European Journal of Pharmaceutics and Biopharmaceutics 142 (2019) 411–420

Contents lists available at ScienceDirect

European Journal of Pharmaceutics and Biopharmaceutics

journal homepage: www.elsevier.com/locate/ejpb

Research paper

Intestinal absorption-modifying excipients: A current update on preclinical T

in vivo evaluations
⁎
D. Dahlgrena, M. Sjöblomb, H. Lennernäsa,
a
Department of Pharmacy, Uppsala University, Uppsala, Sweden
b
Department of Neuroscience, Division of Physiology, Uppsala University, Uppsala, Sweden

A B S T R A C T

Pharmaceutical excipients in drug products are defined as pharmacologically inactive and are integral constituents of all types of oral dosage forms. However, some
excipients may increase drug absorption by interacting with the mucosal membrane. If the strategy is to use an excipient with a potential to affect the processes
determining the rate and/or extent of the intestinal drug absorption, it is defined as an absorption-modifying excipients (AME). These pharmaceutical excipients may
act as AMEs, depending on the amounts applied, and accordingly influence bioequivalence assessment of innovative and generic drug products, as well as enable oral
delivery of peptides and oligonucleotides. This review discusses the mechanisms by which AMEs increase drug absorption, and especially permeation step. The focus
is on the most recent data regarding how AMEs can be evaluated in preclinical models, with an emphasis on in situ and in vivo intestinal absorption models. The in
vivo predictive value of these models is reviewed for five factors of clinical relevance for the intestinal absorption performance: (a) effect and response rate of AMEs,
(b) mucosal exposure time and intestinal transit of AMEs, (c) intraluminal AME dilution and prandial state, (d) mucosal recovery and safety, and (e) variability in the
effects of the AMEs. We argue that any preclinical investigations of AMEs that fail to consider these processes will ultimately be of limited clinical value and add little
to our understanding of how excipients affect intestinal drug absorption.

1. Introduction/Background it is called a absorption-modifying excipient (AME). As fabs is influenced

by the dissolution, solubility, intestinal permeability and/or intestinal
Pharmaceutical excipients are integral constituents of most of oral transit time of the drug compound, AMEs can be sub-classified as per-
dosage forms and approved oral pharmaceutical products. They are meation-modifying excipients and/or transit-modifying excipients, re-
necessary to ensure a consistent and reproducible manufacturing pro- spectively. In addition, post-absorption processes, such as gut-wall
cess, and to optimize the physicochemical stability and various bio- metabolism (i.e., EG) of the API, might also be affected by the excipient
pharmaceutical properties of the drug product. In traditional oral drug [4]. It should be emphasized that pharmaceutical excipients may also
formulations, they are classified as pharmacologically and physiologi- be formulated intentionally to increase the intestinal fabs and bioavail-
cally inactive. The amounts included in these formulations are expected ability for low permeation compounds; they are then commonly re-
to result in luminal concentrations and epithelial exposures that only ferred to as permeation or absorption enhancers.
affect the drug product dissolution and API solubility. Most oral for- The gastrointestinal transit time of an API can be reduced with
mulations are designed not to directly interact with the gastrointestinal osmotically active excipients. The excipients can be used at luminal
(GI) epithelial tissue or its physiological regulation, because epithelial concentrations that increase the intestinal fluid volume to levels that
permeability, pancreatic secretion, and GI transit may affect the frac- stimulate the GI motility [5]. For instance, sorbitol and mannitol (at
tion dose absorbed (fabs) [1–3]. A few representative examples of oral doses between 2.2 and 2.5 g) are two sugar alcohols with low
pharmaceutical excipients and their functions are presented in Table 1. membrane permeability that significantly reduce the bioavailability of
Bioavailability (F) is the key pharmacokinetic parameter describing the BCS class III drugs, ranitidine and cimetidine, by 25% and 31%,
the fraction of an orally administered drug that reaches systemic cir- respectively [6,7]. In contrary, the intestinal transit time of an API may
culation (Eq. (1)): also be increased by an excipient that has transit prolonging properties
(e.g. mucoadhesive or motility inhibitor) [8,9].
F = fabs × (1 − EG ) × (1 − EH ) (1)
Gut-wall metabolism (first-pass extraction, EG) of the API may be
where EG and EH are the fractions extracted in the gut wall and liver, affected if an excipient interacts with any of the phase I (CYP450) and/
respectively. If an excipient has the potential to affect the processes or phase II (transferases) metabolic enzymes in the enterocytes [4]. For
determining the rate and extent of the intestinal F (i.e both fabs and EG), instance, the intrinsic clearance (rat liver and intestinal microsomes) of

⁎
Corresponding author at: Department of Pharmacy, Uppsala University, Box 580, SE-751 23 Uppsala, Sweden.
E-mail address: [email protected] (H. Lennernäs).

https://doi.org/10.1016/j.ejpb.2019.07.013
Received 5 March 2019; Received in revised form 27 May 2019; Accepted 11 July 2019
Available online 12 July 2019
0939-6411/ © 2019 Elsevier B.V. All rights reserved.
D. Dahlgren, et al. European Journal of Pharmaceutics and Biopharmaceutics 142 (2019) 411–420

Table 1
Representative examples of categories of pharmaceutical excipients used in solid oral drug products and their function(s) in the oral formulations [90].
Categories Function(s) Specific examples

Filler, binder, diluent Improve handling and mechanical strength, and secure dosing uniformity lactose, starch, mannitol microcrystalline
cellulose
Preservatives Improve shelf life by securing API and drug product stability antioxidants such as butylated
hydroxytoluene
Glidants and lubricants Reduce particle adhesion, cohesion, and friction magnesium stearate, talc
Disintegrants Promote disintegration of a tablet/granule into smaller particles. starch, crospovidone
Wetting and solubilizing agents Prevent particle aggregation and increase particle dissolution rate sodium dodecyl sulphate (SDS) and
cyclodextrin
Tablet coatings Improve stability and appearance, mask bad taste, facilitate swallowing, and control tablet and hydroxypropyl methylcellulose, cellulose
granule disintegration (e.g., with enteric coatings) acetate

