Gastrointestinal pH and Transit Time Profiling in Healthy Volunteers Using the IntelliCap System Confirms Ileo-Colonic Release of ColoPulse Tablets

Introduction ColoPulse tablets are an innovative development in the field of oral dosage forms characterized by a distal ileum and colon-specific release. Previous studies in humans showed release in the ileo-colonic region, but the relationship between gastrointestinal pH and release was not experimentally proven in vivo. This information will complete the in vivo release-profile of ColoPulse tablets. Materials and Methods Release from ColoPulse tablets was studied in 16 healthy volunteers using the dual label isotope strategy. To determine gastrointestinal pH profiles and transit times the IntelliCap system was used. A ColoPulse tablet containing 13C-urea and an uncoated, immediate release tablet containing 15N2-urea were taken simultaneously followed by a standardized breakfast after three hours. Five minutes after intake of the tablets the IntelliCap capsule was swallowed and pH was measured until excretion in the feces. Breath and urine samples were collected for isotope analysis. Results Full analysis could be performed in 12 subjects. Median bioavailability of 13C -urea was 82% (95% CI 74–94%, range 61–114%). The median lag time (5% release of 13C) was 5:42 h (95% CI 5:18–6:18 h, range 2:36–6:36 h,) There was no statistically significant difference between lag time based on isotope signal and colon arrival time (CAT) based on pH (median 5:42 vs 5:31 h p = 0.903). In all subjects an intestinal pH value of 7.0 was reached before release of 13C from the ColoPulse tablet occurred. Discussion and Conclusions From the combined data from the IntelliCap system and the 13C -isotope signal it can be concluded that release from a ColoPulse tablet in vivo is not related to transit times but occurs in the ileo-colonic region after pH 7.0 is reached. This supports our earlier findings and confirms that the ColoPulse system is a promising delivery system for targeting the distal ileum and colon. Trial Registration ISRCTN Registry 18301880

h (95% CI 5:18-6:18 h, range 2:36-6:36 h,) There was no statistically significant difference between lag time based on isotope signal and colon arrival time (CAT) based on pH (median 5:42 vs 5:31 h p = 0.903). In all subjects an intestinal pH value of 7.0 was reached before release of 13 C from the ColoPulse tablet occurred.

Discussion and Conclusions
From the combined data from the IntelliCap system and the 13 C -isotope signal it can be concluded that release from a ColoPulse tablet in vivo is not related to transit times but occurs in the ileo-colonic region after pH 7.0 is reached. This supports our earlier findings and confirms that the ColoPulse system is a promising delivery system for targeting the distal ileum and colon.

ISRCTN Registry 18301880
Introduction Distal ileum and Colon-specific delivery of medicines is clinically relevant because their efficacy can be improved, side effects can be reduced and the bioavailability of drugs that are metabolized or poorly absorbed in the higher parts of the small intestine can be enhanced. This offers interesting perspectives for the treatment of for instance inflammatory bowel diseases with peptides [1].
In the literature different strategies for colon-specific delivery have been described. They include pH-responsive systems, time-based systems and systems triggered by the colon flora, as well as combinations of such systems [1,2]. The recently developed ColoPulse technology is a promising pH-responsive system being able to specifically deliver the active substance to the ileo-colonic region. Because of the non-percolating incorporation of a super-disintegrant in the coating, by which a more reliable and pulsatile release is achieved, it differs from other pHresponsive systems. The dissolution of this coating is triggered by the physiologically occurring increase in pH from 5.5 in the upper small intestine to 7.5 in the ileo-colonic region [3].
Until now bioavailability from ColoPulse dosage forms has been studied in healthy volunteers and in patients with Crohn's disease using stable isotopes of urea [4,5,6]. All studies showed a mean total bioavailability of 13 C-urea of > 76%, but the relationship between gastrointestinal pH and release from a ColoPulse dosage form has so far not been proven experimentally in vivo. To obtain more insight in the behaviour and functioning of ColoPulse tablets and capsules in vivo, studies on the gastrointestinal pH with the concomitant administration of a ColoPulse tablet are justified.
In the literature different devices for gastrointestinal pH measurement have been described [7]. They can be divided in two subgroups: the "static" devices and the "freely moving" devices. With representatives of the first group only the oesophageal, intra-gastric or peri-mucosal colonic pH can be measured. Their main disadvantage is the inability to measure the pH along the entire length of the gastrointestinal tract. The second group mainly comprises pH sensitive wireless radiotelemetry capsules (RTC). After administration of a RTC the subject is able to perform normal daily activities. Emitted radio signals are detected by a recorder mostly worn around the waist. If the device functions properly, measuring ends when the RTC is excreted with the feces. Disadvantages of the RTCs used so far are the frequent loss of signals, batteries running out of power before excretion, large pH-drift (up to approximately 1 unit) and difficulties in determining the exact location of the capsule in the gastrointestinal tract. In Table 1 a summary of the available literature on freely moving devices studied in inflammatory bowel diseases and / or healthy volunteers is presented [8][9][10][11][12][13][14][15][16][17][18][19][20][21].
Recently a medical device for the in vivo measurement of pH and temperature in the gastrointestinal tract was developed by Medimetrics (Eindhoven, The Netherlands): the IntelliCap system [20,21]. Furthermore the system can be used for electronically controlled drug delivery in defined sections of the gastrointestinal tract to quantify regional drug absorption [22]. It differs from the so far available freely moving RTCs by more accurate and more frequent measurements, minimal signal loss, a built-in drug reservoir and improved battery power (> 72 h) to ensure reliable and complete data acquisition. By combining diagnostic functionalities and the capability to generate adjustable controlled drug release profiles, the IntelliCap system can play a promising role in pharmaceutical drug profiling and formulation development.
In this paper we describe a study performed in healthy volunteers in which we studied the relationship between gastrointestinal pH obtained with the IntelliCap system as well as the release from a ColoPulse tablet to prove that release from a ColoPulse tablet indeed does occur in the ileo-colonic region and after a pH value of 7.0 is reached.

