Pharmaceutical Research, Vol. 16. No. 7, 1999
`
`Research Paper
`
`Regional Differences in Quinine
`Absorption from the Undisturbed
`Human Colon Assessed Using a Timed
`Release Delivery System
`
`distal sites (1,2), but in normal clinical use colonic contents
`may have an important modifying effect. Wc therefore aimed to
`study absorption by measuring the permeation of probe marker
`molecules from different regions of the distal gut, using a
`delivery system, the Pulsincap™ (3), which allowed selective
`delivery to various regions of the undisturbed gut.
`
`John M. Hebden,' Clivc G. Wilson,2
`Robin C. Spiller,1'4 Peter J. Gilchrist,2
`Elaine Blackshaw,3 Malcolm E. Frier,3 and
`Alan C. Perkins3
`
`Received March II. 1999; accepted April 12, 1999
`
`Purpose. To investigate the regional absorption characteristics of'thc
`distal gut using two markers of permeability, quinine (a transcellular
`probe) and "CrEDTA (a paracellular probe).
`Methods. The permeability markers were delivered to the undisturbed
`gastrointestinal tract in 39 healthy volunteers using an oral timed-
`release delivery vehicle which allowed pulsed release within a particu(cid:173)
`lar site of the gut. Site of release was identified using gamma scintigra(cid:173)
`phy. Absorption of quinine and "CrEDTA was assessed by measuring
`the percent excretion in the urine using HPLC and gamma counting
`respectively. Serial plasma samples allowed time-concentration curves
`for quinine to be plotted.
`Results. There was a significant trend for diminished absorption with
`more distal delivery of the transcellular probe, quinine, which was:
`6.26 ± 0.87% (small intestine, n = 10); 4.65 ± 0.93% (ascending
`colon, n = 16); and 2.59 ± 0.52% (transverse colon, n = 10) of the
`ingested dose excreted respectively (p < 0.001). No such gradient was
`seen with the paracellular marker. "CrEDTA.
`Conclusions. These results suggest that delayed release formuations
`shoud aim for release in the distal small bowel and proximal colon if
`absorption is to be miximised. Absorption by the transcellular toutc
`diminishes in the more distal colon, a fact which has implications for
`delayed or sustained release formulations.
`
`KEY WORDS: transcellular; paracellular:
`trointestinal.
`
`absorption: gas(cid:173)
`
`INTRODUCTION
`
`It has long been the aim of the pharmaceutical industry
`to achieve delayed absorption of drug to treat nocturnal and
`early morning exacerbations of disease, e.g., asthma, rheuma(cid:173)
`toid disease, and cardiovascular disease. Such formulations
`taken at bedtime are inevitably situated in the distal small bowel
`or proxi mal colon by the early hours of the momi ng. Knowledge
`of the effect of site of release on absorption characteristics in
`the undisturbed colon is therefore highly relevant in designing
`such preparations.
`Previous studies using a cleansed colon have suggested
`little intrinsic difference in absorption between proximal and
`
`1 Division of Gastroenterology. University Hospital Nottingham, Not(cid:173)
`tingham, UK.
`1 Department of Pharmaceutical Sciences. University of Strathclydc,
`Glasgow, UK.
`3 Department of Medical Physics, University Hospital Nottingham,
`Nottingham, UK.
`4 To whom correspondence should be addressed.
`
`METHODS
`
`Subjects
`
`Thirty-nine healthy volunteers (23 male; 16 females), age
`range 20-40, were recruited into the study. All subjects were
`free from gastrointestinal disease, and were not taking any
`laxatives or drugs known to affect gut motility. All were asked
`to refrain from excess alcohol, curries, and aspirin or non(cid:173)
`steroidal antiinflammatory drugs during the course of the study.
`Females were required to have a negative pregnancy test on
`the morning of the study day. The study was approved by
`Nottingham University Ethical Committee and the Association
`of Radioactive Substances Advisory Committee at the Depart(cid:173)
`ment of Health.