the CYP3A4 substrate, midazolam, is reduced, and consequently the permeation-enhancing excipients are reviewed later in this article.
plasma exposure is increased, by a number of excipients (e.g. Cremo- The major objective of this update is to review the recent data of
phor, Tween, and HPMC) [10]. The mechanism(s) of the reduced in- how AMEs affect passive transcellular and/or paracellular diffusion of
trinsic clearance and first-pass extraction are the altered enzyme reg- drugs in the small intestine. The implications for bioequivalence are
ulation and/or the inhibition of enzymatic activity [10]. Less common discussed with respect to its effects on assessment of innovative and
mechanisms are: excipient-induced induction of CYP activity, inhibition generic drug products, and briefly for oral delivery of new modalities
of phase II enzymes, and interaction of the excipient with hepatic me- such as peptides and oligonucleotides. The focus is on AMEs in pre-
tabolism [4,11]. At present, the in vivo relevance of excipients inter- clinical models, with an emphasis on the ones for in situ and in vivo
fering with intestinal metabolism is uncertain, as it has scarcely been intestinal absorption. The in vivo predictive value of these models is
investigated. To us, this scarcity of reports suggests that interactions reviewed for five factors of clinical relevance in the intestines: (a) effect
between excipients and gut wall enzymes are of limited clinical im- and response rate of AMEs, (b) mucosal exposure time and intestinal
portance and animal findings as above is due to very high amounts of transit of AMEs, (c) intraluminal AME dilution and prandial state, (d)
excipients applied. mucosal recovery and safety, and (e) variability in the effects of the
Intestinal permeability is a key biopharmaceutical variable (to- AMEs.
gether with solubility, dissolution rate, dose, and GI transit) that de-
termine the in vivo rate and extent of intestinal drug absorption
[12–16]. The permeation of a drug across the intestinal epithelium 2. Effects of AMEs on the intestinal barrier
depends on one, or several, of multiple, parallel transport processes that
include passive transcellular/paracellular diffusion, carrier-mediated An AME, such as the mucolytic agent, n-acetylcysteine (NAC), may
transport, and carrier-mediated efflux (Fig. 1) [17]. The net intestinal increase transmucus drug transport by reducing the thickness, visc-
permeability of a drug molecule is determined by its chemical structure osity, and/or porosity of the intestinal epithelial mucus layer (Fig. 2a).
and physicochemical properties, and by the physiological and bio- A reduction of the mucus barrier may result in an overall increase in
chemical conditions along the GI tract [18–21]. The physiologically drug permeability, in those cases where the diffusion across the mucus
active site of an AME determine its classification: transmucus, trans- and aqueous boundary layer is the rate-limiting permeation step [22]. A
cellular, and/or paracellular (Fig. 2a–d) [1]. These three classes of number of in vivo studies in humans and other animals have shown that
these transport processes are not rate-limiting even for drugs that have
high membrane permeability. For instance, the permeability of keto-
API profen is very high in both the jejunum and colon of humans
(> 3.4 × 10−4 cm/s) and rats (> 1.5 × 10−4 cm/s), despite that the
colonic mucus is substantially thicker (human 200 vs. 15 µm; rat 800 vs.
1 2 3 4 1
200 µm) and more strongly adherent than the mucus in the jejunum
Lumen Lumen
[15,23–25]. Further, co-administration of NAC, has no effect on the
absorption rate of ketoprofen in the rat single-pass intestinal perfusion
(SPIP) model SPIP model [1]. Nevertheless, mucus may affect the
permeation of compounds that interact with it physically (i.e. hydro-
phobic or ionic interactions), or sterically. These interactions between
the mucus and smaller solutes are generally insignificant, but an in-
API creased drug lipophilicity correlates slightly in vitro with a reduction in
transmucus diffusion rate [24,26]. Complete steric interaction has been
seen for larger peptides/proteins (> 12 kDa) and solid particles (about
200 nm in diameter, depending on shape and surface properties) [27].
Consequently, NAC has the potential to increase the transmucus diffu-
sion rate of particles, and of peptides/proteins with a MW from about
Blood Blood 2000 g/mol [27,28]. However, the use of mucolytic agents for in-
creasing drug permeation is still only an experimental strategy that will
most likely require combination with a transcellular or paracellular
permeation enhancer to be effective
An AME may increase transcellular drug diffusional transport by
Fig. 1. The transport mechanisms across the intestinal epithelial barrier that increasing the fluidity and/or by reducing the integrity of the mem-
determine the net permeability of a luminally dissolved compound. (1) Passive brane lipid bilayer of the epithelial cells (Fig. 2b) [29]. This leads to an
paracellular diffusion, (2) passive transcellular diffusion, (3) absorptive carrier- increased partitioning into, and/or higher diffusion rate across, the li-
mediated transport, and (4) efflux carrier-mediated transport. poidal membrane. One common property for these AMEs are that they

412
D. Dahlgren, et al. European Journal of Pharmaceutics and Biopharmaceutics 142 (2019) 411–420

Fig. 2. The mechanisms by which an absorption-modifying excipient (AME) can affect the intestinal permeability of an active pharmaceutical ingredient (API). (A)
An AME reducing the thickness, viscosity, and/or porosity of the intestinal mucus layer lining the epithelial barrier may increase the rate at which a drug is presented
to the epithelial membrane. (B) An amphiphilic AME (e.g., surfactant) can be inserted into the lipoidal epithelial cell membrane and thereby increase its fluidity/
porosity, and potentially dissolve phospholipids in the epithelial cell membrane. (C) An AME inhibitor of a membrane protein transporter may increase, or reduce,
drug intestinal permeability, depending on if it is an influx (blue) or efflux (red) transporter. (D) An AME interacting, directly or indirectly, with tight junction protein
components may increase the paracellular diffusion of an API.

are amphiphiles (e.g., bile acids, fatty acids, alkyl sulphates). Conse- bioavailability of the P-gp substrate, etoposide, by reducing its P-gp
quently, they can insert into the plasma membrane, and potentially mediated efflux [33]. However, the contribution from other unspecific
solubilize constituents of the membrane, thereby reducing its integrity surfactant mechanisms cannot be excluded, such as an increased
[30]. Generally, the permeation enhancing effect of a surfactant cor- membrane fluidity, and/or reduced membrane integrity.
relates with its ability to interact with the membrane (where large An AME may increase paracellular drug transport by interacting,
hydrophobic regions favor interaction), and to its critical micelle con- directly or indirectly, with tight junction protein components, such as
centration (CMC) [31]. A high CMC results in a higher concentration of claudins and occludin (Fig. 2d) [34]. These AMEs are classified as first
the surfactant in its free monomeric form, which explains why anionic or second generation, which is determined by the site of action—first
surfactants are usually more potent than non-ionic ones. (the changed generation AMEs such as sodium caprate affect intracellular signalling
headgroups of the anionic surfactants repel each other, resulting in a pathways, and second ones by interacting directly with extracellular
higher CMC [30].) Luminal concentrations above the CMC may cause components of the tight junctions [31]. Caprate has been proposed to
primarily hydrophobic drugs (or drugs that have been hydrophobised affect the intracellular signalling pathways mediated by phospholipase
through interaction with the surfactant) to partition into the micelles. C, a protein that increases intracellular Ca2+ and diacylglycerol
This reduces the free drug concentration, leading to a reduced rate and [35,36]. These signalling pathways control the contraction of actin and
extent of intestinal absorption [32]. Finally, the transcellular perme- myosin II, and removes tricellulin from the tricellular tight junctions,
ability may also be affected if an excipient inhibits a drug transporter which opens the paracellular tight junctions [35]. It is also important to
protein (Fig. 2c) [4]. How such an interaction affects the membrane consider that caprate has amphiphilic properties, and may act as a
permeation depends if the drug is a substrate for an efflux and/or influx transcellular permeation enhancer [36]. Many of the intra/inter-cel-
transporter. For instance, it has been proposed that polysorbate 20 in- lular pathways affected by first-generation paracellular AMEs are
creases the rat intestinal absorption and consequently the oral physiologically regulated. For instance, it is well-established that

