doses in dog (rat) were: acyclovir 5 (0.11) mg, atenolol 5 (0.26) mg, enalaprilat 20 (0.17)

mg, ketoprofen 2.5 (0.27) mg, metoprolol 12.5 (0.27) mg, and phenol red 5 (0.18) mg.

49

The effects of caprate, chitosan, and SDS on the absorption of the representa-tive low permeability compound enalaprilat, are presented in Figure 13. Fig-ure 13 also contains the absorption ratio (Jabs) of enalaprilat from the rat SPIP study in Paper 3 for easier comparison with the rat bolus study.

Figure 13. Mean (±SEM) enalaprilat dog and rat Peff ratios compared to a control so-lution, following intraintestinal administration of caprate, chitosan, and SDS at 2.3 and 11.4 mg/mL (●, darker shade, Paper 4). The rat figure also contains Jabs ratio data obtained from the SPIP model at 1 and 5 mg/mL (■, lighter shade, Paper 3).

In rat, there was a 13-fold significant increase in absorption ratio (Jabs) of en-alaprilat during administration of SDS at 11.4 mg/mL, but there was no effect of SDS at 2.3 mg/mL, or from the other AMEs (chitosan and caprate). This was in contrast to the data from the rat SPIP model in Paper 3, where there was a significant increase in Peff during perfusion of both SDS and chitosan, at comparable concentrations (1 and 5 mg/mL). Similar phenomena have been observed by others, where effects of AMEs are generally reduced in vivo, compared to in vitro.91,129,174 The lower effect in the bolus model compared to the SPIP model is most likely due to in vivo dilution and transit effects, and/or to contribution of luminal colloidal structures.

In summary, the dog has a more permeable intestinal mucosa than rat, and consequently seems to be less affected by AMEs. Mechanistic investigations of AMEs are useful in the SPIP model, but its ability to predict in vivo effects is uncertain. This was illustrated by the substantially lower effects of AMEs in both the rat and dog bolus models in Paper 4, compared to the rat SPIP study in Paper 3.

Time-dependent effects in the rat SPIP model The intraintestinal bolus study in Paper 4 showed that a mucosal exposure time of 75 min in the rat SPIP model in Paper 3 was not predictive of the effect of AMEs in vivo. Paper 6 therefore aimed to evaluate if a shorter mu-cosal exposure time of SDS and chitosan in the rat SPIP model would be more

50

predictive. The values of Jabs over time of enalaprilat (a representative low-permeability compound) and CLCr over time following 15- and 60-min muco-sal exposures of chitosan and SDS are presented in Figure 14.

Figure 14. Mean (SEM) enalaprilat Jabs and CLCr over 150 or 180 min in the rat SPIP model. The experiment started with a 60 min perfusion of the control solution (white). The perfusate was thereafter changed to a solution containing an AME (brown) for 15 or 60 min. The perfusate was then changed back to the control solu-tion (white) for 60 min. The investigated AMEs were SDS (red) and chitosan (blue), and the perfusate concentrations were 1 or 5 mg/mL.

The chitosan and SDS induced increase in total (i.e. area) and maximal (i.e. peak) enalaprilat Jabs, at a mucosal exposure of 15 min, were similar to the increase in Jabs ratio observed in the rat SPIP study in Paper 3 (75-min expo-sure). The total and maximal effect on Jabs in the 15-min exposure experiment could consequently not predict the in vivo effect of chitosan and SDS in the rat intraintestinal bolus study in Paper 4.

However, there was an association with the in vivo data in Paper 4 if time to effect (response rate) was considered, i.e., the increase in Jabs at the end of the 15-min exposure period. This is illustrated in Figure 16 for the low perme-ability compounds. A relative increase in Jabs above six-fold at the end of the 15-min exposure period was a requirement for a corresponding increase in Peff from the rat intraintestinal bolus model in Paper 4. A relative increase above six-fold in the 15-min exposure study was only induced by SDS at 5 mg/mL, and then only for acyclovir, enalaprilat, and phenol red; these compounds in combination with the high dose SDS (11.4 mg/mL) were also the only ones that were affected in the rat bolus study. This emphasizes that dynamic time-

51

dependent AME effects must be considered in preclinical evaluations of in vivo relevant AME effects.

Figure 15. Association between the chitosan- and SDS-induced increases in Peff ratio from the rat bolus study in Paper 4, and the corresponding increase in flux ratio at the end of the 15-min perfusion (Jabs(15)) in the rat SPIP model in Paper 6. The con-centrations of chitosan and SDS in the bolus study were 2.3 and 11.4 mg/mL, and the concentrations in this SPIP study were 1 and 5 mg/mL.

Finally, both a 15- and 60-min exposure of chitosan and SDS resulted in a faster and larger effect for Jabs of all low permeability model compounds, than for CLCr. Further, the mucosal recovery of Jabs was complete within the 60-min recovery period following the 15-min exposure, while CLCr was only 50% recovered (Figure 14). Following the 60-min exposure the recovery of Jabs was 50%, while there was no recovery of CLCr. In Papers 3 and 5, we observe a close correlation between AME-effects on Jabs and CLCr ratio, at various lu-minal conditions in the SPIP model (Figure 9). Those results indicate that CLCr ratio can be used as a marker when investigating the potential effect of an AME on drug absorption. However, the difference in how AMEs affected Jabs and CLCr when time dependence was accounted for indicate that: i) CLCr is not predictive of the rapid increase in Jabs induced by AMEs, and ii) Jabs is not necessarily predictive of mucosal recovery, as CLCr may be substantially in-creased for a long time after Jabs is reduced to baseline.

52

Conclusions

Papers 1 and 2 in this thesis investigated the regional intestinal differences in drug permeability in human and dog, and its implication for preclinical eval-uations of intestinal drug absorption.

Key conclusions: In human, the low permeability drug atenolol had a substantially lower

Peff in the colon than in the jejunum and ileum, whereas the Peff of the high permeability drugs, metoprolol and ketoprofen, were largely unaffected by intestinal region (Paper 1).

The deconvoluted human jejunal Peff values in Paper 1 were highly cor-related to the historically, directly determined values using the intestinal single-pass perfusion model. This suggests that the deconvolution-Peff method based on intraintestinal drug administrations was suitable for de-termination of regional intestinal permeability.

In dog, the jejunal Peff values of atenolol, metoprolol, and ketoprofen, from Paper 2, were highly correlated to the human jejunal Peff values from Paper 1. There was also a correlation between dog and human for metoprolol and ketoprofen in the colon, whereas the low permeability drug atenolol was substantially more permeable in the dog colon than in the human colon.

The difference in human and dog colonic permeability of low permeabil-ity compounds can have implications for the evaluation of intestinal drug absorption when dog is used as model species, especially for the develop-ment of modified-release dosage forms and BCS class III and IV drugs (Papers 1 and 2).

Papers 3–6 in this thesis investigated the effects and mechanisms of AMEs (caprate, chitosan, NAC, SDS) in various preclinical rat and dog intestinal drug absorption models. Key conclusions: There were no AME-induced effects on Jabs of the high permeability

drugs, metoprolol and ketoprofen, for any of the absorption models. This suggests that addition of AME in a novel drug formulation has limited effect on the absorption of BCS I and II drugs. (Papers 3–6).