`
`Study Protocol
`
`Phase I
`
`In the initial phase of the study, 11 healthy volunteers (6
`males; 5 females; age range 20-29) were recruited into a two
`way crossover study in which they were dosed with either a 5
`hour (part A) or 15 hour (part B) release delivery capsule
`in an attempt to selectively target permeability probes to the
`proximal colon or distal colon respectively. There was a 2 week
`washout period, and the volunteers were required to adhere to
`a 20 gm fibre diet for 2 days before each of the study days. In
`part A, subjects ingested the 5 hour release delivery system at
`0800 h following an overnight fast, and then consumed a stan(cid:173)
`dard 200 kcal breakfast of toast, butter and jam once the capsule
`was seen to empty from the stomach. A standard 600 kcal lunch
`and 1000 kcal meal were provided at 1300 h and 1800 h
`respectively. Blood samples were taken prior to, and at 30
`minute intervals after the predicted release time (1300 h), and
`urine was collected in the 3 hour period leading up to the
`expected release time, and for the 0-20 hour period following
`the expected release time. In part B, subjects were dosed at
`2200 h on the evening prior to the study day, so that the predicted
`time of release would be the same as in part A (i.e., 1300 hrs).
`An identical meal pattern, blood and urine sampling protocol
`was followed.
`
`Phase 2
`
`This comprised 12 healthy volunteers who were dosed on
`a separate occasion with a 5 hr release Pulsincap™, and 16
`healthy volunteers who were dosed with a 6 hr release Pulsin(cid:173)
`cap™ . These 28 healthy volunteers(I7M; 11F, age range 20-40)
`followed an identical protocol to phase I part A.
`
`Pulsincap™ Delivery Capsule
`
`The Pulsincap™ delivery capsule consists of a water insol(cid:173)
`uble body with a hydrogel plug inserted at its open end, which
`
`1087
`
`I)724.874l/<W/07(IO-IOK7SIA.(KI/ll » I W" Plenum PuMahin|;C<»p.>rali<«i
`
`Cosmo Ex. 2032-p. 1
`Argentum v Cosmo
`IPR2018-00080
`
`

`

`wmmSz
`
`1088
`
`Hebden et al
`
`swells on contact with water and gradually propels itself from
`the body, thereby allowing the contents to be released. The
`time of release of the plug is dependent on its dimensions.
`Each delivery system in this study contained SO mg of quinine
`hydrochloride. l.SMBq CrEDTA dried onto sucrose, I MBq
`Indium-labelled amberlite resin as an imaging agent, and an
`excipient. A 50 mg dose of quinine was chosen as this was
`sufficient to be easily detected given the highly sensitive HPLC
`method employed, and low enough as to be unlikely to cause
`any significant side-effects. The usual clinical dose for treatment
`of muscular cramps is 300 mg once daily and for the treatment
`of malaria 600 mg every 8 hours, but these doses may be
`associated with significant side-effects such as nausea, tinnilus,
`and cardiotoxicity.
`
`Scintigraphic Imaging
`
`Scintigraphic images were obtained using a GEC Maxi-
`camcra set with a 20% window for simultaneous acquisition
`of the 140-kcV radiation peak of Tc and the 247-keV radiation
`peak of In. Anterior and posterior images of 30 seconds duration
`were taken every 30 minutes after dosing. All images were
`stored for later analysis using a dedicated computer. Alignment
`of serial images was facilitated by taping small radiolabelled
`markers (0.2 MBq of Tc) anteriorly and posteriorly over the
`hepatic area. The position within the gastrointestinal tract, and
`the time of release of markers from the delivery system was
`determined by visual inspection of the serial images. Release
`could easily be inferred from the serial images as the area of
`the hot spot increased and its intensity concomitantly decreased.