Subjects
Sixteen healthy volunteers (10 male, 6 female, age 18-65) were initially included in this study ( Table 2). Participant recruitment started January 2011 and ended April 2011. Written informed consent was obtained from all participants. They had no history of gastrointestinal diseases or gastrointestinal surgery. None of the subjects used antibiotics or drugs influencing the gastrointestinal transit time for at least three months prior to the start of the study. A possible Helicobacter pylori infection was excluded with a 13 C-urea breath test (INFAI, Köln, Germany).
A flowchart summarizing recruitment and analysis is shown in Fig 1.

Study-design
This bioavailability study was performed as an open-label, non-randomized, single arm clinical trial and was part of a study described by Maurer et al [6]. The design was based on a previous feasibility study in which two stable isotopes of urea are administered simultaneously [23]. Subjects were fasted from 8 p.m. the day before the test day. Only water, apple juice (until 11 p. m.) and unsweetened tea without sugar were allowed. On the test day an uncoated tablet containing 50 mg 15 N 2 -urea and a ColoPulse tablet containing 50 mg 13 C-urea were taken simultaneously at around 8 a.m. with 100 mL apple juice. Five minutes thereafter the IntelliCap capsule was swallowed with another 100 mL apple juice. A standardized breakfast was taken three hours after the intake of the tablets. The meal consisted of a standardized double sandwich and 200 mL unsweetened tea. Approximately 6 and 10 hours after tablet intake, respectively lunch and dinner were taken. There were no food-restrictions for lunch and dinner except foods rich in 13 C, like corn products, cane sugar and pineapple. During the test day, (that ended at 8 a.m. the next morning) water, apple juice and tea were the only beverages allowed. Sampling, administration of the tablets and IntelliCap capsule took place in a controlled facility until 5 p.m. Thereafter subjects went home were they continued sampling of breath and urine according to the study protocol. All necessary information was recorded in a diary. A summary of the study design is shown in Table 3.

Ethics statement
The study was approved by the ethical committee of the University Medical Center Groningen (ref 2009.188 / EudraCT 2009-01347121) and was performed according to the principles of the Declaration of Helsinki. The study has been registered in the ISRCTN register (ISRCTN18301880). This was not done before the start of recruitment because at that time this was not required by internal procedures and the ethical committee that approved this study.

IntelliCap system
The IntelliCap system was supplied by Medimetrics (Eindhoven, The Netherlands) and consisted of a capsule and a portable unit. The size of the IntelliCap capsule was 27 x 11 mm. A complete description including illustrations of the IntelliCap system can be found in the literature [20]. The drug reservoir was filled with normal saline solution which had no function in this study and was not expelled from the IntelliCap capsule during the experiments. Data were measured until excretion of the IntelliCap capsule or until the battery ran out of power. The excretion of the IntelliCap capsule from the body had to be confirmed by collecting the device from the stool. If the IntelliCap capsule was not retrieved within 96 hours it was probably missed and its absence from the body was confirmed with an abdominal X-ray. No follow up was required after retrieving the IntelliCap capsule or confirmation of its absence by X-ray.