53

In the rat SPIP model, the AME-induced increase in lumen-to-blood Jabs ratio of the low permeability model compounds, and blood-to-lumen CLCr ratio, at isotonic conditions, had an order of SDS > chitosan, and no effect of caprate and NAC (Paper 3).

The AME-induced increase in Jabs and CLCr ratio was highly correlated at both isotonic and hypotonic conditions, showing that CLCr can be used as a marker in the rat SPIP model for evaluation of effect-potential of an AME (Papers 3 and 5).

Luminal hypotonicity alone increased CLCr in the rat SPIP model. The effect was potentiated by parecoxib, a COX-2 inhibitor, and reduced by mecamylamine, a nicotinergic receptor antagonist (Paper 5). In vivo pre-dictions of drug absorption and relevance of the SPIP model can be in-creased by considering such physiological effects, for instance by pre-treatment of the rats with parecoxib, which reduces the negative effect of abdominal surgery on intestinal functions (Paper 5).

The effect of AMEs was substantially lower in the in vivo rat and dog intraintestinal bolus models, than in the rat SPIP model (Papers 3 and 4). This is most likely explained by the influence of physiological factors, such as intestinal transit, dilution, and colloidal structures, in the bolus models, compared to the SPIP model for which the luminal AME expo-sure time and intestinal free concentration were controlled. AME effects in transit-independent preclinical models should therefore be verified in more in vivo-like models.

Use of a shorter AME exposure-time in the rat SPIP model resulted in better in vivo predictions of AME in the rat bolus model. A more than six-fold increase in Jabs at the end of a 15-min exposure period seemed to be a requirement for an increase in Jabs in the bolus model, at comparable con-centrations of AMEs (Papers 4 and 6).

The correlation between Jabs and CLCr ratio from Papers 3 and 5 was not observed when dynamic time-dependent effects on the mucosa were in-vestigated in Paper 6. Jabs was more sensitive than CLCr in detecting re-sponse rate, while CLCr was more sensitive than Jabs at determining muco-sal recovery times. Both Jabs and CLCr are therefore useful parameters when investigating dynamic time-dependent effects on the intestinal mu-cosa in the rat SPIP model.

54

Populärvetenskaplig sammanfattning

Det vanligaste sättet att dosera ett läkemedel som ska verka i kroppen är via munnen. För att ett sådant läkemedel ska ha önskad effekt måste det absorbe-ras, det vill säga passera över tarmväggen till blodet, med en tillräckligt hög hastighet för att läkemedlet ska hinna tas upp innan det passerat hela mag-tarmkanalen. Ett läkemedels absorptionshastighet har ofta lokala skillnader i tarmen och är vanligtvis snabbare i tunntarmen jämfört med tjocktarmen. Hur mycket av ett läkemedel som absorberas kan därför bero på var i tarmen ta-bletten löses upp. Om ett läkemedel har låg absorptionshastighet i alla delar av tarmen spelar det däremot mindre roll var tabletten löses upp. Som tur är kan man öka absorptionen av ett sådant läkemedel genom att ge det tillsam-mans med ämnen som tillfälligt ökar tarmväggens genomsläpplighet, så kal-lade absorptionsförbättrare.

Inom läkemedelsindustrin utvärderar man läkemedel genom att testa på djur för att bland annat få information om effekten och säkerheten av ett nytt läkemedel innan det testas på människa. Kostnaden för att utveckla nya läke-medel har senaste åren skenat, delvis beroende på att många läkemedel som testas i människa inte visar sig vara tillräckligt effektiva och säkra. En av de bidragande orsakerna till detta är att man misslyckas med att förutsäga hur ett läkemedel kommer absorberas i människa baserat på djurstudier. Detta beror delvis på att det finns för lite kunskap om hur absorptionen av läkemedel ser ut i tjocktarmen på människa, eftersom denna del av tarmen är svårtillgänglig. Kunskapsbristen har gjort det svårt att veta hur bra olika djur är på att förut-säga läkemedelsabsorptionen i olika delar av tarmen på människa. På liknande sätt är kunskapen låg om hur väl effekten av en absorptionsförbättrare i olika djur avspeglar den faktiska effekten på människa. Om dessa kunskapsluckor täpps till så kan det i förlängningen leda till att nya effektiva läkemedel kan nå patienterna snabbare och till en lägre kostnad.

Denna avhandling hade två huvudsyften. Det första syftet var att för första gången studera absorptionshastigheten av läkemedel i både tunntarmen och tjocktarmen på människa och hund. Hund är ett vanligt djur att studera läke-medelsabsorption i eftersom dess storlek möjliggör dosering av hela tabletter. Det andra syftet var att öka den allmänna förståelsen av hur absorptionsför-bättrare påverkar absorptionen av läkemedel, samt utvärdera hur användbara råtta och hund är för att studera dessa effekter. Råtta är också ett vanligt djur att studera läkemedelsabsorption i då råttor är lätta att arbeta med samt att det finns mycket tidigare kunskap att luta sig mot.

55

Skillnader i absorptionshastighet mellan tunntarmen och tjocktarmen i människa och hund studerades genom att dosera läkemedel direkt in i dessa delar av tarmen via en slang. Absorptionshastigheten utvärderades sedan ge-nom att jämföra hur snabbt de olika läkemedlen kunde mätas i blodet.

Effekten av absorptionsförbättrare studerades genom att jämföra hur snabbt olika läkemedel kunde mätas i blodet, beroende på om de doserats tillsammans med en absorptionsförbättrare eller inte. Doseringen skedde i antingen direkt in i tunntarmen på både råtta och hund via en slang, eller i en liten del av tunntarmen på råtta som isolerats genom att båda tarmändarna snörpts av.

Hur snabbt olika läkemedel absorberades i tunntarmen överensstämde mycket väl mellan människa och hund. I människa var däremot absorptions-hastigheten mycket lägre i tjocktarmen än tunntarmen för läkemedel med låg absorptionshastighet, medan skillnaden mellan tjocktarmen och tunntarmen i hund för dessa läkemedel var mindre. Denna skillnad i tjocktarmsabsorption mellan arterna är viktig att känna till när hund används för att förutsäga hur mycket av ett läkemedel som kommer absorberas i människa, framförallt för tabletter som löses upp i just tjocktarmen.

Effekten av absorptionsförbättrare var generellt större i råtta än hund, vilket troligtvis avspeglar att råttans tarm var mindre genomsläpplig i sitt naturliga tillstånd, och därmed påverkas starkare. I råtta var även effekten av absorpt-ionsförbättrare större när de studerades i en isolerad del av tunntarmen, i jäm-förelse när de doserades in i tarmen via en slang. Detta avspeglar troligtvis den längre kontakten mellan absorptionsförbättraren och tarmväggen i den av-gränsade tarmen. I den andra modellen transporteras absorptionsförbättraren kontinuerligt framåt med tarmens naturliga rörelser. Ett bevis för detta är att effekten av absorptionsförbättraren i den avgränsade tarmen överensstämde bättre med resultaten från doseringen direkt in i tarmen via en slang, om kon-takttiden minskades i den avgränsade tarmen. I avhandlingen visades även att nerver och naturliga fysiologiska processer i tarmen kan ha en stor effekt på absorptionshastigheten av läkemedel. Det är viktigt att tänka på det när man studerar läkemedelabsorption för att säkerställa att studier i djur ger tillförlit-liga förutsägelser för hur det ser ut i människa.