413
D. Dahlgren, et al. European Journal of Pharmaceutics and Biopharmaceutics 142 (2019) 411–420

luminal hypotonicity and glucose regulate paracellular ion diffusion modalities–are generally not orally administered because of their low
and convection, as well as the permeation of low permeability low intestinal stability, permeability and/or extensive first-pass metabolism
molecular mass marker compounds [37,38]. These physiological reg- [48]. Nonetheless, by incorporating permeation-enhancing excipients
ulations of paracellular permeability are generally regarded to be of in a drug formulation, or using mucus/cell penetrating particles, ad-
little quantitative importance for in vivo drug absorption, as the pro- vocates of oral peptide delivery are revisiting this drug delivery issue,
cesses are fast and transient. However, they may still have an impact on hoping to enable oral administration of these compounds [49–51].
results in preclinical models [39,40]. The second-generation para- However, this experimental strategy has already been investigated
cellular AMEs, whereof many are experimental peptides, interact di- during several decades, with limited success, and there are currently no
rectly with extracellular motifs of tight junction proteins, such as oc- approved oral drug products using pharmaceutical excipients to in-
cludin and claudin [41,42]. crease intestinal drug permeation of any type of compound [52–54].
This has been attributed to the weak, and highly variable, effects of
3. AMEs – relevance for bioequivalence and drug delivery AMEs in vivo, as well as to safety issues [54,55]. For permeation-en-
hancing drug delivery technologies to reach a broad clinical break-
As outlined above, AMEs incorporated into oral drug formulations through, these issues must be resolved. The evaluation of AMEs and
may affect the intestinal permeability of any API, regardless of the BCS their impact on oral drug delivery and safety in different preclinical
classification of the API. This has implications for regulatory bioequi- models are therefore discussed in this review.
valence assessments when pharmaceutical excipients are used at
amounts that affects the rate and/or extent of intestinal absorption of 4. Preclinical models used to evaluate AMEs
the drug. In addition, an AME may have permeability enhancement as a
key effect in a drug formulation if intentionally used at higher doses There are many preclinical models in pharmaceutical science to
and luminal concentration, and/or at a special local condition (for in- investigate membrane permeation and predict human intestinal drug
stance, a drug delivery system with muco-adhesion properties). The absorption [56,57]. These range from simple in vitro tools to complex
implications of AMEs for both bioequivalence assessment and oral de- in vivo animal models. One commonly used in vitro system is the dif-
livery of drugs are discussed below. fusion chamber, in which the drug flux is determined across a biological
membrane separating two compartments. This barrier can be composed
3.1. Bioequivalence assessment of a monolayer of cells, such as Caco-2 or MDCK cells, or an excised
tissue, such as intestinal segments derived from animals or humans
Two drug products are defined as bioequivalent if the test and re- [58,59]. An example of a more in vivo relevant tool is the single-pass
ference drug products display the same rate and extent of F, and this is intestinal perfusion (SPIP) model, in which a drug solution is perfused
demonstrated in vivo in healthy subjects by plasma PK (AUC, Cmax, through a defined intestinal segment, and the luminal drug dis-
Tmax) [43,44]. Bioequivalence may also be established on the basis of in appearance is evaluated, and/or appearance in plasma, per intestinal
vitro dissolution tests—this is called a BCS-based biowaiver [45]. A surface area [1]. Intestinal perfusions are most often performed in rat,
BCS-based biowaiver can only be used for immediate-release drug mouse, dog and pig, but extensive experiments have been performed in
products, and only if the drug compound has a high solubility, and high human as well [14,20,60,61]. The most clinically relevant preclinical
or low permeability (BCS I and III, respectively). The FDA and EMA also approach is to evaluate GI drug absorption by assessing plasma PK
require that the test products contain known excipients, at doses following intraintestinal, intragastric, or oral, administration to ani-
documented as safe, that do not affect intestinal transit and/or per- mals, such as rat and dog [62]. The biopharmaceutical evaluation of
meability [44,45]. Further, only small differences ( ± 10%) in excipient AMEs in all the above models can be performed in a variety of ways,
doses between the test and reference drug products are allowed. As the depending on the question being investigated. The typical study design
exact amounts of excipients are undisclosed for an approved product, is based on an evaluation of changes in, for example, drug absorption/
this consequently results in many unnecessary clinical bioequivalence flux or mucosal injury/recovery, induced by an AME compared to a
studies being performed, as the risk for bioinequivalence is minor for control period/group.
BCS class I drugs. However, in an evaluation of excipient effects there
are claims that AMEs may affect the bioequivalence of BCS class I drugs, 4.1. Potential of the different models
based on undisclosed human bioequivalence data from a drug product
application of 1 mg risperidone formulated with 3.7 mg SDS in the re- The cell/tissue-based diffusion-chamber model is suitable for high
ference product; reductions in AUC and Cmax were observed, but there throughput because multiple AMEs can be investigated at a range of
were no effects on the dissolution profiles [46]. This led to a claim that concentrations and exposure times [49,63]. It is also suitable for a
an excipient effect cannot be generalised, but rather that it is both dose mechanistic evaluation of AMEs, as some experimental conditions can
and/or drug dependent [47]. Such a claim would inevitably lead to a be controlled. Changes can be visualized, for instance, in cytoskeletal
situation where in vivo bioequivalence assessment in human healthy and tight junction protein conformation, as well as epithelial membrane
subjects is routinely performed on all oral formulations containing ei- damage [29,63–65]. The disadvantage of the diffusion-chamber models
ther new AME concentrations, or known AMEs in combination with is the absence of neural and hormonal feedback mechanisms, because
new APIs. On the basis of our long experimental experience with animal these are considered to affect epithelial functions. The SPIP model al-
and human in vivo studies, we consider that such a conservative ap- lows for controlled conditions in the lumen of the intestinal segment
proach is exaggerated. This review therefore discusses what properties and visualization of tissue following AME exposure, but it is associated
and concentrations of AMEs that may result in bioequivalence issues for with a lower throughput and higher cost than the in vitro systems
different drug compounds and products, and the possibility to evaluate [66,67]. The major advantage is that the perfusions are performed in
this in preclinical models. However, a clear definitions of a safe space vivo, with a dynamic intestinal blood flow and feedback mechanisms,
for the use of excipients and critical excipients (AMEs) formulation which increases the robustness and in vivo relevance of the model. The
require an international collaborative project involving academia, most relevant in vivo preclinical model is the evaluation of the plasma
regulatory agencies and to some extent pharmaceutical industry. PK of a drug in an animal following oral/intraintestinal administration
of the drug together with an AME formulated as, for instance, a solu-
3.2. New modalities – peptides and oligonucleotides tion, immediate-release or multiple-unit enteric coated dosage form
[52,62]. The effect of the AME in the PK models is expected to be af-
Pharmaceutical peptides and oligonucleotides–i.e. new fected by, among other things: intraintestinal spreading and dilution of

414
D. Dahlgren, et al. European Journal of Pharmaceutics and Biopharmaceutics 142 (2019) 411–420

Fig. 3. (A) The lowest reported concentration at which sodium dodecyl sulfate (SDS) increases the permeability of low-permeability small model compounds
(mannitol, phenol red, and amoxicillin). The exposure time was 20 min in the Caco-2 model, 210 min in the rat jejunal Ussing model, and 75 min in the rat single-pass
intestinal perfusion (SPIP) model [1,91,92]. The administered volume in the rat intraintestinal bolus model was 0.5 mL (5.7 mg SDS), and 2 mL (20 mg SDS) in the rat
oral bolus study [62,93]. (B) Reported mean ( ± SEM) increase in jejunal absorption ratio of the low-permeability drug enalaprilat, induced by SDS and chitosan in
the rat SPIP (red) and intraintestinal bolus (blue) models, at different SDS concentrations (mg/mL). The intestinal exposure time was 75 min in the rat SPIP model,
and the administered volume in the rat intraintestinal bolus model was 0.5 mL (5.7 mg SDS) [1,62]. (For interpretation of the references to colour in this figure
legend, the reader is referred to the web version of this article.)

the AME; precipitation; complex binding with colloidal structures; ab- a strong impact on the local concentration–time profiles along the in-
sorption of the AME; and intestinal transit of the AME/drug. This testines, and consequently on both intra- and interindividual varia-
multitude of factors renders the PK model less suitable for mechanistic bility. Large GI fluid volumes reduce the likelihood of effective luminal
evaluations— even if the model is the most in vivo relevant—as it can AME concentrations due to dilution and rapid GI transit, whereas fatty
be challenging to differentiate the individual contribution of the dif- chyme and endogenous bile salts/acids may reduce the free monomeric
ferent factors. concentrations of AME surfactants acting on the membrane [69–71].
For safety reasons, the intestinal mucosa should have a rapid recovery
4.2. Effect of AMEs in different preclinical models following an AME exposure to avoid translocation of harmful luminal
contents, which could initiate intestinal inflammation [55,72]. The
It is a well-recognized pattern that the luminal (donor) concentra- AME itself should have a good local and systemic safety profile, which
tion at which an AME is effective needs to be increased the more in is especially important if the AME itself is absorbed and causes a sys-
vivo-like the applied model is (e.g., the effect of caprate: Caco- temic exposure. Consequently, advanced formulation designs of com-
2 > Ussing > SPIP > oral bolus) [36]. The higher sensitivity (i.e. plicated AME-containing products require the use of relevant in vivo
lower AME exposure required) in in vitro models can be attributed to predictive absorption models. The above mentioned physiological and
the factors outlined in the previous section, whereas the more in vivo- biopharmaceutical processes determine the clinical effect of an AME.
like models have intact blood flow and mucosal repair mechanisms. In These processes—and how they can be investigated in preclinical
addition, the oral/intraintestinal bolus models are influenced by in- models—are discussed in the following sections.
testinal AME dissolution/release, dilution, absorption, and transit. This
is exemplified by the AME concentration needed to increase the per- 5.1. Response rate, mucosal exposure time, and intestinal transit
meability of low-permeability model compounds in different preclinical
models (Fig. 3a and b). The exposure times and administered volumes If the response rate of an AME on the intestinal mucosal membrane
vary in the different models, but there is a clear trend that higher is slower (including delayed onset time) than the intestinal transit, the
epithelial exposure (concentrations or doses) is needed in the more in AME is predicted not to be able to increase the intestinal absorption of
vivo relevant models. the API to clinically effective plasma levels [73]. Therefore, any model
(in vitro or in vivo) using long AME exposure times may therefore
5. Ideal biopharmaceutical properties of an AME formulation generate data that is not clinically relevant. This is illustrated by the rat
intended for an absorption enhancing effect SPIP model, where a mucosal AME exposure time of 60-min generated
an increase in model compound absorptive fluxes that were sub-
If the purpose of an AME incorporated into an oral drug product is stantially higher than those observed in the rat intraintestinal bolus
to increase the absorption of an API with low intestinal permeation, model [1,62]. However, when a more in vivo relevant exposure time of
some ideal properties can be postulated for it (conversely, these prop- 15-min was used in the same SPIP model, the data was predictive of the
erties should be avoided if the pharmaceutical excipients are not to rat bolus data, but only if both maximum absorptive flux, and time to
enhance intestinal absorption.) The AME needs to be released from the maximum absorptive flux, were considered [68]. Consequently, only a
formulation sufficiently fast and synchronized with the API release, as high AME induced flux (Fig. 4a) was not predictive of the effect in the
the API needs to be present at the intestinal mucosal barrier that has bolus model, and neither was only a rapid time to maximum flux
been compromised. The onset-time for the AME effect should be short (Fig. 4b). This combination of both a rapid onset and a high effect on
to obtain increased intestinal drug absorption. If the local effect is slow, the intestinal mucosa was only observed for SDS at a high dose in the
the API might be transported away from the targeted intestinal site 15-min SPIP model, and SDS at a high dose was the only AME that was
prior the AME has exerted its sufficient effect on the intestinal mucosa effective at increasing drug absorption in the rat intraintestinal bolus
[68]. The maximum duration during which an AME exerts its effect(s) is model [1,68].
also reduced if the AME itself is absorbed, which is the case for caprate These three studies - the 60- and 15-min SPIP studies, and the rat
and SDS [64,65]. Luminal water volumes and contents, which vary intraintestinal bolus study - altogether indicate that the increase in
substantially between intestinal segments and prandial states, also have absorptive flux (or permeability) observed in preclinical models, such