`
`Quantification of Marker Probes
`
`The analysis of quinine was carried out using an estab(cid:173)
`lished revcrscd-phase HPLC method (4). The HPLC apparatus
`(HP 1050) was fitted with an auto sampler and a fluorescence
`detector. For the assay of quinine optimum settings were: excita(cid:173)
`tion = 350 nm, and emission wavelength = 450 tim. The
`mobile phase consisted of an acctonitrile-aqueous phosphate
`buffer (10 mM) mixture (70/30 v:v), containing 3 mM tetra-
`butylammonium bromide (TBA) and 20 mM sodium dodecyl
`sulphate (SDS). pH 2.5. The stationary phase consisted of a
`Hypersil C-18 column (5 mm) 150 X 3.2 mm protected by a
`guard column 30 X 3.2 mm (Phenomenex).
`A protein precipitation technique was employed in the
`preparation of the urine samples. To 200 u.1 of sample, methanol
`(400 u.1) was added, the mixture vortcxed and then centrifuged
`at 1800 g for 15 minutes to remove the precipitate. The superna(cid:173)
`tant was transferred to a siliconiscd glass vial prior to injection
`from autosamplcr. Sample injection volume was 10 u.1 and
`flow rate was 0.5 p-l/min. Chromatographic separations were
`performed at room temperature. The inter- and intra- assay
`coefficients of variation were found to be less lltan 4%. The
`lowest limit of detection for quinine in plasma was 3.5 ng/ml.
`A 10 ml sample of urine was counted in a gamma-counter
`(LKB Wallac 1280) for determination of Cr-EDTA content.
`Reference standard solutions of Cr-EDTA were prepared at the
`beginning of the trials for the calculation of decay corrections.
`After correcting for the total volume of urine in each time
`interval, the results were expressed as the % of administered
`dose excreted.
`
`Table 1. Number. Release Time, and Site of Release of the 5 and 6
`Hour Pulsincap Delivery Systems
`
`Number
`Release
`Site
`
`5hr Pulsincap
`
`23
`5.5 ± 0.2 h
`SI 5
`AC 11
`TC4
`
`6hr Pulsincap
`
`16
`7.3 ± 0.3 h
`SI 5
`AC 5
`TC6
`
`Note: SI = small intestine; AC = ascending colon; TC = transverse
`colon.
`
`Statistics
`
`Permeation of the marker probes from the different sites
`was analysed using one way analysis of variance (ANOVA).
`and Jonckhccrc's test for ordered alternatives.
`
`RESULTS
`
`The site and times of deli very arc summarised in Table I (5
`and 6 hr delivery systems) and Table 2 (15 hr delivery systems).
`In 20 out of the 23 subjects ingesting the 5 hr release
`system, scintigraphic release occurred in the distal small bowel
`or colon. In one subject, the delivery system lay in the rectosig(cid:173)
`moid colon and did not appear to release, and in two subjects
`the delivery system was retained in the stomach and therefore
`released in this region. Release occurred successfully in the
`distal small bowel or colon in all of the 16 subjects ingesting
`the 6 hr release delivery system.
`In the II studies involving the 15 hr release delivery
`system, scintigraphic release was only observed in 2 subjects
`(in the ascending and transverse colons at 19.4 and 20.5 hours
`after ingestion). Al the predicted release time, 9 of the 11
`delivery systems resided in the colon (2 had been excreted).
`The majority of the systems, however, were located in the
`proximal colon (rather than the intended distal colon).
`Since release of probe marker was by and large unsuccess(cid:173)
`ful using the 15 hr release delivery systems, we have restricted
`our analysis to the 5 and 6 hr release data. We were thus able
`to compare delivery of marker probes to the small intestine
`(n = 10), ascending colon (n = 16), and transverse colon (n
`= 10).
`
`Table 2. Number, and Site at Expected Release Time of the IS Hour
`Pulsincap Delivery Systems
`
`IShr Pulsincap
`
`Number
`Release
`Site
`
`11
`n/a
`SI
`AC
`TC
`DC
`exer
`
`0
`6
`1
`2
`2
`
`Note: Release time is not applicable (n/a) as only 2 were evidenced
`to release scintigraphically. (SI = small intestine; AC = ascending
`colon: TC = transverse colon).