Sammantaget har denna avhandling, som bygger på sex referensgranskade vetenskapliga artiklar, ökat den grundläggande förståelsen för hur fysiologin i tarmen påverkar upptaget av läkemedel, samt hur detta samverkar med ta-blettens egenskaper, såsom var den löses upp och ifall den innehåller absorpt-ionsförbättrare. Detta tillsammans med de nya kunskaperna om fördelar och nackdelar med olika djur för att studera läkemedelsabsorption kan leda till en snabbare och mer kostnadseffektiv utveckling av nya läkemedel.

56

Acknowledgements

The work in this thesis was carried out at the Department of Pharmacy, Faculty of Pharmacy, Uppsala University, Sweden. Thanks to Apotekare CD Carlssons stiftelse (Apotekarsocieteten) for money allowing me to travel to Edinburgh to attend a conference. Foremost I want to thank my supervisor Hans Lennernäs and co-supervisor Erik Sjögren. Hans, your energy and positivity is inspiring and it often gave the energy needed to push on with the projects in rough periods. The high degree of freedom you gave me in the projects has allowed me to take initia-tive, which has made me a better and more independent scientist. I hope that I have also been able to inspire you, into becoming a better football player, and that one-day you will lift the Korpen div. 1 trophy. Erik, the everyday talks in the corridor were amazing, a perfect blend of seriousness and jokes. Your brilliant brain has made you an excellent scientist and Warhammer player, two invaluable traits for any supervisor. At AstraZeneca R&D in Göteborg I want to thank Christer Tannergren, Bertil Abrahamsson, Pernilla Johansson, Anders Lundqvist, Charlotta Vedin, and many more. This thesis would not have been possible without your valuable help and contributions. A special thanks to Markus Sjögren, Olof Nylander, and Tobias Olander, at the Department of Neuroscience. Working with you has been a joy. I appreci-ate all practical help from you in the projects, but foremost I appreciate all the interesting and fun discussions, both at work and at bars. Carl Roos. Working with you has been a heavenly experience. You have kept me running these six years and I cannot see myself achieving even half of what I have done without you. You are my Samwise Gamgee (the true hero of Middle-earth). To the other PhD students in the Biopharmaceutics group: Ilse, Elsa, Emelie, Johanna, Fredrik, and Oliver. One could not have hoped for more harmonious and wonderful colleagues.

57

A big thanks to Johan Gråsjö for all the help with (for me) complicated statis-tics and math. Thanks to all wonderful people at the department of pharmacy not specifically named. Team Dotter, tack för alla verkligt stora gäddor ni fångar. Thomas, du behöver köpa mer drag, och Jakob, jag ber till Gud att sommaren 2018 då du inte fång-ade någon fisk aldrig kommer upprepas. Tack till påläggskalvarna i Bamm: Jeremy, Tomas, Johan, Hanna, Sepand, Oskar, Kalle. Era framgångar, alla fysiska och psykiska tillkortakommanden till trots, är en stor inspirationskälla. Tack till Mamma, Pappa och Tova. Bättre familj kan man inte önska sig. Att ni alltid finns där betyder allt!

58

References

1. Dahlgren D, Roos C, Johansson P, Lundqvist A, Tannergren C, Abrahamsson B, Sjögren E, Lennernäs H. Regional intestinal permeability in dogs: biopharmaceutical aspects for development of oral modified-release dosage forms. Mol Pharm. 2016, 13:3022-3033.

2. Marchiando AM, Shen L, Graham WV, Edelblum KL, Duckworth CA, Guan Y, Montrose MH, Turner JR, Watson AJ. The epithelial barrier is maintained by in vivo tight junction expansion during pathologic intestinal epithelial shedding. Gastroenterology. 2011, 140:1208-1218. e1202.

3. Marchiando AM, Graham WV, Turner JR. Epithelial barriers in homeostasis and disease. Annual Review of Pathological Mechanical Disease. 2010, 5:119-144.

4. Hughes JP, Rees S, Kalindjian SB, Philpott KL. Principles of early drug discovery. Br J Pharmacol. 2011, 162:1239-1249.

5. Morgan S, Grootendorst P, Lexchin J, Cunningham C, Greyson D. The cost of drug development: a systematic review. Health Policy. 2011, 100:4-17.

6. DiMasi JA, Grabowski HG, Hansen RW. The cost of drug development. N Engl J Med. 2015, 372:1972-1972.

7. Lennernäs H, Aarons L, Augustijns P, Beato S, Bolger M, Box K, Brewster M, Butler J, Dressman J, Holm R, Julia Frank K, Kendall R, Langguth P, Sydor J, Lindahl A, McAllister M, Muenster U, Müllertz A, Ojala K, Pepin X, Reppas C, Rostami-Hodjegan A, Verwei M, Weitschies W, Wilson C, Karlsson C, Abrahamsson B. Oral biopharmaceutics tools–time for a new initiative-an introduction to the IMI project OrBiTo. Eur J Pharm Sci. 2013, 57:292-299.

8. Sjögren E, Abrahamsson B, Augustijns P, Becker D, Bolger MB, Brewster M, Brouwers J, Flanagan T, Harwood M, Heinen C, Holm R, Juretschke H, Kubbinga M, Lindahl A, Lukacova V, Münster U, Neuhoff S, Nguyen M, Peer A, Reppas C, Hodjegan A, Tannergren C, Weitschies W, Wilson C, Zane P, Lennernäs H, Langguth P. In vivo methods for drug absorption–comparative physiologies, model selection, correlations with in vitro methods (IVIVC), and applications for formulation/API/excipient characterization including food effects. Eur J Pharm Sci. 2014, 57:99-151.

9. Kostewicz ES, Abrahamsson B, Brewster M, Brouwers J, Butler J, Carlert S, Dickinson PA, Dressman J, Holm R, Klein S. < i> In vitro</i> models for the prediction of< i> in vivo</i> performance of oral dosage forms. Eur J Pharm Sci. 2013.

10. Waring MJ, Arrowsmith J, Leach AR, Leeson PD, Mandrell S, Owen RM, Pairaudeau G, Pennie WD, Pickett SD, Wang J. An analysis of the attrition of drug candidates from four major pharmaceutical companies. Nat Rev Drug Discov. 2015, 14:475.

11. Rowland M, Noe CR, Smith DA, Tucker GT, Crommelin DJ, Peck CC, Rocci Jr ML, Besançon L, Shah VP. Impact of the pharmaceutical sciences on health care: a reflection over the past 50 years. J Pharm Sci. 2012, 101:4075-4099.

59

12. Amidon KS, Langguth P, Lennernas H, Yu L, Amidon GL. Bioequivalence of oral products and the biopharmaceutics classification system: science, regulation, and public policy. Clin Pharmacol Ther. 2011, 90:467-470.

13. FDA.BCS Guidance for Industry www.fda.gov/downloads/Drugs/Guidances/ucm070246.pdf 12 August 2014 14. Amidon GL, Lennernas H, Shah VP, Crison JR. A theoretical basis for a

biopharmaceutic drug classification: the correlation of in vitro drug product dissolution and in vivo bioavailability. Pharm Res. 1995, 12:413-420.

15. Abrahamsson B, Lücker P, Olofsson B, Regårdh CG, Sandberg A, Wieselgren I, Bergstrand R. The Relationship Between Metoprolol Plasma Concentration and Beta1‐Blockade in Healthy Subjects: A Study on Conventional Metoprolol and Metoprolol CR/ZOK Formulations. J Clin Pharmacol. 1990, 30:S46-S54.