415
D. Dahlgren, et al. European Journal of Pharmaceutics and Biopharmaceutics 142 (2019) 411–420

Fig. 4. Reported mean ( ± SEM) acyclovir

absorptive flux (Jabs) during a 150-min rat
single-pass intestinal perfusion (SPIP) [68].
Sodium dodecyl sulfate (SDS) and chitosan
were only added to the perfusate in the 15-
min test period (brown); the periods before
and after were control periods in which the
base-line value and recovery time were
evaluated, respectively. (A) Comparison of
SDS and chitosan at 5 mg/mL. (B) Compar-
ison of SDS at 1 and 5 mg/mL.

as Caco-2 and SPIP, has limited in vivo relevance, unless time–-

dependent parameters, such as response rate and intestinal transit, are
considered [1,62,68]. However, it should be stressed that time–-
independent models, such as evaluation of absorption ratio in the Us-
sing and SPIP models, may still be useful screening tools for new AMEs.
They may also be useful in the investigation of the mechanisms of ac-
tion of an AME, and for investigation of the potential effect of an AME in
vivo [1]. Readers with a broad and focused interest in AME response
rate and intestinal transit, and their impact on oral drug delivery, are
referred to a recent review by Maher et al., 2019 [74].

5.2. Intraluminal effects

The primary determinant of an AME effect on the mucosal mem-

brane is the free concentration adjacent to the epithelial barrier. This
concentration can be reduced by different in vivo factors, the most Fig. 5. The increase in jejunal rat 51Cr-EDTA clearance (CLCr-EDTA) ratio of a test
obvious one being dilution in the intestinal fluid. Dilution may be even formulation compared to a control solution perfused in the same animal. The
control solution contained buffer at pH 7.4 with no caprate, while the test
more pronounced if the formulation is dissolved and diluted already in
formulation contained buffer at pH 7.4 with caprate at 5, 10, and 20 mg/mL.
the stomach, and only gradually released into the small intestine, which
Caprate at 5 mg/mL was at the saturation concentration at pH 7.4, while cap-
is usually the targeted intestinal region for AME effects and drug ab-
rate at 10 and 20 mg/mL was above its solubility (suspension).
sorption [69]. These gastric dilution effects can be resolved by enteric
coating of a multiple unit formulation [54]. Gastric emptying, as well as
random intestinal motility, may spread the AME/drug over a larger form mixed micelles with endogenous surfactants (e.g., bile salts) or
section of the intestines. This may result in a lower total dose at any food components (e.g., phospholipids). Such effects have been observed
local mucosal site, and reduce the effect of the AME, even if the free in the Caco-2 and rat closed-loop models, in which surface active al-
AME concentration is unchanged. The risk is higher if the AME itself is kylmaltosides were co-administered with mixed micelles [78]. Pre-
absorbed. A reservoir of undissolved AME is then necessary to replenish liminary results from our laboratory also found this for SDS in the rat
any absorbed AME to maintain a constant free concentration adjacent SPIP model; there was a substantial reduction in the permeation en-
to the membrane. Preliminary results from our laboratory demon- hancing effect of SDS at 1 and 5 mg/mL when co-administered with
strated this effect—a saturated solution of caprate (5 mg/mL) at pH 7.4 FeSSIF, compared to FaSSIF, which has a lower content of fatty acids,
was insufficient to increase the permeability of the mucosal barrier phospholipids, and bile salts (Fig. 6). The effects of ion concentration
marker, 51Cr-EDTA, in the rat SPIP model. There must be an excess of and food components clearly illustrate the importance of considering
undissolved caprate (i.e., 10 and 20 mg/mL) in the perfusate (Fig. 5). relevant in vivo luminal conditions when evaluating AMEs [78].
Prandial state and diet are expected to have a strong effect on the
outcome of an AME on intestinal absorption, which is thoroughly dis- 5.3. Gastrointestinal mucosal repair processes and safety
cussed by Maher et al, 2019 [74]. For instance, ion concentration af-
fects the CMC of ionic surfactants, (e.g., SDS has a CMC between 2 and The effect and safety of an AME are directly related to the function
8 mM depending on NaCl concentration) [75,76]. High ion concentra- of the GI mucosa to uphold homeostasis. The risk for safety issues is
tions shield the repulsion between ionic head groups, reducing the CMC highest for potent AMEs such as claudin inhibitors, or AMEs adminis-
and hence free monomer concentration, which is the primary de- tered at high doses (and high local concentrations), where the ambition
terminant of their effect on the mucosa [31]. Ion content may also af- is to affect the mucosal barrier [74]. High doses may not only increase
fect the physicochemical properties of charged AME polymers, such as drug permeation, but also translocate bacteria, viruses, and proteins.
chitosan, by de-hydrating the polymer and by reduced charge repulsion Such events may initiate intestinal inflammation, as well as dysregulate
[77]. However, changing the NaCl concentration in a rat SPIP study did hormonal and neural processes involved in mucosal homeostasis. The
not affect the SDS-and-chitosan-induced increase in membrane perme- risk for these effects depends on the frequency and duration of the oral
ability [40]. In contrast, membrane permeability increased when the drug treatment, as harmful epithelial effects may only show following
ion concentration was reduced with caprate, another AME with sur- long-term exposure.
factant properties. The mechanism behind this is unclear, but it is most Simpler absorption predictive systems, such as Caco-2 and the
likely not related to alterations of the CMC of caprate (between 25 and Ussing chamber, lack many fundamental physiological, neural, and
140 mM in saline, depending on reference), as the CMC is higher than hormonal processes involved in the homeostasis and repair of the epi-
the concentrations used in the perfusion study (5–25 mM) [36,40]. The thelial barrier [79–81]. The intact blood supply in the in vivo models
free concentration of amphiphilic AMEs can also be reduced if they (i.e., SPIP- and PK-models), effectively oxygenates and brings nutrients