`
`fi
`
`:i
`
`Cosmo Ex. 2032-p. 2
`Argentum v Cosmo
`IPR2018-00080
`
`

`

`Regional Differences in Colonic Absorption of Quinine
`
`Scintigraphic release was seen a little earlier in those indi(cid:173)
`viduals with delivery to the small intestine compared to the
`ascending and transverse colons (small intestine 5.7 i 0.4 hrs;
`ascending colon 6.0 ± 0.2 hrs; transverse colon 7.3 ± 0.5 hrs),
`resulting in a slightly greater time for permeation to occur.
`Excretion of all or a proportion of the released isotope
`had occurred in half of the subjects by the end of the 0-20 hr
`urine collection period, with approximately equal proportions
`in each set of individuals grouped by initial release site (small
`intestine 5/10; ascending colon 7/16; transverse colon 6/10).
`
`Permeation of Marker Probes
`
`Quinine absorption diminished with progressively more
`distal release of capsular contents, as assessed by the 0-20 hr
`urine collection: small intestinal release, 6.26 ± 0.87% of the
`ingested dose excreted; ascending colon release, 4.65 ± 0.93%
`of the ingested dose excreted; and transverse colon release,
`2.59 ± 0.52% of the ingested dose excreted (Fig. I). One way
`analysis of variance showed a difference between these sites,
`with Jonckheerc's test for ordered alternatives showing a signifi(cid:173)
`cant trend for decreased absorption with progressively more
`distal delivery (p < 0.001). Permeation of EDTA, however,
`showed no such gradient: small intestinal release, 1.43 ± 0.43%
`of the ingested dose excreted; ascending colon release, 0.60 ±
`0.18% of the ingested dose excreted; and transverse colon,
`1.17 ± 0.64% of the ingested dose excreted (Fig. 2).
`Although time concentration profiles of plasma quinine
`seemed to show a trend for a more rapid upstroke with more
`proximal delivery (median [range] times to peak concentration:
`small intestine 1.9 10. 5-3.3] hours, ascending colon 2.3 [0.6-
`17] hours, transverse colon 2.5 [1.0-8.0| hrs), this was not
`statistically significant. There was however a significant trend
`for greater peak concentrations with more proximal delivery
`(median [range) concentration: small intestine0.6610.13-1.06],
`ascending colon 0.38 [0.02-0.61 ], transverse colon 0.1810.08-
`0.60|, p < 0.005, Jonckhecre's test for ordered alternatives, sec
`Fig. 3).
`
`DISCUSSION
`
`The Pulsincap™ delivery system allowed us the opportu(cid:173)
`nity to examine the absorption characteristics of the colon in
`
`%EDTA
`recovered
`in urine
`
`7"
`
`6"
`
`0
`
`1089
`
`O
`
`TC
`AC
`SI
`Fig. 2. Percent of "CrEDTA appearing in the urine following release
`in the distal small bowel (SI), ascending colon (AC), and transverse
`colon (TC).
`
`a non-invasive physiological manner, by delivering a bolus
`quantity of marker probe to a selected region. A previous study
`had suggested that a 5 hour release time would allow successful
`targeting of the proximal colon (5).
`The present study shows that precise targeting was not
`possible due to interindividual differences in gastrointestinal
`transit rates. Indeed, despite the reasonably precise release times
`obtained from the 5 and 6 hour Pulsincaps™, delivery occurred
`in the small inlcstinc in almost l/3rd of cases, and as far distally
`as the transverse colon in another l/3rd. Nevertheless, selective
`release in the ileocaecal region (distal small bowel or ascending
`colon) occurred in 26/39 (67%) individuals (using timed 5 and
`6 hour delivery systems. This compares with a figure of 86%
`in a study employing a mcthylacrylale ('Eudragit') coating to
`protect gelatin capsules against disintegration within the stom(cid:173)
`ach in 14 subjects (6).
`Selective delivery to the distal colon represented a more
`difficult challenge. The descending and sigmoid colon have
`thick muscular walls (7) designed primarily for propulsion, and
`recent stud ics have suggested that this region serves as a conduit,
`in contrast to the storage role of the more proximal colon (6,8).