16. Lennernäs H, Skrtic S, Johannsson G. Replacement therapy of oral hydrocortisone in adrenal insufficiency: the influence of gastrointestinal factors. 2008.

17. Leung PS. 2014. The Gastrointestinal System: Gastrointestinal, Nutritional and Hepatobiliary Physiology. ed.: Springer.

18. Madara JL. Functional morphology of epithelium of the small intestine. Comprehensive Physiology. 2011.

19. Van Itallie CM, Anderson JM. The molecular physiology of tight junction pores. Physiology. 2004, 19:331-338.

20. Johansson ME, Sjövall H, Hansson GC. The gastrointestinal mucus system in health and disease. Nature Reviews Gastroenterology and Hepatology. 2013, 10:352-361.

21. Johansson ME, Ambort D, Pelaseyed T, Schütte A, Gustafsson JK, Ermund A, Subramani DB, Holmén-Larsson JM, Thomsson KA, Bergström JH. Composition and functional role of the mucus layers in the intestine. Cell Mol Life Sci. 2011, 68:3635-3641.

22. Wang Y, Brasseur JG. Three-dimensional mechanisms of macro-to-micro-scale transport and absorption enhancement by gut villi motions. Physical Review E. 2017, 95:062412.

23. Furness JB. The enteric nervous system and neurogastroenterology. Nature Reviews Gastroenterology and Hepatology. 2012, 9:286.

24. Turner JR. Intestinal mucosal barrier function in health and disease. Nature Reviews Immunology. 2009, 9:799.

25. Sellers RS, Morton D. The Colon From Banal to Brilliant. Toxicol Pathol. 2013:67-81.

26. Helander HF, Fändriks L. Surface area of the digestive tract-revisited. Scand J Gastroenterol. 2014:1-9.

27. Wilson J. Surface area of the small intestine in man. Gut. 1967, 8:618. 28. Mudie DM, Murray K, Hoad CL, Pritchard SE, Garnett MC, Amidon GL,

Gowland PA, Spiller RC, Amidon GE, Marciani L. Quantification of gastrointestinal liquid volumes and distribution following a 240 mL dose of water in the fasted state. Mol Pharm. 2014, 11:3039-3047.

29. Schiller C, Fröhlich CP, Giessmann T, Siegmund W, Mönnikes H, Hosten N, Weitschies W. Intestinal fluid volumes and transit of dosage forms as assessed by magnetic resonance imaging. Aliment Pharmacol Ther. 2005, 22:971-979.

30. Wang YT, Mohammed SD, Farmer AD, Wang D, Zarate N, Hobson AR, Hellström PM, Semler JR, Kuo B, Rao SS. Regional gastrointestinal transit and pH studied in 215 healthy volunteers using the wireless motility capsule: influence of age, gender, study country and testing protocol. Aliment Pharmacol Ther. 2015, 42:761-772.

60

31. Fallingborg J, Christensen L, Ingeman‐Nielsen M, Jacobsen B, Abildgaard K, Rasmussen H. pH‐Profile and regional transit times of the normal gut measured by a radiotelemetry device. Aliment Pharmacol Ther. 1989, 3:605-614.

32. Davis S, Hardy J, Fara J. Transit of pharmaceutical dosage forms through the small intestine. Gut. 1986, 27:886-892.

33. Abrahamsson B, Alpsten M, Jonsson UE, Lundberg P, Sandberg A, Sundgren M, Svenheden A, Tölli J. Gastro-intestinal transit of a multiple-unit formulation (metoprolol CR/ZOK) and a non-disintegrating tablet with the emphasis on colon. Int J Pharm. 1996, 140:229-235.

34. Ensign LM, Cone R, Hanes J. Oral drug delivery with polymeric nanoparticles: the gastrointestinal mucus barriers. Advanced drug delivery reviews. 2012, 64:557-570.

35. Harwood M, Neuhoff S, Carlson G, Warhurst G, Rostami‐Hodjegan A. Absolute abundance and function of intestinal drug transporters: a prerequisite for fully mechanistic in vitro–in vivo extrapolation of oral drug absorption. Biopharm Drug Dispos. 2013, 34:2-28.

36. Estudante M, Morais JG, Soveral G, Benet LZ. Intestinal drug transporters: an overview. Advanced drug delivery reviews. 2012.

37. Giacomini KM, Huang S-M, Tweedie DJ, Benet LZ, Brouwer KL, Chu X, Dahlin A, Evers R, Fischer V, Hillgren KM. Membrane transporters in drug development. Nat Rev Drug Discov. 2010, 9:215-236.

38. Fan J, Inés L. Pharmacokinetics. Biochem Pharmacol. 2014, 87:93-120. 39. Gibson GG, Skett P. 2013. Introduction to drug metabolism. ed.: Springer. 40. EMA.Bioeqvivalence and Bioavailability Guideline http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guidelin

e/2009/09/WC500003519.pdf 41. FDA.Bioeqvivalence and Bioavailability Guideline https://www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinforma

tion/guidances/ucm389370.pdf 42. Wu C-Y, Benet LZ. Predicting drug disposition via application of BCS:

transport/absorption/elimination interplay and development of a biopharmaceutics drug disposition classification system. Pharm Res. 2005, 22:11-23.

43. Adkin DA, Davis S, Sparrow R, Huckle P, Phillips A, Wilding I. The effects of pharmaceutical excipients on small intestinal transit. Br J Clin Pharmacol. 1995, 39:381-387.

44. Basit AW, Newton JM, Short MD, Waddington WA, Ell PJ, Lacey LF. The effect of polyethylene glycol 400 on gastrointestinal transit: implications for the formulation of poorly-water soluble drugs. Pharm Res. 2001, 18:1146-1150.

45. Thelen K, Dressman JB. Cytochrome P450‐mediated metabolism in the human gut wall. J Pharm Pharmacol. 2009, 61:541-558.

46. Fukao M, Ishida K, Horie A, Taguchi M, Nozawa T, Inoue H, Hashimoto Y. Variability of Bioavailability and Intestinal Absorption Mechanisms of Metoprolol. Drug Metab Pharmacokinet. 2014, 29:162-167.

47. Rowland M. Models to identify sources of pharmacokinetic variability. Va—riability in drug therapy: Description, estimation, and control Raven Press, New York. 1985:11-28.

48. Silas J, McGourty J, Lennard M, Tucker G, Woods H. Polymorphic metabolism of metoprolol: clinical studies. Eur J Clin Pharmacol. 1985, 28:85-88.

61

49. Kirchheiner J, Heesch C, Bauer S, Meisel C, Seringer A, Goldammer M, Tzvetkov M, Meineke I, Roots I, Brockmöller J. Impact of the ultrarapid metabolizer genotype of cytochrome P450 2D6 on metoprolol pharmacokinetics and pharmacodynamics. Clin Pharmacol Ther. 2004, 76:302-312.

50. Smith D, Artursson P, Avdeef A, Di L, Ecker GF, Faller B, Houston JB, Kansy M, Kerns EH, Krämer SD. Passive lipoidal diffusion and carrier-mediated cell uptake are both important mechanisms of membrane permeation in drug disposition. Mol Pharm. 2014, 11:1727-1738.

51. Lipinski CA, Lombardo F, Dominy BW, Feeney PJ. Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Advanced drug delivery reviews. 1997, 23:3-25.