416
D. Dahlgren, et al. European Journal of Pharmaceutics and Biopharmaceutics 142 (2019) 411–420

mL). Complete recovery of marker compounds to their baseline per-

meability value was 250 min in Caco-2, and 30–60 min in the rat SPIP
study [64,68]. It is uncertain which of the physiological mechanisms
(blood supply, villus contraction, restitution, or sealing of tight junc-
tions) is primarily responsible for the faster GI mucosal recovery in the
SPIP model - or if it was related to a greater membrane damage in the in
vitro model - but the results are a good illustration of the important
differences between the two intestinal absorption models.
However, there is a limit to the repair of the GI mucosa in the in
vivo models. Following a 60-min intestinal mucosal exposure of SDS
and chitosan (5 mg/mL each) in the rat SPIP model, there was only a
50% recovery of the model compounds absorptive flux during a 60-min
recovery period [68]. The same experiment also showed that the blood-
to-lumen transport of the clinical intestinal barrier marker, 51Cr-EDTA,
was elevated, and unaffected, during the whole 60-min recovery period.
A reduction in drug flux does not necessarily equate with a healthy
Fig. 6. Preliminary results showing the increase in mean ( ± SEM) enalaprilat mucosa, and safety evaluations should preferably be coupled to histo-
absorption flux (Jabs) ratio of a test formulation compared to a control solution logical examinations [63,71]. Nevertheless, the most difficult task is to
perfused in the same rat jejunum. The control solution contained FaSSIF or
categorize the risk of long-term treatment with an AME. Such studies
FeSSIF with no sodium dodecyl sulfate (SDS), while the test formulation con-
need repetitive treatment—using in vivo relevant models—to monitor
tained FaSSIF or FeSSIF with SDS at 1 or 5 mg/mL.
mucosal inflammation and the general health of the intestines and the
animal. Complete evaluation of these long-term effects and their im-
to the intestinal mucosa. It also removes any absorbed AME and cellular plication for different patient populations is difficult to assess, and the
waste, which ultimately results in a lower degree of their local accu- final risk-benefit analysis has to be evaluated in large, clinical studies, if
mulation. The hormonal and neural regulation of the supportive tissue pre-clinical toxicity studies allow human studies.
in the different models strongly influence the response of the epithelial
barrier to damage and AME exposure. For instance, enteric neurons
may induce contraction of smooth muscle fibers extending into the 5.4. Variability in absorption-modifying excipient effects
villus core, which reduces the height of the villus as well as the exposed
villus surface area [82]. This defence mechanism mitigates epithelial A major hurdle for oral peptide and nucleotide oral delivery is the
damage by reducing the surface area of the epithelial defect [83]. The notoriously low and highly variable rate and extent of GI absorption
impact of this phenomenon on drug absorption was possibly observed [48]. Incorporation of an AME in an oral drug delivery system might
in a recent rat SPIP study with SDS and chitosan at 5 mg/mL. The in- resolve both these issues. For instance, oral administration of an oli-
creased absorptive flux of the model compounds induced by SDS and gonucleotide (ISIS 104838) together with 660 mg caprate in fasted
chitosan was temporarily reduced during a 60-min perfusion [68]. humans resulted in an oral bioavailability of 12.0 ± 8.78%, without
Arguably, the most important epithelial mechanism for preventing/ resolving variability (F ranged between 2.0 and 27.5%) [86]. In another
resisting AME effects is its ability to repair itself. Every 72 h the epi- example, the bioavailability of semaglutide (MW 4113 g/mol, HBD/A
thelial barrier is renewed, and every day billions of cells are shed 57/63, cLogP −5.8) following oral single-dose administration in fasted
without affecting the mucosal barrier [55]. These cells are regenerated dogs was 0.29% together with 600 mg SNAC, but with a coefficient of
by stem cells situated in the crypt region of the intestinal epithelium, a variance close to 200% (data from patent: wo2012080471). It is not
process that is accelerated by epithelial damage [84]. Barrier defects easy to identify the source of this interanimal variability of ISIS 104,838
are also rapidly healed by a process called restitution, in which adjacent and semaglutide, It is most likely a combination of biopharmaceutical
healthy epithelial cells are mobilized [83]. Restitution is followed by a and physiological factors, such as failure to release drug and AME at the
final process whereby the paracellular space between the cells is sealed same place, fast intestinal transit, luminal AME dilution, low intestinal
tight. This process is proposed to be the most important process in permeability and luminal drug degradation.
barrier recovery [83,85]. Another factor worth noting is the surprisingly large variation in
The effect on AMEs by the different GI mucosal repair mechanisms intestinal mucosal response upon AME exposure–consistently there are
has not been systematically investigated in detail in the different ab- a few animals not responding. This is exemplified by the plasma con-
sorption models. However, two studies compared a 15-min exposure of centration of enalaprilat, when perfused together with SDS at 5 mg/mL
SDS to Caco-2 cells (0.1 mg/mL) and in the rat SPIP model (1 and 5 mg/ during 75 min in the rat SPIP model, and following an 0.5 mL in-
traintestinal bolus together with SDS at 11.4 mg/mL (Fig. 7a and b) [1].

Fig. 7. (A) Reported individual plasma

concentration–time profiles of enalaprilat,
perfused during 150 min in the rat SPIP
model [1]. The AME, sodium dodecyl sulfate
(SDS) at 5 mg/mL, was added to the perfu-
sate solution between 75 and 150 min. (B)
Reported individual rat plasma con-
centration–time profiles of enalaprilat fol-
lowing an intrajejunal bolus administration
of enalaprilat alone (control), and together
with SDS at 11.4 mg/mL [62].