`
`15
`
`10-
`
`5-
`
`% Quinine
`recovered
`in urine
`
`o
`
`O
`
`O
`
`plasma
`quinine
`(Pg/ml)
`
`0
`
`8
`TC
`AC
`SI
`Fig. 1. Percent of quinine appearing in the urine following release in
`the distal small bowel (SI), ascending colon (AC), and transverse
`colon (TC).
`
`20
`10
`time (hours)
`FIR. 3. Graph showing the mean concentration of quinine in plasma
`versus time, following small intestinal release (circles), ascending colon
`release (triangles), and transverse colon release (squares). Standard
`error bars are shown.
`
`30
`
`Cosmo Ex. 2032-p. 3
`Argentum v Cosmo
`IPR2018-00080
`
`

`

`^^^p:f^|p^tPP^S%^'
`
`1090
`
`Hebden et aL
`
`Thus there is likely to be a much smaller 'window' of time
`available for release to result ineffective targeting of this region.
`Additionally, as residence time in the ascending colon has been
`shown to be approximately 12 hours>(9), a substantial delay in
`release lime would be required. In vitro work had shown the
`Pulsincap™ delivery system successfully releasing after a delay
`of up to 15 hours, and it was therefore this configuration which
`was used to effect selective delivery to the distal colon.
`Our attempts to target permeability probes to the distal
`colon using the 15 hour delivery system were however unsuc(cid:173)
`cessful on two counts. First and unexpectedly, the majority of
`the delivery systems were situated in the proximal colon at
`their predicted release time (15 hours), and no more further
`advanced when compared to the 5 hour systems viewed 6 hours
`after dosing (Fig. 4). This relative stagnation may in part be
`explained by the different dosing times in the two arms of the
`study. In the 15 hour arm, subjects were dosed at 2200 hrs so
`that the predicted release time (1300 hrs) would be identical
`
`(a)
`
`(b)
`
`to the 5 hour arm of the study, and therefore meal patterns
`would be the same. Thus in this arm of the study, subjects
`would have slept for approximately 8 of the 15 hours, and sleep
`has been shown to reduce colonic electrical and contractile
`activity (10-14). Delayed nocturnal gastric emptying (15) and
`reduced propagation velocity of the intestinal migrating motor
`complex (16) may also have been contributory, as supported
`by the finding that in 2 individuals the delivery system did
`not enter the colon until 12.5 and 13.5 hours after ingestion.
`Secondly, although reliable release from a 15 hour release Pul(cid:173)
`sincap™ had been demonstrated in simulated intestinal contents
`in vitro, there had been no previous in vivo experience with
`this configuration. In fact, scintigraphic release was only evident
`in 2 of the 11 volunteers in phase I of the study (although plasma
`samples showed evidence of release in a further 3 individuals).
`Targeted delivery to the distal colon using the 15 hour
`delivery capsule was unsuccessful for two reasons. Firstly, the
`capsules were no further advanced 15 hours after ingestion than
`the 5 hour capsules viewed 6 hours after dosing (Fig. 4). This
`was probably related to the period of sleep following dosing,
`which is known to reduce colonic electrical and contractile
`activity (10-14), delay gastric emptying (15), and reduce the
`propagation velocity of the intestinal migrating motor complex
`(16). Secondly, release was only successful in 5 individuals (2
`scintigraphically; 3 additionally on plasma measurements of
`quinine). Despite these shortcomings, we were still able to
`compare regional absorption characteristics of the distal gut as
`a result of the intcrindividual variations in transit of the 5 and
`6 hour delivery systems, which led to a spread of initial release
`sites. Our results show a clear trend for reduced absorption of
`quinine as one moves aborally from the distal small bowel to
`the transverse colon. A similar gradient however was not
`observed for the permeation of CrEDTA.