52. Winiwarter S, Ax F, Lennernäs H, Hallberg A, Pettersson C, Karlén A. Hydrogen bonding descriptors in the prediction of human in vivo intestinal permeability. J Mol Graph Model. 2003, 21:273-287.

53. Winiwarter S, Bonham NM, Ax F, Hallberg A, Lennernäs H, Karlén A. Correlation of human jejunal permeability (in vivo) of drugs with experimentally and theoretically derived parameters. A multivariate data analysis approach. J Med Chem. 1998, 41:4939-4949.

54. Nylander O, Sjöblom M. Modulation of mucosal permeability by vasoactive intestinal peptide or lidocaine affects the adjustment of luminal hypotonicity in rat duodenum. Acta Physiologica. 2007, 189:325-335.

55. Taverner A, Dondi R, Almansour K, Laurent F, Owens S-E, Eggleston IM, Fotaki N, Mrsny RJ. Enhanced paracellular transport of insulin can be achieved via transient induction of myosin light chain phosphorylation. J Control Release. 2015, 210:189-197.

56. Lennernas H. Intestinal permeability and its relevance for absorption and elimination. Xenobiotica. 2007, 37:1015-1051.

57. Lennernas H, Ahrenstedt Ö, Ungell A. Intestinal drug absorption during induced net water absorption in man; a mechanistic study using antipyrine, atenolol and enalaprilat. Br J Clin Pharmacol. 1994, 37:589-596.

58. He YL, Murby S, Warhurst G, Gifford L, Walker D, Ayrton J, Eastmond R, Rowland M. Species differences in size discrimination in the paracellular pathway reflected by oral bioavailability of poly (ethylene glycol) and D‐peptides. J Pharm Sci. 1998, 87:626-633.

59. Nicholls G, Youdim K. 2016. Drug Transporters: Volume 1: Role and Importance in ADME and Drug Development. ed.: Royal Society of Chemistry.

60. Bras A, Sitar D, Aoki F. Comparative bioavailability of acyclovir from oral valacyclovir and acyclovir in patients treated for recurrent genital herpes simplex virus infection. The Canadian journal of clinical pharmacology= Journal canadien de pharmacologie clinique. 2001, 8:207-211.

61. Beutner KR. Valacyclovir: a review of its antiviral activity, pharmacokinetic properties, and clinical efficacy. Antiviral Res. 1995, 28:281-290.

62. Thomsen AE, Christensen MS, Bagger MA, Steffansen B. Acyclovir prodrug for the intestinal di/tri-peptide transporter PEPT1: comparison of in vivo bioavailability in rats and transport in Caco-2 cells. Eur J Pharm Sci. 2004, 23:319-325.

62

63. Kell DB, Dobson PD, Bilsland E, Oliver SG. The promiscuous binding of pharmaceutical drugs and their transporter-mediated uptake into cells: what we (need to) know and how we can do so. Drug discovery today. 2013, 18:218-239.

64. Kell DB, Dobson PD, Oliver SG. Pharmaceutical drug transport: the issues and the implications that it is essentially carrier-mediated only. Drug discovery today. 2011, 16:704-714.

65. Sugano K, Kansy M, Artursson P, Avdeef A, Bendels S, Di L, Ecker GF, Faller B, Fischer H, Gerebtzoff G, Lennernäs H, Senner F. Coexistence of passive and carrier-mediated processes in drug transport. Nat Rev Drug Discov. 2010, 9:597-614.

66. Smithson KW, Millar DB, Jacobs LR, Gray GM. Intestinal diffusion barrier: unstirred water layer or membrane surface mucous coat? Science. 1981, 214:1241-1244.

67. Winne D. Rat jejunum perfused in situ: effect of perfusion rate and intraluminal radius on absorption rate and effective unstirred layer thickness. Naunyn-Schmiedeberg's archives of pharmacology. 1979, 307:265-274.

68. Sugano K. Aqueous boundary layers related to oral absorption of a drug: from dissolution of a drug to carrier mediated transport and intestinal wall metabolism. Mol Pharm. 2010, 7:1362-1373.

69. Levitt MD, Strocchi A, Levitt DG. Human jejunal unstirred layer: evidence for extremely efficient luminal stirring. American Journal of Physiology-Gastrointestinal and Liver Physiology. 1992, 262:G593-G596.

70. Fagerholm U, Lennernäs H. Experimental estimation of the effective unstirred water layer thickness in the human jejunum, and its importance in oral drug absorption. Eur J Pharm Sci. 1995, 3:247-253.

71. Levitt MD, Furne J, Strocchi A, Anderson BW, Levitt DG. Physiological measurements of luminal stirring in the dog and human small bowel. J Clin Invest. 1990, 86:1540.

72. Roos C, Dahlgren D, Berg S, Westergren J, Abrahamsson B, Tannergren C, Sjögren E, Lennernäs H. In Vivo Mechanisms of Intestinal Drug Absorption from Aprepitant Nanoformulations. Mol Pharm. 2017, 14:4233-4242.

73. Dahlgren D, Roos C, Sjögren E, Lennernäs H. Direct In Vivo Human Intestinal Permeability (Peff) Determined with Different Clinical Perfusion and Intubation Methods. J Pharm Sci. 2014, 104:2702-2726.

74. Andersson K-E, Nyberg L, Dencker H, Göthlin J. Absorption of digoxin in man after oral and intrasigmoid administration studied by portal vein catheterization. Eur J Clin Pharmacol. 1975, 9:39-47.

75. Steffansen B, Nielsen CU, Brodin B, Eriksson AH, Andersen R, Frokjaer S. Intestinal solute carriers: an overview of trends and strategies for improving oral drug absorption. Eur J Pharm Sci. 2004, 21:3-16.

76. Lipinski C. Poor aqueous solubility—an industry wide problem in drug discovery. Am Pharm Rev. 2002, 5:82-85.

77. Kawabata Y, Wada K, Nakatani M, Yamada S, Onoue S. Formulation design for poorly water-soluble drugs based on biopharmaceutics classification system: basic approaches and practical applications. Int J Pharm. 2011, 420:1-10.

78. Dokoumetzidis A, Macheras P. A century of dissolution research: from Noyes and Whitney to the biopharmaceutics classification system. Int J Pharm. 2006, 321:1-11.

63

79. Tannergren C, Karlsson E, Sigfridsson K, Lindfors L, Ku A, Polentarutti B, Carlert S. Biopharmaceutic Profiling of Salts to Improve Absorption of Poorly Soluble Basic Drugs. J Pharm Sci. 2016, 105:3314-3323.

80. Kanaujia P, Poovizhi P, Ng W, Tan R. Amorphous formulations for dissolution and bioavailability enhancement of poorly soluble APIs. Powder Technology. 2015, 285:2-15.

81. Taniguchi C, Kawabata Y, Wada K, Yamada S, Onoue S. Microenvironmental pH-modification to improve dissolution behavior and oral absorption for drugs with pH-dependent solubility. Expert opinion on drug delivery. 2014, 11:505-516.

82. Fine-Shamir N, Beig A, Zur M, Lindley D, Miller JM, Dahan A. Toward Successful Cyclodextrin Based Solubility-Enabling Formulations for Oral Delivery of Lipophilic Drugs: Solubility–Permeability Trade-Off, Biorelevant Dissolution, and the Unstirred Water Layer. Mol Pharm. 2017, 14:2138-2146.