417
D. Dahlgren, et al. European Journal of Pharmaceutics and Biopharmaceutics 142 (2019) 411–420

Our group has observed that it was always at least one animal in each References
study group for which there was no increase in absorptive flux.
{Dahlgren, 2018 #515;Dahlgren, 2018 #559;Dahlgren, 2018 [1] D. Dahlgren, C. Roos, A. Lundqvist, P. Langguth, C. Tannergren, M. Sjöblom,
#585;Dahlgren, 2017 #445} These studies covered different AMEs, E. Sjögren, H. Lennernas, Preclinical effect of absorption modifying excipients on
rat intestinal transport of five model compounds and the intestinal barrier marker
AME concentrations and intestinal exposure times, and two models (rat 51Cr-EDTA, Mol. Pharm. 14 (2017) 4243–4251.
SPIP and intraintestinal bolus) [1,40,62,68]. These non-responders [2] S. Vaithianathan, S.H. Haidar, X. Zhang, W. Jiang, C. Avon, T.C. Dowling, C. Shao,
occurred even in spite of the controlled luminal conditions in the SPIP M. Kane, S.W. Hoag, M.H. Flasar, Effect of common excipients on the oral drug
absorption of biopharmaceutics classification system class 3 drugs cimetidine and
model, in which the same mucosal segment is exposed to a defined acyclovir, J. Pharm. Sci. 105 (2016) 1355–1357.
concentration. On the basis of these studies and published data from [3] D.A. Adkin, S. Davis, R. Sparrow, P. Huckle, A. Phillips, I. Wilding, The effects of
clinical studies evaluating AMEs, we conclude that variability between pharmaceutical excipients on small intestinal transit, Br J Clin Pharmacol. 39
(1995) 381–387.
and within animals is a major hurdle in designing a robust AME-based [4] W. Zhang, Y. Li, P. Zou, M. Wu, Z. Zhang, T. Zhang, The effects of pharmaceutical
oral formulation strategy for new modalities, such as peptides and excipients on gastrointestinal tract metabolic enzymes and transporters—an update,
oligonucleotides. AAPS J. 18 (2016) 830–843.
[5] M.-L. Chen, N. Sadrieh, L. Yu, Impact of osmotically active excipients on bioa-
The knowledge of factors that explain the variability between and
vailability and bioequivalence of BCS class III drugs, AAPS J. 15 (2013) 1043–1050.
within individuals needs to be improved. Alternatively, the API must be [6] D.A. Adkin, S.S. Davis, R.A. Sparrow, P.D. Huckle, I.R. Wilding, The effect of
designed with pharmacokinetic/pharmacodynamic properties for mannitol on the oral bioavailability of cimetidine, J. Pharm. Sci. 84 (1995)
which a large day-to-day variability in intestinal absorption does not 1405–1409.
[7] M.-L. Chen, A. Straughn, N. Sadrieh, M. Meyer, P. Faustino, A. Ciavarella,
affect the clinical outcome. For instance, variability issues are cir- B. Meibohm, C. Yates, A. Hussain, A modern view of excipient effects on bioequi-
cumvented for the peptide drug, desmopressin, because it has a flat valence: case study of sorbitol, Pharm. Res. 24 (2007) 73–80.
concentration–effect curve, where a maximum effect is achieved al- [8] H.L. Lueßen, B.J. de Leeuw, M.W. Langemeÿer, A.B.G. de Boer, J.C. Verhoef,
H.E. Junginger, Mucoadhesive polymers in peroral peptide drug delivery. VI.
ready at oral doses of 250 ng [87]. Semaglutide, a peptide being ex- Carbomer and chitosan improve the intestinal absorption of the peptide drug bu-
plored for oral delivery together with an AME (SNAC), is chemically serelin in vivo, Pharm. Res. 13 (1996) 1668–1672.
modified to have a terminal half-life of nearly a week [88,89]. This [9] A. Mayavanshi, S. Gajjar, Floating drug delivery systems to increase gastric reten-
tion of drugs: a review, Res. J. Pharm. Technol. 1 (2008) 345–348.
makes the steady-state plasma concentration less sensitive to changes in [10] X. Ren, X. Mao, L. Cao, K. Xue, L. Si, J. Qiu, A.D. Schimmer, G. Li, Nonionic sur-
day-to-day variability in intestinal absorption. However, it does not factants are strong inhibitors of cytochrome P450 3A biotransformation activity in
resolve the issue with inter- and intraindividual variability, as low, or vitro and in vivo, Eur. J. Pharm. Sci. 36 (2009) 401–411.
[11] T.R. Buggins, P.A. Dickinson, G. Taylor, The effects of pharmaceutical excipients on
high, absorption of semaglutide several days in a row will result in a drug disposition, Adv. Drug Deliv. Rev. 59 (2007) 1482–1503.
new steady-state plasma concentration. For any such drug design to be [12] G.L. Amidon, H. Lennernas, V.P. Shah, J.R. Crison, A theoretical basis for a bio-
regulatory acceptable, it must be shown that these extremes in fabs do pharmaceutic drug classification: the correlation of in vitro drug product dissolu-
tion and in vivo bioavailability, Pharm. Res. 12 (1995) 413–420.
not affect the clinical efficacy or safety of the drug. It is also important
[13] P.J. Sinko, G.D. Leesman, G.L. Amidon, Predicting fraction dose absorbed in hu-
to consider that the absorption modifying drug delivery system do not mans using a macroscopic mass balance approach, Pharm. Res. 8 (1991) 979–988.
affect the absorption of concomitant administered drugs. [14] H. Lennernäs, Ö. Ahrenstedt, R. Hällgren, L. Knutson, M. Ryde, L.K. Paalzow,
Regional jejunal perfusion, a new in vivo approach to study oral drug absorption in
man, Pharm. Res. 9 (1992) 1243–1251.
6. Concluding remarks [15] D. Dahlgren, C. Roos, A. Lundqvist, B. Abrahamsson, C. Tannergren,
P.M. Hellström, E. Sjögren, H. Lennernäs, Regional intestinal permeability of three
model drugs in human, Mol. Pharm. 13 (2016) 3013–3021.
AMEs incorporated into oral drug products may increase the rate [16] D. Dahlgren, C. Roos, P. Johansson, A. Lundqvist, C. Tannergren, B. Abrahamsson,
and extent of intestinal drug absorption by increasing the membrane E. Sjögren, H. Lennernäs, Regional intestinal permeability in dogs: biopharmaceu-
tical aspects for development of oral modified-release dosage forms, Mol Pharm. 13
permeability of low intestinal permeability drugs (traditional low mo- (2016) 3022–3033.
lecular mass drugs and new modalities)API. These excipients are [17] K. Sugano, M. Kansy, P. Artursson, A. Avdeef, S. Bendels, L. Di, G.F. Ecker, B. Faller,
therefore interesting from both regulatory and drug delivery perspec- H. Fischer, G. Gerebtzoff, H. Lennernäs, F. Senner, Coexistence of passive and
carrier-mediated processes in drug transport, Nat. Rev. Drug. Discov. 9 (2010)
tives, as they may affect bioequivalence. They may also be used in- 597–614.
tentionally to increase drug absorption of low-permeability compounds [18] D. Dahlgren, C. Roos, E. Sjögren, H. Lennernäs, Direct in vivo human intestinal
from different classes (traditional small drugs and new modalities). permeability (Peff) determined with different clinical perfusion and intubation
methods, J. Pharm. Sci. 104 (2014) 2702–2726.
Currently, there are controversies regarding how to investigate and
[19] A.L. Ungell, S. Nylander, S. Bergstrand, Å. Sjöberg, H. Lennernäs, Membrane
characterize the biopharmaceutical properties of AMEs for affecting in transport of drugs in different regions of the intestinal tract of the rat, J. Pharm. Sci.
vivo intestinal drug absorption. These issues must be resolved to fa- 87 (1998) 360–366.
[20] H. Lennernas, Intestinal permeability and its relevance for absorption and elim-
cilitate the drug approval process of test products (e.g., generics), and
ination, Xenobiotica 37 (2007) 1015–1051.
to enable oral administration of new modalities by using AMEs. This [21] S. Winiwarter, N.M. Bonham, F. Ax, A. Hallberg, Lennernäs H, Karlén A. Correlation
requires that the in vivo processes influencing the effect of an AME be of human jejunal permeability (in vivo) of drugs with experimentally and theore-
better understood, and that mechanistic investigations of AMEs in tically derived parameters. a multivariate data analysis approach, J. Med. Chem. 41
(1998) 4939–4949.
preclinical models consider these processes. Key factors to focus on are [22] K. Sugano, Aqueous boundary layers related to oral absorption of a drug: from
gastrointestinal mucosal response rate of an AME, intestinal transit, dissolution of a drug to carrier mediated transport and intestinal wall metabolism,
luminal AME effects, intestinal epithelial repair, and variability in AME Mol. Pharm. 7 (2010) 1362–1373.
[23] C. Atuma, V. Strugala, A. Allen, L. Holm, The adherent gastrointestinal mucus gel
effects. Any preclinical investigations of AMEs that fail to consider these layer: thickness and physical state in vivo, American J. Physiol. Gastrointestinal
processes will ultimately be of limited clinical value and add little to Liver Physiol. 280 (2001) G922–G929.
our understanding of how excipients affect drug absorption. [24] P. Lundquist, P. Artursson, Oral absorption of peptides and nanoparticles across the
human intestine: Opportunities, limitations and studies in human tissues, Adv. Drug
Deliv. Rev. 106 (2016) 256–276.
[25] C. Roos, D. Dahlgren, C. Tannergren, B. Abrahamsson, E. Sjögren, H. Lennernas,
Acknowledgement
Regional intestinal permeability in rats: a comparison of methods, Mol Pharm. 14
(2017) 4252–4261.
This work received support from the Innovative Medicines Initiative [26] A.W. Larhed, P. Artursson, J. Gråsjö, E. Björk, Diffusion of drugs in native and
purified gastrointestinal mucus, J. Pharm. Sci. 86 (1997) 660–665.
Joint Undertaking (http://www.imi.europa.eu) under grant agreement
[27] A. Bernkop-Schnürch, R. Fragner, Investigations into the diffusion behaviour of
number 115369, the resources of which are composed of financial polypeptides in native intestinal mucus with regard to their peroral administration,
contributions from the European Union’s Seventh Framework Pharm. Pharmacol. Commun. 2 (1996) 361–363.
Programme (FP7/2007-013) and EFPIA companies. We also thank the [28] C. Roos, D. Dahlgren, E. Sjögren, M. Sjöblom, M. Hedeland, H. Lennernäs, Jejunal
absorption of aprepitant from nanosuspensions: Role of particle size, prandial state
reviewer of this manuscript for their important input.