Analysis of pH profiles
The pH profiles were analysed for gastrointestinal landmarks (ingestion, pylorus, ileocecal valve and excretion) using the following criteria:

Chemicals, isotopes and coated tablets
All substances were of pharmacopoeial grade (Ph. Eur. or USP) and were obtained via a certified wholesaler as described before [6]. The stable isotopes 13 C-urea and 15 N 2 -urea (AP 99%) were obtained from an FDA-controlled facility (Isotec, USA). Tablet cores containing 50 mg 13 C-or 50 mg 15 N 2 -urea and 25 mg caffeine were compounded in the Department of Hospital and Clinical Pharmacy of the University Medical Center Groningen and analysed according to the European Pharmacopoeia 7 th edition A ColoPulse coating of 13-17 mg/cm 2 was applied on the tablets containing 13 C-urea.The coating was composed of a mixture of Eudragit S-100:PEG 6000:Ac-di-sol:talc in a ratio of 7:1:3:2 (w/w). The solvent was an acetone-water 97:3 mixture (w/w). Coating thickness was determined and expressed as the amount of Eudragit S100 applied per cm 2 . Caffeine was added to the 13 C-urea tablet cores for quality control purposes and was used as a marker substance for the in vitro determination of the release characteristics lag-and pulse time in the in vitro dissolution test. Caffeine was also added to the 15 N 2 -urea tablet cores to obtain comparable tablet cores, with no particular function in this tablet. All tablets, coated and uncoated, met established pharmaceutical quality control criteria [6].

Urea-kinetics
To study the bioavailability from a ColoPulse tablet in the ileo-colonic region the difference in kinetics and fate between 13 C-urea and 15 N 2 -urea was used. An overview of the relevant kinetic steps can be found in Maurer et al [23,24]. Release of 13 C-urea in the ileo-colonic region (urease-rich) from a ColoPulse tablet leads to in situ fermentation of 13 C-urea into 13 CO 2 which is subsequently exhaled in breath. The delivery of the isotope in the colon can therefore be established by measuring the 13 CO 2 response in breath. Unfermented urea (i.e. release in the small intestine, urease-poor) can be measured as the amount of 13 C-urea in urine. The second stable isotope of urea, 15 N 2 -urea, in an uncoated tablet functions as a reference and reflects 100% release in a urease-poor region. Release of 15 N 2 -urea in the small intestine from an uncoated capsule leads to recovery of 15 N 2 -urea in urine. Bioavailability can be described by the difference between kinetics of 13 C-and 15 N 2 -urea [23,24].

Sample collections and analysis
Breath samples were collected every 0.5-1 h up to 15 h after intake of the tablets (Table 2) and were analysed as described before [23]. Briefly, 13 C/ 12 C isotope ratios in the CO 2 of breath samples were analysed by using a validated breath 13 C-analyser (Thermo Fisher Scientific, Bremen, Germany) based on isotope ratio mass spectrometry (IRMS). Urine samples were collected during 24 h after intake of the tablets at prescribed intervals (Table 2) in 500 or 1000 mL containers containing an aliquot of 6M HCl. Urine volumes were recorded and 20 mL samples were stored at -80°C until analysis. Concentrations of total 15 N and 13 C were determined as described before using an elemental analyzer interfaced with IRMS [23].

Calculations
The Percentage of the administered Dose Recovered (PDR) of 13 C and 15 N in each urine sample, the ratio of the PDRs 13 C versus 15 N (the 13 C/ 15 N-ratio), the fermented (F fermented ) and not-fermented (F not-fermented ) fraction of 13 C urea were calculated as described before [23,24]. In short: ▪. F fermented was calculated as the cumulative (c)PDR of 13  All data were corrected for baseline-concentrations of 13 C and 15 N in breath and /or urine. Furthermore, breath data were corrected for CO 2 -retention as described before [4].

Statistical procedures
This study was performed as a bioavailability study. Based on previous data on transit times of a ColoPulse tablet a sample size of 10 patients is needed to detect a clinically relevant difference of 15% between lag-time based on isotope-signal and colon arrival time based on pH with 80% power and a significance level of α = 0.05 (two sided). Because this study was part of another study [6] requiring a higher sample size and anticipating some drop-out 16 subjects were included.
The results were evaluated by descriptive statistics with SPSS version 22. Normal distribution of the data was investigated with the Shapiro-Wilk test. The center was characterized by the mean and standard deviation (pH-data) or the median and corresponding bootstrap based 95% confidence intervals (95% CI). The dispersion was characterized by the coefficient of variation (CV) and range because not all data were normally distributed. A (parametric) pairedsamples t-test (two tailed, α = 0.05) was used to compare the results within groups when data were normally distributed for both variables. A (non-parametric) Wilcoxon signed rank test was performed to compare the results when at least one of the variables was not normally distributed. Differences were considered significant when p < 0.05.