83. Chen H, Khemtong C, Yang X, Chang X, Gao J. Nanonization strategies for poorly water-soluble drugs. Drug discovery today. 2011, 16:354-360.

84. Liversidge GG, Cundy KC. Particle size reduction for improvement of oral bioavailability of hydrophobic drugs: I. Absolute oral bioavailability of nanocrystalline danazol in beagle dogs. Int J Pharm. 1995, 125:91-97.

85. Williams AC, Barry BW. Penetration enhancers. Advanced drug delivery reviews. 2012, 64:128-137.

86. Whitehead K, Karr N, Mitragotri S. Safe and effective permeation enhancers for oral drug delivery. Pharm Res. 2008, 25:1782-1788.

87. Anderberg EK, Lindmark T, Artursson P. Sodium caprate elicits dilatations in human intestinal tight junctions and enhances drug absorption by the paracellular route. Pharm Res. 1993, 10:857-864.

88. McCartney F, Gleeson JP, Brayden DJ. Safety concerns over the use of intestinal permeation enhancers: A mini-review. Tissue barriers. 2016, 4:e1176822.

89. Aguirre TA, Teijeiro-Osorio D, Rosa M, Coulter I, Alonso M, Brayden D. Current status of selected oral peptide technologies in advanced preclinical development and in clinical trials. Advanced drug delivery reviews. 2016, 106:223-241.

90. Muheem A, Shakeel F, Jahangir MA, Anwar M, Mallick N, Jain GK, Warsi MH, Ahmad FJ. A review on the strategies for oral delivery of proteins and peptides and their clinical perspectives. Saudi Pharmaceutical Journal. 2016, 24:413-428.

91. Maher S, Mrsny RJ, Brayden DJ. Intestinal permeation enhancers for oral peptide delivery. Advanced drug delivery reviews. 2016, 106:277-319.

92. Danielsen EM, Hansen GH. Intestinal surfactant permeation enhancers and their interaction with enterocyte cell membranes in a mucosal explant system. Tissue barriers. 2017, 5:e1361900.

93. Klein J, Arora A, Liu P. 2017. Gastric acid modulators for oral delivery of peptides and proteins. ed.: Google Patents.

94. Lennernäs H, Regårdh C-G. Evidence for an interaction between the β-blocker pafenolol and bile salts in the intestinal lumen of the rat leading to dose-dependent oral absorption and double peaks in the plasma concentration–time profile. Pharm Res. 1993, 10:879-883.

95. Gozalbes R, Jacewicz M, Annand R, Tsaioun K, Pineda-Lucena A. QSAR-based permeability model for drug-like compounds. Bioorg Med Chem. 2011, 19:2615-2624.

64

96. Akamatsu M, Fujikawa M, Nakao K, Shimizu R. In silico prediction of human oral absorption based on QSAR analyses of PAMPA permeability. Chem Biodivers. 2009, 6:1845-1866.

97. Lipinski CA, Lombardo F, Dominy BW, Feeney PJ. Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings1. Advanced drug delivery reviews. 2001, 46:3-26.

98. Awoonor-Williams E, Rowley CN. Molecular simulation of nonfacilitated membrane permeation. Biochimica et Biophysica Acta (BBA)-Biomembranes. 2016, 1858:1672-1687.

99. Lee CT, Comer J, Herndon C, Leung N, Pavlova A, Swift RV, Tung C, Rowley CN, Amaro RE, Chipot C. Simulation-based approaches for determining membrane permeability of small compounds. Journal of chemical information and modeling. 2016, 56:721-733.

100. Kostewicz ES, Aarons L, Bergstrand M, Bolger MB, Galetin A, Hatley O, Jamei M, Lloyd R, Pepin X, Rostami-Hodjegan A, Sjögren E, Tannergren C, Turner D, Wagner C, Weitschies W, Dressman J. PBPK models for the prediction of< i> in vivo</i> performance of oral dosage forms. Eur J Pharm Sci. 2013, 57:300-321.

101. Sugano K. Introduction to computational oral absorption simulation. Expert Opin Drug Metab Toxicol. 2009, 5:259-293.

102. Sjögren E, Thörn H, Tannergren C. In silico modeling of gastrointestinal drug absorption: predictive performance of three physiologically based absorption models. Mol Pharm. 2016, 13:1763-1778.

103. Jones H, Chen Y, Gibson C, Heimbach T, Parrott N, Peters S, Snoeys J, Upreti V, Zheng M, Hall S. Physiologically based pharmacokinetic modeling in drug discovery and development: a pharmaceutical industry perspective. Clin Pharmacol Ther. 2015, 97:247-262.

104. Sjöberg Å, Lutz M, Tannergren C, Wingolf C, Borde A, Ungell A-L. Comprehensive study on regional human intestinal permeability and prediction of fraction absorbed of drugs using the Ussing chamber technique. Eur J Pharm Sci. 2012.

105. Artursson P. Epithelial transport of drugs in cell culture. I: A model for studying the passive diffusion of drugs over intestinal absorbtive (Caco‐2) cells. J Pharm Sci. 1990, 79:476-482.

106. Nielsen CU, Andersen R, Brodin B, Frokjaer S, Taub ME, Steffansen B. Dipeptide model prodrugs for the intestinal oligopeptide transporter. Affinity for and transport via hPepT1 in the human intestinal Caco-2 cell line. J Control Release. 2001, 76:129-138.

107. Neuhoff S, Ungell A-L, Zamora I, Artursson P. pH-dependent bidirectional transport of weakly basic drugs across Caco-2 monolayers: implications for drug–drug interactions. Pharm Res. 2003, 20:1141-1148.

108. Larregieu CA, Benet LZ. Drug Discovery and Regulatory Considerations for Improving In Silico and In Vitro Predictions that Use Caco-2 as a Surrogate for Human Intestinal Permeability Measurements. The AAPS journal. 2013:1-15.

109. Yamashita S, Furubayashi T, Kataoka M, Sakane T, Sezaki H, Tokuda H. Optimized conditions for prediction of intestinal drug permeability using Caco-2 cells. Eur J Pharm Sci. 2000, 10:195-204.

110. Steffansen B, Grandvuinet AS. 2013. In Vitro Kinetic Characterization of Transporter-Mediated Permeability. Transporters in Drug Development, ed.: Springer. p 23-35.

65

111. Lozoya-Agullo I, Gonzalez-Alvarez I, Zur M, Fine-Shamir N, Cohen Y, Markovic M, Garrigues TM, Dahan A, Gonzalez-Alvarez M, Merino-Sanjuán M. Closed-Loop Doluisio (Colon, Small Intestine) and Single-Pass Intestinal Perfusion (Colon, Jejunum) in Rat—Biophysical Model and Predictions Based on Caco-2. Pharm Res. 2018, 35:2.

112. Amidon GL, Kou J, Elliott RL, Lightfoot EN. Analysis of models for determining intestinal wall permeabilities. J Pharm Sci. 1980, 69:1369-1373.

113. Lennernäs H, Lee ID, Fagerholm U, Amidon GL. A Residence‐Time Distribution Analysis of the Hydrodynamics within the Intestine in Man during a Regional Single‐pass Perfusion with Loc‐I‐Gut: In‐vivo Permeability Estimation. J Pharm Pharmacol. 1997, 49:682-686.