418
D. Dahlgren, et al. European Journal of Pharmaceutics and Biopharmaceutics 142 (2019) 411–420

and mucus layer, Eur J Pharm Biopharm. 132 (2018) 222–230. Eur. J. Pharm. Sci. 57 (2014) 99–151.
[29] E.M. Danielsen, G.H. Hansen, Intestinal surfactant permeation enhancers and their [58] Å. Sjöberg, M. Lutz, C. Tannergren, C. Wingolf, A. Borde, A.-L. Ungell,
interaction with enterocyte cell membranes in a mucosal explant system, Tissue Comprehensive study on regional human intestinal permeability and prediction of
Barriers 5 (2017) e1361900. fraction absorbed of drugs using the Ussing chamber technique, Eur. J. Pharm. Sci
[30] D. Lichtenberg, R.J. Robson, E.A. Dennis, Solubilization of phospholipids by de- (2012).
tergents structural and kinetic aspects, Biochimica et Biophysica Acta (BBA)-Rev. [59] P. Artursson, Epithelial transport of drugs in cell culture. I: A model for studying the
Biomembr. 737 (1983) 285–304. passive diffusion of drugs over intestinal absorbtive (Caco-2) cells, J. Pharm. Sci. 79
[31] S. Maher, R.J. Mrsny, D.J. Brayden, Intestinal permeation enhancers for oral pep- (1990) 476–482.
tide delivery, Adv. Drug Deliv. Rev. 106 (2016) 277–319. [60] E. Bergman, A. Lundahl, P. Fridblom, M. Hedeland, U. Bondesson, L. Knutson,
[32] H. Lennernäs, C.-G. Regårdh, Evidence for an interaction between the β-blocker H. Lennernäs, Enterohepatic disposition of rosuvastatin in pigs and the impact of
pafenolol and bile salts in the intestinal lumen of the rat leading to dose-dependent concomitant dosing with cyclosporine and gemfibrozil, Drug Metabolism Disposit.
oral absorption and double peaks in the plasma concentration–time profile, Pharm. 37 (2009) 2349–2358.
Res. 10 (1993) 879–883. [61] E. Lipka, H. Spahn-Langguth, E. Mutschler, G. Amidon, In vivo non-linear intestinal
[33] A.A.A. Al-Ali, J.R.C. Quach, C. Bundgaard, B. Steffansen, R. Holm, C.U. Nielsen, permeability of celiprolol and propranolol in conscious dogs: evidence for intestinal
Polysorbate 20 alters the oral bioavailability of etoposide in wild type and mdr1a secretion, Eur. J. Pharm. Sci. 6 (1998) 75–81.
deficient Sprague-Dawley rats, Int J Pharm. 543 (2018) 352–360. [62] D. Dahlgren, C. Roos, P. Johansson, C. Tannergren, A. Lundqvist, P. Langguth,
[34] T.Y. Ma, P. Nighot, R. Al-Sadi, Tight junctions and the intestinal barrier, Physiology M. Sjöblom, E. Sjögren, H. Lennernas, The effects of three absorption-modifying
of the Gastrointestinal Tract, Sixth ed., Elsevier, 2018, pp. 587–639. critical excipients on the in vivo intestinal absorption of six model compounds in
[35] S.M. Krug, M. Amasheh, I. Dittmann, I. Christoffel, M. Fromm, S. Amasheh, Sodium rats and dogs, Int. J. Pharm. 547 (2018) 158–168.
caprate as an enhancer of macromolecule permeation across tricellular tight junc- [63] S. Maher, J. Heade, F. McCartney, S. Waters, S.B. Bleiel, D.J. Brayden, Effects of
tions of intestinal cells, Biomaterials 34 (2013) 275–282. surfactant-based permeation enhancers on mannitol permeability, histology, and
[36] S. Maher, T.W. Leonard, J. Jacobsen, D.J. Brayden, Safety and efficacy of sodium electrogenic ion transport responses in excised rat colonic mucosae, Int. J. Pharm.
caprate in promoting oral drug absorption: from in vitro to the clinic, Adv. Drug 539 (2018) 11–22.
Deliv. Rev. 61 (2009) 1427–1449. [64] E.K. Anderberg, T. Lindmark, P. Artursson, Sodium caprate elicits dilatations in
[37] O. Nylander, L. Pihl, M. Perry, Hypotonicity-induced increases in duodenal mucosal human intestinal tight junctions and enhances drug absorption by the paracellular
permeability facilitates adjustment of luminal osmolality, American J. Physiol. route, Pharm. Res. 10 (1993) 857–864.
Gastrointest Liver Physiol. 285 (2003) G360–G370. [65] N.G. Schipper, S. Olsson, J.A. Hoogstraate, K.M. Vårum, P. Artursson, Chitosans as
[38] J.R. Turner, Show me the pathway!: Regulation of paracellular permeability by Na absorption enhancers for poorly absorbable drugs 2: mechanism of absorption en-
+-glucose cotransport, Adv. Drug Deliv. Rev. 41 (2000) 265–281. hancement, Pharm. Res. 14 (1997) 923–929.
[39] H. Lennernäs, Does fluid flow across the intestinal mucosa affect quantitative oral [66] E.S. Swenson, W.B. Milisen, W. Curatolo, Intestinal permeability enhancement:
drug absorption? Is it time for a reevaluation? Pharm. Res. 12 (1995) 1573–1582. efficacy, acute local toxicity, and reversibility, Pharm. Res. 11 (1994) 1132–1142.
[40] D. Dahlgren, C. Roos, A. Lundqvist, C. Tannergren, M. Sjöblom, E. Sjögren, [67] N.G. Schipper, K.M. Vårum, P. Stenberg, G. Ocklind, H. Lennernäs, P. Artursson,
H. Lennernas, Effect of absorption-modifying excipients, hypotonicity, and enteric Chitosans as absorption enhancers of poorly absorbable drugs: 3: Influence of
neural activity in an in vivo model for small intestinal transport, Int J Pharm. 549 mucus on absorption enhancement, Eur. J. Pharm. Sci. 8 (1999) 335–343.
(2018) 239–248. [68] D. Dahlgren, C. Roos, A. Lundqvist, C. Tannergren, M. Sjöblom, E. Sjögren,
[41] S. Tavelin, K. Hashimoto, J. Malkinson, L. Lazorova, I. Toth, P. Artursson, A new H. Lennernäs, Time-dependent effects on small intestinal transport by absorption-
principle for tight junction modulation based on occludin peptides, Mol Pharmacol. modifying excipients, Eur J Pharm Biopharm (2018).
64 (2003) 1530–1540. [69] D.M. Mudie, K. Murray, C.L. Hoad, S.E. Pritchard, M.C. Garnett, G.L. Amidon,
[42] M. Kondoh, A. Masuyama, A. Takahashi, N. Asano, H. Mizuguchi, N. Koizumi, P.A. Gowland, R.C. Spiller, G.E. Amidon, L. Marciani, Quantification of gastro-
M. Fujii, T. Hayakawa, Y. Horiguchi, Y. Watanbe, A novel strategy for the en- intestinal liquid volumes and distribution following a 240 mL dose of water in the
hancement of drug absorption using a claudin modulator, Mol Pharmacol. 67 fasted state, Mol. Pharm. 11 (2014) 3039–3047.
(2005) 749–756. [70] C. Schiller, C.P. Fröhlich, T. Giessmann, W. Siegmund, H. Mönnikes, N. Hosten,
[43] FDA, Bioeqvivalence and Bioavailability Guideline. < https://www.fda.gov/ W. Weitschies, Intestinal fluid volumes and transit of dosage forms as assessed by
downloads/drugs/guidancecomplianceregulatoryinformation/guidances/ magnetic resonance imaging, Aliment Pharmacol Ther. 22 (2005) 971–979.
ucm389370.pdf > . [71] S.B. Petersen, G. Nolan, S. Maher, U.L. Rahbek, M. Guldbrandt, D.J. Brayden,
[44] EMA, Bioeqvivalence and Bioavailability Guideline. < http://www.ema.europa.eu/ Evaluation of alkylmaltosides as intestinal permeation enhancers: comparison be-
docs/en_GB/document_library/Scientific_guideline/2009/09/WC500003519. tween rat intestinal mucosal sheets and Caco-2 monolayers, Eur. J. Pharm. Sci. 47
pdf > . (2012) 701–712.
[45] FDA, BCS Guidance for Industry, 12 August 2014. < www.fda.gov/downloads/ [72] J.R. Turner, Intestinal mucosal barrier function in health and disease, Nature Rev.
Drugs/Guidances/ucm070246.pdf > . Immunol. 9 (2009) 799.
[46] A. García-Arieta, Interactions between active pharmaceutical ingredients and ex- [73] M. Baluom, M. Friedman, A. Rubinstein, The importance of intestinal residence
cipients affecting bioavailability: impact on bioequivalence, Eur. J. Pharm. Sci. 65 time of absorption enhancer on drug absorption and implication on formulative
(2014) 89–97. considerations, Int. J. Pharm. 176 (1998) 21–30.
[47] A. García-Arieta, J. Gordon, H. Potthast, On the Effect of Common Excipients on the [74] S. Maher, D.J. Brayden, L. Casettari, L. Illum, Application of permeation enhancers
Oral Absorption of Class 3 Drugs, J. Pharm. Sci. 105 (2016) 1353–1354. in oral delivery of macromolecules: an update, Pharmaceutics 11 (2019) 41.
[48] P. Tyagi, S. Pechenov, J.A. Subramony, Oral peptide delivery: Translational chal- [75] M. Thongngam, D.J. McClements, Influence of pH, ionic strength, and temperature
lenges due to physiological effects, J Control Release. (2018). on self-association and interactions of sodium dodecyl sulfate in the absence and
[49] K. Whitehead, N. Karr, S. Mitragotri, Safe and effective permeation enhancers for presence of chitosan, Langmuir 21 (2005) 79–86.
oral drug delivery, Pharm. Res. 25 (2008) 1782–1788. [76] C. Mihali, G. Oprea, E. Cical, Determination of critical micelar concentration of
[50] A. Muheem, F. Shakeel, M.A. Jahangir, M. Anwar, N. Mallick, G.K. Jain, anionic surfactants using surfactants-sensible electrodes, Measurement. 1 (2008) 3.
M.H. Warsi, F.J. Ahmad, A review on the strategies for oral delivery of proteins and [77] K.P. Kwang-hee, Acid-base equilibria and related properites of chitosan, Bulletin
peptides and their clinical perspectives, Saudi Pharmaceut. J. 24 (2016) 413–428. Korean Chem. Soc. 4 (1983) 68–72.
[51] S. Rehmani, J.E. Dixon, Oral delivery of anti-diabetes therapeutics using cell pe- [78] K. Gradauer, A. Nishiumi, K. Unrinin, H. Higashino, M. Kataoka, B.L. Pedersen,
netrating and transcytosing peptide strategies, Peptides 100 (2018) 24–35. S.T. Buckley, S. Yamashita, Interaction with mixed micelles in the intestine at-
[52] D.L. Burcham, B.A. Aungst, M. Hussain, M.A. Gorko, C.Y. Quon, S.-M. Huang, The tenuates the permeation enhancing potential of alkyl-maltosides, Mol. Pharm. 12
effect of absorption enhancers on the oral absorption of the GP IIB/IIIA receptor (2015) 2245–2253.
antagonist, DMP 728, in rats and dogs, Pharm. Res. 12 (1995) 2065–2070. [79] A.-C. Luissint, C.A. Parkos, A. Nusrat, Inflammation and the intestinal barrier:
[53] T. Nishihata, J.H. Rytting, T. Higuchi, L. Caldwell, Enhanced rectal absorption of Leukocyte–epithelial cell interactions, cell junction remodeling, and mucosal repair,
insulin and heparin in rats in the presence of non-surfactant adjuvants, J. Pharm. Gastroenterology 151 (2016) 616–632.
Pharmacol. 33 (1981) 334–335. [80] J.B. Furness, The enteric nervous system and neurogastroenterology, Nature Rev.
[54] T.A. Aguirre, D. Teijeiro-Osorio, M. Rosa, I. Coulter, M. Alonso, D. Brayden, Current Gastroenterol. Hepatol. 9 (2012) 286.
status of selected oral peptide technologies in advanced preclinical development [81] A.M. Marchiando, W.V. Graham, J.R. Turner, Epithelial barriers in homeostasis and
and in clinical trials, Adv. Drug Deliv. Rev. 106 (2016) 223–241. disease, Annual Rev. Patholog. Mech. Dis. 5 (2010) 119–144.
[55] F. McCartney, J.P. Gleeson, D.J. Brayden, Safety concerns over the use of intestinal [82] R. Moore, S. Carlson, J.L. Madara, Villus contraction aids repair of intestinal epi-
permeation enhancers: a mini-review, Tissue Barriers 4 (2016) e1176822. thelium after injury. American J. Physiol. Gastrointest Liver Physiol. 257 (1989)
[56] E.S. Kostewicz, B. Abrahamsson, M. Brewster, J. Brouwers, J. Butler, S. Carlert, G274–G283.
P.A. Dickinson, J. Dressman, R. Holm, S. Klein, < i > In vitro < /i > models for [83] A.T. Blikslager, A.J. Moeser, J.L. Gookin, S.L. Jones, J. Odle, Restoration of barrier
the prediction of < i > in vivo < /i > performance of oral dosage forms, Eur. J. function in injured intestinal mucosa, Physiol Rev. 87 (2007) 545–564.
Pharm. Sci. (2013). [84] N. Barker, Adult intestinal stem cells: critical drivers of epithelial homeostasis and
[57] E. Sjögren, B. Abrahamsson, P. Augustijns, D. Becker, M.B. Bolger, M. Brewster, regeneration, Nature Rev. Mol. Cell Biol. 15 (2014) 19.
J. Brouwers, T. Flanagan, M. Harwood, C. Heinen, R. Holm, H. Juretschke, [85] A.M. Marchiando, L. Shen, W.V. Graham, K.L. Edelblum, C.A. Duckworth, Y. Guan,
M. Kubbinga, A. Lindahl, V. Lukacova, U. Münster, S. Neuhoff, M. Nguyen, A. Peer, M.H. Montrose, J.R. Turner, A.J. Watson, The epithelial barrier is maintained by in
C. Reppas, A. Hodjegan, C. Tannergren, W. Weitschies, C. Wilson, P. Zane, vivo tight junction expansion during pathologic intestinal epithelial shedding,
H. Lennernäs, P. Langguth, In vivo methods for drug absorption–comparative Gastroenterology 140 (1208–1218) (2011) e1202.
physiologies, model selection, correlations with in vitro methods (IVIVC), and ap- [86] L.G. Tillman, R.S. Geary, G.E. Hardee, Oral delivery of antisense oligonucleotides in
plications for formulation/API/excipient characterization including food effects, man, J. Pharm. Sci. 97 (2008) 225–236.