Endpoints
The endpoint was to investigate the relationship between the gastrointestinal pH-profile obtained with the IntelliCap system and release of 13 C-urea from a ColoPulse tablet and to confirm that release occurs in the ileo-colonic region after pH 7.0 has been reached.

Results
The results of 15 out of 16 healthy volunteers initially included in the study were evaluated (Fig  1). One volunteer could not swallow the IntelliCap capsule and was therefore excluded without replacement. Two other subjects (6 and 15) appeared to have a prolonged gastric residence time and the IntelliCap capsule was still in the stomach when breakfast and the following meals were taken. Because their gastric residence time was respectively 17 and 22 h data of these subjects were excluded from any analysis comprising transit times. Their lag time based on the isotope signal was within the normal range. Subject 5 was also excluded from analysis comprising lag time and bioavailability, because of a probably altered urea metabolism. This was concluded from the fact that the cPDR 13 C in breath was <6.5% after 15h combined with a cDPR of unfermented 13 C in urine of 70% at t = 24 h. The pH profile of this subject was normal with a GRT of 0:15 h and a SBTT of 3.15 h. The coating functioned well, because the appearance of 13 C in the urine sample could be seen in the sample collected between t = 4 and 7 h and not earlier. This means that 13 C-urea in stead of being fermented was absorbed into the bloodstream when it was released at the ileo-colonic region. The data from the remaining 12 subjects were available for all analyses.
IntelliCap capsules could be recovered from the feces within 72 hours after intake in 13 out of 15 subjects. For two subjects the temperature data indicated that the IntelliCap capsule had left the body, but the subjects failed to retrieve it from the feces. Absence from the body was confirmed with an abdominal X-ray. No adverse events potentially related to the IntelliCap system were observed during the study.
In three subjects the portable unit ran out of power after circa 60 hours. This did not influence the data collection necessary for the endpoint analysis because in all subjects the Intelli-Cap capsule already passed the cecum. However, for these subjects time of excretion and whole gut transit time (WGTT) could not be determined.
In three other subjects the communication between the capsule and portable unit was interrupted varying from 4 to 12 h because the subjects did not wear the portable unit close enough to the body or did not wear it. This also did not influence the data collection for endpoint analysis because all interruptions occurred more than 24 h after intake, when the IntelliCap capsule already had passed the cecum. For one of these subjects excretion, CTT and WGTT could not be determined because excretion occurred during the period of interrupted communication. No other loss of data was encountered in the study.
All gastrointestinal pH profiles recorded with the IntelliCap system were analysed according to the mentioned methods. A summary of the gastrointestinal transit times is shown in Fig 2  and a representative example of a pH and temperature profile is shown in Lag time and bioavailability of 13 C-urea from a ColoPulse tablet were calculated as described. The median lag time was 5:42 h (95% CI 5:18-6:18 h, range 2:36-6:36 h, CV 18%) and median bioavailability was 82% (95% CI 74-94%) range 61-114%, CV 10%). More detailed results can be found in S1 summary and in Maurer et al [6].
There was no statistically significant difference between CAT based on pH-data (IntelliCap) and the lag time of the ColoPulse tablet based on the stable isotope signal of 13 C-urea in breath (median 5:31 vs 5:42 h, p = 0.903, parametric test). A representative example is shown in Fig 5. Information about all subjects can be found in Fig 6. In all subjects a pH value of 7.0 was reached before release of 13 C from the ColoPulse tablet occurred, as measured in exhaled breath. There was a statistically significant difference between the time when pH 7.0 was reached and the lag time (185 vs 342 minutes, p = 0.002, nonparametric test). Discussion This is the first study in humans combining release from ColoPulse tablets and in vivo gastrointestinal pH measurements using the IntelliCap system. There was no difference between CAT (pH-data) and lag time (isotope data) found in this study which delivers proof that release of ColoPulse tablets occurs in the ileo-colonic region. Furthermore the results show that release from the ColoPulse tablet does not occur before an intestinal pH value of 7.0 is reached.
Several studies have been published with radiotelemetry capsules (RTC). Through the years the functioning of RTCs improved considerably. The first RTCs were used in the 1970's and had logging intervals from 5 to 120 minutes to save battery capacity [8,10]. Furthermore high data loss (75%) has been described [9]. Thereafter RTCs with shorter logging intervals up to 5 seconds became available and battery life improved considerably [16,17,19]. However in some studies data loss was still described mostly attributed to the angle between the RTC and the antenna [16]. In the current study with the IntelliCap system data loss was observed in three subjects out of 15. This didn't occur in the first 8 hours after intake of the capsule, when data were also sent to the control center, but only when the subjects were at home and did not keep their receiver close enough to the body. Therefore this event did not influence the outcome of our study, because the IntelliCap capsule already passed the cecum at the time data loss occurred. On the other hand, this shows that an overnight stay in a controlled facility is preferable when longer pH profiling is needed.