114. Volpe DA. Application of method suitability for drug permeability classification. The AAPS journal. 2010, 12:670-678.

115. Sjögren E, Dahlgren D, Roos C, Lennernas H. Human in vivo regional intestinal permeability: quantitation using site-specific drug absorption data. Mol Pharm. 2015, 12:2026-2039.

116. Burcham DL, Aungst BA, Hussain M, Gorko MA, Quon CY, Huang S-M. The effect of absorption enhancers on the oral absorption of the GP IIB/IIIA receptor antagonist, DMP 728, in rats and dogs. Pharm Res. 1995, 12:2065-2070.

117. Bickerton GR, Paolini GV, Besnard J, Muresan S, Hopkins AL. Quantifying the chemical beauty of drugs. Nature chemistry. 2012, 4:90.

118. De Miranda P, Good S. Species differences in the metabolism and disposition of antiviral nucleoside analogues: 1. Acyclovir. Antiviral Chemistry and Chemotherapy. 1992, 3:1-8.

119. Reeves PR, Barnfield DJ, Longshaw S, McIntosh DA, Winrow MJ. Disposition and metabolism of atenolol in animals. Xenobiotica. 1978, 8:305-311.

120. Reeves PR, McAinsh J, McIntosh DA, Winrow MJ. Metabolism of atenolol in man. Xenobiotica. 1978, 8:313-320.

121. Ulm EH. Enalapril maleate (MK-421), a potent, nonsulfhydryl angiotensin-converting enzyme inhibitor: absorption, disposition, and metabolism in man. Drug Metab Rev. 1983, 14:99-110.

122. Foster R, Jamali F. Stereoselective pharmacokinetics of ketoprofen in the rat. Influence of route of administration. Drug Metabolism and Disposition. 1988, 16:623-626.

123. Shohin IE, Kulinich JI, Ramenskaya GV, Abrahamsson B, Kopp S, Langguth P, Polli JE, Shah VP, Groot D, Barends DM, Dressman J. Biowaiver monographs for immediate‐release solid oral dosage forms: Ketoprofen. J Pharm Sci. 2012, 101:3593-3603.

124. Regårdh CG, Borg KO, Johansson R, Johnsson G, Palmer L. Pharmacokinetic studies on the selectiveβ 1-receptor antagonist metoprolol in man. J Pharmacokinet Biopharm. 1974, 2:347-364.

125. Yoon I-S, Choi M-K, Kim JS, Shim C-K, Chung S-J, Kim D-D. Pharmacokinetics and first-pass elimination of metoprolol in rats: contribution of intestinal first-pass extraction to low bioavailability of metoprolol. Xenobiotica. 2011, 41:243-251.

126. Fleck C, Bräunlich H. Relation between renal and hepatic excretion of drugs: II. Age-dependence of phenol red excretion in comparison with those of p-aminohippurate and indocyanine green. Exp Pathol. 1986, 29:235-247.

66

127. Yin J, Duan H, Shirasaka Y, Prasad B, Wang J. Atenolol Renal Secretion Is Mediated by Human Organic Cation Transporter 2 and Multidrug and Toxin Extrusion Proteins. Drug Metabolism and Disposition. 2015, 43:1872-1881.

128. Driscoll JL, Hayner NT, Williams-Holland R, Spies-Karotkin G, Galletti PM, Jauregui HO. Phenolsulfonphthalein (phenol red) metabolism in primary monolayer cultures of adult rat hepatocytes. In Vitro-Plant. 1982, 18:835-842.

129. Maher S, Leonard TW, Jacobsen J, Brayden DJ. Safety and efficacy of sodium caprate in promoting oral drug absorption: from in vitro to the clinic. Advanced drug delivery reviews. 2009, 61:1427-1449.

130. Schipper NG, Olsson S, Hoogstraate JA, Vårum KM, Artursson P. Chitosans as absorption enhancers for poorly absorbable drugs 2: mechanism of absorption enhancement. Pharm Res. 1997, 14:923-929.

131. Takatsuka S, Kitazawa T, Morita T, Horikiri Y, Yoshino H. Enhancement of intestinal absorption of poorly absorbed hydrophilic compounds by simultaneous use of mucolytic agent and non-ionic surfactant. Eur J Pharm Biopharm. 2006, 62:52-58.

132. Davies B, Morris T. Physiological parameters in laboratory animals and humans. Pharm Res. 1993, 10:1093-1095.

133. Talattof A, Price JC, Amidon GL. Gastrointestinal Motility Variation and Implications for Plasma Level Variation: Oral Drug Products. Mol Pharm. 2015, 13:557–567.

134. Sarosiek I, Selover K, Katz L, Semler J, Wilding G, Lackner J, Sitrin M, Kuo B, Chey W, Hasler W. The assessment of regional gut transit times in healthy controls and patients with gastroparesis using wireless motility technology. Aliment Pharmacol Ther. 2010, 31:313-322.

135. Di Colo G, Zambito Y, Zaino C. Polymeric enhancers of mucosal epithelia permeability: synthesis, transepithelial penetration-enhancing properties, mechanism of action, safety issues. J Pharm Sci. 2008, 97:1652-1680.

136. Magnuson BL, Clifford TM, Hoskins LA, Bernard AC. Enteral nutrition and drug administration, interactions, and complications. Nutr Clin Pract. 2005, 20:618-624.

137. Nyholm D, Askmark H, Gomes–Trolin C, Knutson T, Lennernäs H, Nyström C, Aquilonius S-M. Optimizing levodopa pharmacokinetics: intestinal infusion versus oral sustained-release tablets. Clin Neuropharmacol. 2003, 26:156-163.

138. Sommansson A, Nylander O, Sjöblom M. Melatonin decreases duodenal epithelial paracellular permeability via a nicotinic receptor–dependent pathway in rats in vivo. J Pineal Res. 2013, 54:282-291.

139. Sedin J, Sjöblom M, Nylander O. The selective cyclooxygenase‐2 inhibitor parecoxib markedly improves the ability of the duodenum to regulate luminal hypertonicity in anaesthetized rats. Acta Physiologica. 2012, 205:433-451.

140. Nylander O, Kvietys P, Granger DN. Effects of hydrochloric acid on duodenal and jejunal mucosal permeability in the rat. American Journal of Physiology-Gastrointestinal and Liver Physiology. 1989, 257:G653-G660.

141. Sandberg A, Blomqvist I, Jonsson U, Lundborg P. Pharmacokinetic and pharmacodynamic properties of a new controlled-release formulation of metoprolol: a comparison with conventional tablets. Eur J Clin Pharmacol. 1988, 33:S9-S14.

142. Mason W, Winer N, Kochak G, Cohen I, Bell R. Kinetics and absolute bioavailability of atenolol. Clin Pharmacol Ther. 1979, 25:408-415.

67

143. Geisslinger G, Menzel S, Wissel K, Brune K. Pharmacokinetics of ketoprofen enantiomers after different doses of the racemate. Br J Clin Pharmacol. 1995, 40:73-75.

144. Blum MR, Liaq SH, De Miranda P. Overview of acyclovir pharmacokinetic disposition in adults and children. The American journal of medicine. 1982, 73:186-192.

145. Kerns EH, Di L, Petusky S, Farris M, Ley R, Jupp P. Combined application of parallel artificial membrane permeability assay and Caco‐2 permeability assays in drug discovery. J Pharm Sci. 2004, 93:1440-1453.