419
D. Dahlgren, et al. European Journal of Pharmaceutics and Biopharmaceutics 142 (2019) 411–420

[87] K.V. Juul, D.G. Bichet, J.P. Nørgaard, Desmopressin duration of antidiuretic action Pharmaceutical press London, 2006.
in patients with central diabetes insipidus, Endocrine 40 (2011) 67–74. [91] I. Legen, M. Kračun, M. Salobir, J. Kerč, The evaluation of some pharmaceutically
[88] T.C. Marbury, A. Flint, J.B. Jacobsen, J.D. Karsbøl, K. Lasseter, Pharmacokinetics acceptable excipients as permeation enhancers for amoxicillin, Int. J. Pharm. 308
and tolerability of a single dose of semaglutide, a human glucagon-like peptide-1 (2006) 84–89.
analog, in subjects with and without renal impairment, Clin. Pharmacokinet. 56 [92] E.K. Anderberg, P. Artursson, Epithelial transport of drugs in cell culture. VIII: ef-
(2017) 1381–1390. fects of sodium dodecyl sulfate on cell membrane and tight junction permeability in
[89] S.T. Buckley, T.A. Bækdal, A. Vegge, S.J. Maarbjerg, C. Pyke, J. Ahnfelt-Rønne, human intestinal epithelial (Caco-2) cells, J. Pharm. Sci. 82 (1993) 392–398.
K.G. Madsen, S.G. Schéele, T. Alanentalo, R.K. Kirk, Transcellular stomach ab- [93] Y. Narkar, R. Burnette, R. Bleher, R. Albrecht, A. Kandela, J.R. Robinson, Evaluation
sorption of a derivatized glucagon-like peptide-1 receptor agonist, Sci Transl Med. of mucosal damage and recovery in the gastrointestinal tract of rats by a penetra-
10 (2018) eaar7047. tion enhancer, Pharm. Res. 25 (2008) 25–38.
[90] R.C. Rowe, P.J. Sheskey, S.C. Owen, Handbook of pharmaceutical excipients,

420