`Intestinal permeability tests investigate the unmediatcd
`diffusion across the intestinal epithelium of medium and large
`sized, inert, non-metabolised, water soluble molecules. There
`are two main pathways: transcellular and paracellular. The for(cid:173)
`mer are thought to be small aqueous 'pores' (<0.4-0.7 nm)
`of high incidence, whereas the latter are thought to be larger
`aqueous 'channels' (>6.5 nm) of low incidence allowing the
`permeation of larger molecules such as CrEDTA (17,18). Lipid
`soluble substances can diffuse directly through the cell surface
`membrane. Monosaccharides, such as L-rhamnose (molecular
`mass 164 Daltons; molecular diameter = 0.83 nm) and o-
`mannitol (molecular mass = 182 Daltons; molecular diameter =
`0.67 nm) are thought to permeate
`largely
`transccllularly,
`whereas disaccharides, such as lactulose (molecular mass = 342
`Daltons; molecular diameter = 0.95 nm) and the radioligand
`51 CrEDTA (molecular mass = 359 Daltons; molecular diame(cid:173)
`ter = 1.05 nm) arc thought to permeate paracellularly. The
`relative abundance of the small transcellular pores results ir.
`greater permeation of the monosaccharides (11-19% L-rham(cid:173)
`nose excreted 5 hrs after oral dosing) compared to the disaccha(cid:173)
`rides (0.3-0.4% lactulose excreted 5 hrs after oral dosing)
`(18,19).
`
`The permeability markers chosen in our study were quinine
`hydrochloride and Cr-51 EDTA. These materials were used
`because of ease of detection and lack of pharmaceutical action.
`Fig. 4. Postion of (a) 15 hour timed-release delivery systems within
`The absorption of orally dosed quinine salts occurs rapidly
`the colon 15 hours after ingestion, and (b) 5 hour timed-release systems
`within around 2 hours, with a bioavailability of 76-88%
`(or released contents) 6 hours after ingestion. The distribution is remark(cid:173)
`ably similar despite the differing lengths of time post ingestion.
`(19,20,21). Quinine hydrochloride is very soluble in both water
`
`ft
`
`Cosmo Ex. 2032-p. 4
`Argentum v Cosmo
`IPR2018-00080
`
`

`

`Regional Differences in Colonic Absorption of Quinine
`
`1091
`
`and lipid (1 gram dissolves in 16 mis water, I ml chloroform,
`and in 350 mis ether). Quinine has pKa's of 4.2 and 8.8 and
`is therefore partially ionised in the small and large bowel,
`where pH ranges from 5.5 to 7.8. Since the log P of quinine
`is approximately 2.1, at pH 6.7 it will be equally partitioned
`between water and lipid. It is therefore likely to partition both
`through the lipid membrane and equally through small water
`soluble pores within the membrane (i.e., transcellularly). Its
`appearance within erythrocytes (20) is consistent with these
`concepts. Generally, it is representative of many classes of
`pharmacological compounds which are absorbed through the
`transcellular pathway. Cr-51 EDTA is a large chelate (molecular
`mass = 359 Daltons) carrying a small negative charge, and is
`thought to permeate via the paracellular pathway. It adopts a
`strictly extracellular distribution when given intravenously, and
`is therefore thought to be unable to pass through cell membranes
`(ie transcellularly).
`There arc several possible explanations for the gradient
`observed with quinine. Firstly, the frequency of aqueous pores
`may be different between cells in different parts of the gastroin(cid:173)
`testinal tract. Secondly, the mucous layer in the colon limits
`access to the absorptive surface, and this layer has been shown to
`become progressively thicker distally (22). Finally and perhaps
`most importantly, the luminal contents differ between the proxi(cid:173)
`mal and distal colon. The right side of the colon contains liquid
`stool which would be predicted to promote drug dispersion,
`diffusion and mucosal contact whereas the left side of the
`colon contains viscous, dehydrated stool within which drug
`may become sequestered, and hence unavailable for permeation.