146. Swaan PW, Stehouwer MC, Tukker JJ. Molecular mechanism for the relative binding affinity to the intestinal peptide carrier. Comparison of three ACE-inhibitors: enalapril, enalaprilat, and lisinopril. Biochimica et Biophysica Acta (BBA)-Biomembranes. 1995, 1236:31-38.

147. Friedman DI, Amidon GL. Passive and carrier-mediated intestinal absorption components of two angiotensin converting enzyme (ACE) inhibitor prodrugs in rats: enalapril and fosinopril. Pharm Res. 1989, 6:1043-1047.

148. Legen I, Kristl A. pH and energy dependent transport of ketoprofen across rat jejunum in vitro. Eur J Pharm Biopharm. 2003, 56:87-94.

149. Fara J, Myrback R, Swanson D. Evaluation of oxprenolol and metoprolol Oros systems in the dog: comparison of in vivo and in vitro drug release, and of drug absorption from duodenal and colonic infusion sites. Br J Clin Pharmacol. 1985, 19:91S.

150. Tannergren C, Bergendal A, Lennernäs H, Abrahamsson B. Toward an increased understanding of the barriers to colonic drug absorption in humans: implications for early controlled release candidate assessment. Mol Pharm. 2009, 6:60-73.

151. Barr WH, Zola EM, Candler EL, Hwang S-M, Tendolkar AV, Shamburek R, Parker B, Hilty MD. Differential absorption of amoxicillin from the human small and large intestine. Clin Pharmacol Ther. 1994, 56:279-285.

152. Fuhr U, Staib A, Harder S, Becker K, Liermann D, Schollnhammer G, Roed I. Absorption of ipsapirone along the human gastrointestinal tract. Br J Clin Pharmacol. 1994, 38:83-86.

153. Lennernäs H. Regional intestinal drug permeation: Biopharmaceutics and drug development. Eur J Pharm Sci. 2013.

154. Sutton SC. The use of gastrointestinal intubation studies for controlled release development. Br J Clin Pharmacol. 2009, 68:342-354.

155. Oswald H, Sharkey M, Pade D, Martinez MN. Canine gastrointestinal physiology: Breeds variations that can influence drug absorption. Eur J Pharm Biopharm. 2015, 97:192-203.

156. Dressman JB. Comparison of canine and human gastrointestinal physiology. Pharm Res. 1986, 3:123-131.

157. Kesisoglou F, Balakrishnan A, Manser K. Utility of PBPK Absorption Modeling to Guide Modified Release Formulation Development of Gaboxadol, a Highly Soluble Compound with Region‐Dependent Absorption. J Pharm Sci. 2016, 105:722-728.

158. Lipka E, Spahn-Langguth H, Mutschler E, Amidon G. In vivo non-linear intestinal permeability of celiprolol and propranolol in conscious dogs: evidence for intestinal secretion. Eur J Pharm Sci. 1998, 6:75-81.

159. Sutton SC, Evans LA, Fortner JH, McCarthy JM, Sweeney K. Dog colonoscopy model for predicting human colon absorption. Pharm Res. 2006, 23:1554-1563.

68

160. Chiou WL, Jeong HY, Chung SM, Wu TC. Evaluation of using dog as an animal model to study the fraction of oral dose absorbed of 43 drugs in humans. Pharm Res. 2000, 17:135-140.

161. García-Arieta A. Interactions between active pharmaceutical ingredients and excipients affecting bioavailability: impact on bioequivalence. Eur J Pharm Sci. 2014, 65:89-97.

162. Vaithianathan S, Haidar SH, Zhang X, Jiang W, Avon C, Dowling TC, Shao C, Kane M, Hoag SW, Flasar MH. Effect of common excipients on the oral drug absorption of biopharmaceutics classification system class 3 drugs cimetidine and acyclovir. J Pharm Sci. 2016, 105:1355-1357.

163. Ghosh A, Scott DO, Maurer TS. Towards a unified model of passive drug permeation I: Origins of the unstirred water layer with applications to ionic permeation. Eur J Pharm Sci. 2014, 52:109-124.

164. Lichtenberg D, Robson RJ, Dennis EA. Solubilization of phospholipids by detergents structural and kinetic aspects. Biochimica et Biophysica Acta (BBA)-Reviews on Biomembranes. 1983, 737:285-304.

165. Artursson P, Lindmark T, Davis SS, Illum L. Effect of chitosan on the permeability of monolayers of intestinal epithelial cells (Caco-2). Pharm Res. 1994, 11:1358-1361.

166. Schipper NG, Vårum KM, Stenberg P, Ocklind G, Lennernäs H, Artursson P. Chitosans as absorption enhancers of poorly absorbable drugs: 3: Influence of mucus on absorption enhancement. Eur J Pharm Sci. 1999, 8:335-343.

167. Raoof AA, Chiu P, Ramtoola Z, Cumming IK, Teng C, Weinbach SP, Hardee GE, Levin AA, Geary R. Oral bioavailability and multiple dose tolerability of an antisense oligonucleotide tablet formulated with sodium caprate. J Pharm Sci. 2004, 93:1431-1439.

168. Lennernäs H, Gjellan K, Hällgren R, Graffner C. The influence of caprate on rectal absorption of phenoxymethylpenicillin: experience from an in‐vivo perfusion in humans. J Pharm Pharmacol. 2002, 54:499-508.

169. Nylander O, Pihl L. Luminal hypotonicity increases duodenal mucosal permeability by a mechanism involving 5‐hydroxytryptamine. Acta Physiologica. 2006, 186:45-58.

170. Nylander O. The impact of cyclooxygenase inhibition on duodenal motility and mucosal alkaline secretion in anaesthetized rats. Acta Physiologica. 2011, 201:179-192.

171. Nylander O, Pihl L, Perry M. Hypotonicity-induced increases in duodenal mucosal permeability facilitates adjustment of luminal osmolality. American Journal of Physiology-Gastrointestinal and Liver Physiology. 2003, 285:G360-G370.

172. Schwarz NT, Kalff JC, Türler A, Engel BM, Watkins SC, Billiar TR, Bauer AJ. Prostanoid production via COX-2 as a causative mechanism of rodent postoperative ileus. Gastroenterology. 2001, 121:1354-1371.

173. Fagerholm U, Johansson M, Lennernäs H. Comparison between permeability coefficients in rat and human jejunum. Pharm Res. 1996, 13:1336-1342.

174. Kubbinga M, Nguyen MA, Staubach P, Teerenstra S, Langguth P. The Influence of Chitosan on the Oral Bioavailability of Acyclovir—a Comparative Bioavailability Study in Humans. Pharm Res. 2015, 32:2241-2249.

Acta Universitatis UpsaliensisDigital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Pharmacy 257

Editor: The Dean of the Faculty of Pharmacy

A doctoral dissertation from the Faculty of Pharmacy, UppsalaUniversity, is usually a summary of a number of papers. A fewcopies of the complete dissertation are kept at major Swedishresearch libraries, while the summary alone is distributedinternationally through the series Digital ComprehensiveSummaries of Uppsala Dissertations from the Faculty ofPharmacy. (Prior to January, 2005, the series was publishedunder the title “Comprehensive Summaries of UppsalaDissertations from the Faculty of Pharmacy”.)

Distribution: publications.uu.seurn:nbn:se:uu:diva-358467

ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2018