The IntelliCap system was able to record complete gastrointestinal pH and temperature profiles as well as derived transit-times from intake to excretion. The observed gastrointestinal transit times for the small intestine and the colon are within the range of earlier published data of healthy volunteers, only the median GRT appeared to be increased [10,25]. Published data collected with RTC and dosage forms labeled with gamma emitting radionuclides mention an average GRT of about 1 h in fasted, healthy volunteers while we observed an average gastric residence time of 1:30 h in the evaluable subjects. A likely explanation is the intake of the tablets with apple juice, since it is known that food increases the GRT of pharmaceutical dosage forms [25]. Due to its caloric content apple juice apparently also has a delaying effect. However in this study, the apple juice was given to get the same study design as previous studies with ColoPulse formulations. This was done to be able to compare the functioning of ColoPulse tablets used in this study with capsules which were used in previous studies. In the future Colo-Pulse formulations can also be administered with water.
All subjects showed an elevated gastric pH (pH 3-4) immediately after administration probably due to administration with apple juice, which decreased to a more acidic level of around pH 1.6 in about 30 minutes. Because in our study the determination of the location of the Intel-liCap capsule was only based on pH, pH values of the different segments of the small intestine and colon could not be determined. The median pH values of the stomach, small intestines and colon as observed in the majority of subjects are consistent with published data from fasted, healthy volunteers [7].
No difference was found between the colon arrival time based on the 13 C-isotope signal (lag time of ColoPulse tablets) and pH-measurements (from IntelliCap system). This proves the site-specific release of the active substance from the ColoPulse tablets in the ileo-colonic region. Simultaneous migration of the ColoPulse tablet and the IntelliCap capsule after leaving the stomach is supported by the literature. According to Davis et al [25] no difference in intestinal transit times was seen between solid dosage forms with the same size of ColoPulse tablets and the IntelliCap capsule. Gastric emptying of large single unit systems however, was highly influenced by the presence of food in the stomach. Even a light breakfast delayed emptying in some subjects. This may explain the increased stomach residence time of the IntelliCap capsule in two subjects as seen in the current study. In these subjects colon arrival time and bioavailability based on isotope signal were within the normal range. However the relatively large IntelliCap capsule was retained in the stomach for respectively 17 and 21 h, probably because they returned to their "normal" meal intakes when the IntelliCap capsule was still in the stomach. After pylorus passage of these two IntelliCap capsules intestinal transit times were comparable to those of the other subjects.
In this study we observed no relation between the lag-time of a ColoPulse tablet and the time when pH 7.0 was reached or the CAT. However, in none of the subjects release from the ColoPulse tablet occurred before pH 7.0 was reached.
The difference between the time when pH 7.0 was reached and the CAT was approximately 2.5 hours, supporting the fact that release occurs in the distal ileum and colon. Remarkably, this difference in time is relatively large and differs from parameters used in vitro dissolution tests that were performed with the ColoPulse tablets in the gastrointestinal simulation system (GISS) [26]. We use this in vitro test for quality control of ColoPulse tablets and normally dissolution of the coating and subsequent release occurs within 30 minutes after raising pH from 6.8 to 7.5. However, the volumes of intestinal fluid in vivo differ from the volumes used in the GISS. According to Schiller et al [27] the fluid volume of the small intestine has a maximum of 319 mL while the volume in this stage of the GISS is as high as 940 mL. Furthermore the fluid is not distributed homogenously along the small intestine in vivo with water pockets and "dry" segments randomly scattered. This contributes to the relatively slow dissolution of the Eudragit-S coating and is probably the cause of the relatively large difference between the time point at which pH 7 was reached and the CAT. The clinical relevance of this phenomenon seems to be limited because median bioavailability was 82%.

Conclusion
Based on the combined data from the IntelliCap system and the urea-isotope signal from a ColoPulse tablet as obtained in this study in healthy volunteers it can be concluded that release from ColoPulse tablets indeed occurs in the distal ileum and colon and after pH 7.0 is reached. This supports our earlier observations and confirms that the ColoPulse system is a promising delivery system for site-specific delivery and local therapy in inflammatory bowel diseases present in the distal ileum and colon. Supporting Information S1