`An earlier pilot study involving 10 subjects had suggested
`decreased absorption of CrEDTA with more distal colonic
`release (23). We found no such gradient when a larger number of
`subjects were studied (n = 36). It is possible that the preliminary
`study suffered from a type I error, as the numbers in each
`region were small (small intestine 4, ascending colon 3, trans(cid:173)
`verse colon 3). It is however noteworthy that several EDTA
`absorption values seen in the larger study were much higher
`than expected. The usual upper limit of absorption of CrEDTA
`following ingestion as a solution is around 2.6% (24). and
`several EDTA absorption values seen in this study far exceed
`this figure. One explanation of the abnormally high values seen
`of these results'is that pulsed delivery of sucrose and CrEDTA
`may result in high local concentrations which alter local perme(cid:173)
`ability at the site of release, and hence confound any true
`differences in absorption by region.
`In summary, this study has shown the feasibility of selec(cid:173)
`tively targeting material to the ilco-colonic region using the
`Pulsincap™ timed release delivery system. Selectively targeting
`material to the distal colon presents particular difficulties due
`to a combination of temporal factors, interindividual variability
`in transit rates, and the reservoir function of the intervening
`proximal colon. Using the 5 and 6 hour timed release delivery
`systems to target the distal gastrointestinal tract, we have shown
`a gradient of absorption for (he transcellular probe quinine.
`Whether this gradient is primarily determined by differences
`in the water content of the stool or properties of the mucosa-
`mucus barrier remains to be determined.
`
`ACKNOWLEDGMENTS
`The authors thank Schercr DDS, Clydebank, Scotland for
`supply of the Pulsincap™ timed release delivery system, and
`
`to Dr Harry Seager and Dr Julie Binns for their contributions
`to the project. The work was funded as part of an MRC (Medical
`Research Council) Link Project with Scherer DDS,
`Clydebank, Scotland.
`
`REFERENCES
`
`1. K. H. Anlonin, P. Bieck, M. Scheurlcn, M. Jedrychowski, and H.
`Malchow. Oxprenolol absorption in man after single bolus dosing
`into two segments of the colon compared with that after oral
`dosing. Br. J. Clin. Pharm. 19:I37S-42S (1985).
`2. C. H. Gleiter. K. H. Antonin, P. Bieck. J. Godbillon, W. Schon-
`Icber, and H. Malchow. Colonoscopy in the investigation of drug
`absorption
`in healthy volunteers. Gastrointest. Endoscopy
`31:71-3(1985).
`3. C. G. Wilson, M. Bakhshaee, H. E. Stevens, A. C. Perkins. M.
`Frier, P. E. Blackshaw. and J. S. Binns. Evaluation of a gastro-
`rcsistant pulsed release delivery system (Pulsincap) in humans.
`Drug Del. 4:201-6(1997).
`4. A. R. Zoest. S. Wanwimolruk, and C. T. Hung. Simple high-
`performance liquid chromatographic method for the analysis of
`quinine in human plasma without extraction. J. Liquid Chro-
`matogr. 13:3481-91 (1990).
`5. J. G. Hardy, C. G. Wilson, and E. Wood. Drug delivery to the
`proximal colon. J. Pharm. Pharmacol. 37:874-77 (1985).
`6. M. Proano, M. Camilleri, S. F. Phillips, M. L. Brown, and G. M.
`Thomforde. Transit of solids through the human colon: regional
`quantification
`in the unprepared bowel. Am. J. Physiol.
`258:G856-62(I990).
`7. I. Torsoli, M. L. Famorino, and V. Crucioli. The relationships
`between anatomy and motor activity of the colon. Am. J. Dig.
`Dis. 13:462-67(1968).
`8. P. S. Kamath, S. F. Phillips, M. K. O'Connor, M. L. Brown,
`and A. R. Zinsmcister. Colonic capacitance and transit in man:
`modulation by luminal contents and drugs. Gut 31:443-49 (1990).
`9. L. Barrow, K. P. Steed, R. C. Spiller, P. J. Watts, C. D. Melia,
`M. C. Davies, and C. G. Wilson. Scintigraphic demonstration of
`accelerated proximal colonic transit by lactulose and its modifica(cid:173)
`tion by gelling agents. Gastroenterology 103:1167-73 (1992).
`10. J. Prcxinos, L. Bueno, and J. Fioramonti. Diurnal changes in
`myoelectric spiking activity of the human colon. Gastroenterology
`88:1104-10(1985).
`11. F Narducci, G. Bassotti, M. Gaburri, and A. Morrelli. Twenty
`four hour manometric recording of colonic motor activity in
`healthy man. Gut 28:17-25 (1987).
`12. E. E. Soffer, P. Scalabrini, and D. Wingate. Prolonged ambulant
`monitoring of human colonic motility. Am. J. Physiol.
`2S7:G60I-6(1989).
`13. G. Bassotti, C. Bctti, B. P. Imbimbo. M. A. Pelli, and A. Morelli.
`Colonic motor response to eating: a manometric investigation in
`the proximal and distal portions of the viscus in man. Am. J.
`Gastroenterol. 84:118-22 (1989).
`14. G. Bassotti, C. Betti, C. Fusaro, and A. Morelli. Colonic high-
`amplitude propagated contractions (mass movements): repeated
`24-h manometric studies in healthy volunteers. J. Gastrointest.
`Mot. 4:187-91 (1992).
`15. R. H. Ghoo, J. G. Moore, E. Grccnberg, and N. P. Alazraki.
`Orcadian variation in gastric emptying of meals in humans. Gas(cid:173)
`troenterol 93:515 -18 (1987).
`16. D. Kumar. D. Wingate, and Y. Ruckcbusch. Orcadian variation
`in the propagation velocity of the migrating motor complex. Gas(cid:173)
`troenterology 91:926-30 (1986).
`17. S. Travis and I. Mcnzies. Intestinal permeability: functional
`assessment and significance. Clin. Sci. 82:471-88 (1992).
`18. DC. Maxton. I. Bjarnason, A. P. Reynolds. S. D. Catt, T. J. Peter,
`and I. S. Menzies. Lactulose, "Cr-labelled cthylenediaminetetra-
`acetatc, L-rhamnosc and polyclhyleneglycol 500 as probe markers
`for assessment in vivo of human intestinal permeability. Clin. Sci.
`71:71-80(1986).
`19. I. Bjarnason, D. Maxton, A. P. Reynolds, S. Catt, T. J. Peters,
`and I. S. Mcnzies. Comparison of four markers of intestinal
`permeability in control subjects and patients with coeliac disease.
`Scand. J. Gastroenterol. 29:630-39 (1994).
`
`Cosmo Ex. 2032-p. 5
`Argentum v Cosmo
`IPR2018-00080
`
`

`

`1092
`
`Hebden et al
`
`20. L. A. Salako and A. Sowunmi. Disposition of quinine in plasma,
`red blood cells and saliva after oral and intravenous administera-
`tion to healthy adult Africans. Eur. J. Clin. Pharmacol.
`42:171-74(1992).
`21. G. Paintaud, G. Alvan, and O. Ericsson. The reproducibility
`of quinine bioavailability. Br. J. Clin. Pharm. 35:305-7
`(1993).
`22. R. D. Pullan, G. A. O. Thomas. M. Rhodes, R. G. Newcombe,
`G. T. Williams, A. Allen, and J. Rhodes. Thickness of adherent
`
`mucus gel on colonic mucosa in humans and its relevance to
`colitis. Gut 35:353-59 (1994).
`23. J. M. Hebden, A. C. Perkins, P. J. Gilchrist. P. E. Blackshaw, C.
`G. Wilson, and R. C. Spiller. A dual radionuclide study for the
`investigation of colonic drag delivery and absorption. Nucl. Med.
`Commmt. 16:232(1995).
`24. I. Bjarnason, T. J. Peter, and N. Veall. A persistent defect in
`intestinal permeability in coeliac disease demonstrated by a "Cr-
`labclled EDTA absorption test. Lancet i:323-25 (1983).
`
`Cosmo Ex. 2032-p. 6
`Argentum v Cosmo
`IPR2018-00080
`
`