Feasibility of capsule endoscopy for direct imaging of drug delivery systems in the fasted upper-gastrointestinal tract

Screening for Differences in Early Exposure in the Fasted State with in Vitro Methodologies can be Challenging: Experience with the BioGIT System

The Biorelevant Gastrointestinal Transfer (BioGIT) system is a useful screening tool for assessing the impact of dose and/or formulation on early exposure after administration of immediate release or enabling drug products with a glass of water in the fasted state. The objective of this study was to investigate potential limitations. BioGIT experiments were performed with five low solubility active pharmaceutical ingredients with weakly alkaline characteristics: mebendazole (tablet and chewable tablet), Compound E (aqueous solutions, three doses), pazopanib-HCl (Votrient™ tablet, crushed Votrient™ tablet and aqueous suspension), Compound B-diHCl (hard gelatin capsule, three doses) and Compound C (hard gelatin capsule containing nanosized drug and hard gelatin capsule containing micronized drug). For all formulation or dose comparisons the ratio of mean BioGIT AUC0-50 min values was not predictive of the ratio of mean plasma AUC0-60 min values which became available after completion of BioGIT experiments. BioGIT experimental conditions have not been designed to simulate the gastrointestinal drug transfer process after administration of chewable tablets or aqueous solutions, therefore, BioGIT may not be useful for the assessment of intraluminal performance early after administration of such drug products. Also, based on this study, BioGIT may not be useful in investigating the impact of dose and/or formulation on early exposure when the dose is not administered with a glass of water to fasted healthy individuals or when BioGIT data are highly variable. Finally, the rapid dissolution of nanocrystals after administration of low solubility weak bases may require adjustment of the pH in the gastric compartment of BioGIT to slightly higher pH values. Limitations identified in this study for the BioGIT system may be also relevant to other in vitro systems proposed for similar evaluations.

Classification of endoscopic image and video frames using distance metric-based learning with interpolated latent features

Conventional Endoscopy (CE) and Wireless Capsule Endoscopy (WCE) are well known tools for diagnosing gastrointestinal (GI) tract related disorders. Defining the anatomical location within the GI tract helps clinicians determine appropriate treatment options, which can reduce the need for repetitive endoscopy. Limited research addresses the localization of the anatomical location of WCE and CE images using classification, mainly due to the difficulty in collecting annotated data. In this study, we present a few-shot learning method based on distance metric learning which combines transfer-learning and manifold mixup schemes to localize and classify endoscopic images and video frames. The proposed method allows us to develop a pipeline for endoscopy video sequence localization that can be trained with only a few samples. The use of manifold mixup improves learning by increasing the number of training epochs while reducing overfitting and providing more accurate decision boundaries. A dataset is collected from 10 different anatomical positions of the human GI tract. Two models were trained using only 78 CE and 27 WCE annotated frames to predict the location of 25,700 and 1825 video frames from CE and WCE respectively. We performed subjective evaluation using nine gastroenterologists to validate the need of having such an automated system to localize endoscopic images and video frames. Our method achieved higher accuracy and a higher F1-score when compared with the scores from subjective evaluation. In addition, the results show improved performance with less cross-entropy loss when compared with several existing methods trained on the same datasets. This indicates that the proposed method has the potential to be used in endoscopy image classification.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11042-023-14982-1.

Usefulness of the BioGIT system in screening for differences in early exposure in the fasted state on an a priori basis

The objective of the present study was to confirm the usefulness of BioGIT data in the evaluation of the impact of dose and/or formulation on early exposure after oral administration of immediate release or enabling products of low solubility active pharmaceutical ingredients (APIs) with a glass of water in the fasted state. BioGIT experiments were performed with four APIs: Compound Α (tablet, three dose levels), Compound E (capsule PiC1, capsule PiC2 and tablet), fenofibrate (Lipidil® capsule and Lipidil 145 ONE® tablet) and Compound F (HP-β-CD aqueous solution and tablet). Based on mean plasma AUC_0-60min values which became available after completion of the BioGIT experiments, mean BioGIT AUC_0-50min values were useful for the evaluation of the impact of dose and/or formulation on early exposure. The log-transformed ratios of mean BioGIT AUC_0-50min values for two doses and/or two formulations estimated in this study and in a recent study for two diclofenac potassium products (Cataflam® tablet and Voltfast® sachet, same dose) vs. the corresponding log-transformed ratios of mean plasma AUC_0-60min values (n = 7 pairs of ratios), were included in a previously established correlation between log-transformed ratios of mean BioGIT AUC_0-50min values and log-transformed ratios of plasma AUC_0-60min values (n = 9 pairs of ratios). The correlation between log-transformed plasma AUC_0-60min ratios vs. log-transformed BioGIT AUC_0-50min ratios was confirmed (n = 16 pairs of ratios, R = 0.90). Compared with the previously established correlation the statistical characteristics were improved. Based on this study, the BioGIT system could be useful as a screening tool for assessing the impact of dose and/or formulation differences on early exposure, after administration of immediate release or enabling drug products of low solubility APIs with a glass of water in the fasted state, on an a priori basis.

Direct visualizing of paracetamol immediate release tablet disintegration in vivo and in vitro

The purpose of the present study was to study tablet disintegration by direct visualization, in vivo and in vitro. Based on literature data, a standard conventional paracetamol (CP) tablet, Panodil®, and a rapidly absorbed paracetamol (RP) tablet, Panodil® Zapp, were chosen as model systems to study tablet disintegration in the human stomach. Based on the obtained in vivo results, an in vitro disintegration method was designed to reproduce the visualized disintegration process occurring in the human stomach. For the clinical study, CP and RP tablets fastened to digital endoscopic camera capsules were administered to fasted human volunteers (n=4). The disintegration time and process were visualized by the real time video recordings, using the endoscopic camera capsule. The average disintegration time was found to be 26 ± 13 min and 10 ± 7 min, for CP (n=4) and RP (n=4) tablets, respectively. It was possible to reproduce the in vivo disintegration data in vitro using a USP 2 dissolution apparatus with 250 mL of viscous Fasted State Simulated Gastric Fluid (vFaSSGF*), simulating the rheological profile of human fasted state gastric fluid following administration of a glass of water. The viscosity of the simulated fasted state gastric fluid was found to have a large impact on the disintegration time of the tested immediate release tablets. Therefore, it is recommended to mimic gastric fluid viscosity during in vitro tablet disintegration studies.

Ingestible devices for studying the gastrointestinal physiology and their application in oral biopharmaceutics

Ingestible sensor systems are unique tools for obtaining physiological data from an undisturbed gastrointestinal tract. Since their dimensions correspond to monolithic oral dosage forms, such as enteric coated tablets or hydrogel matrix tablets, they also allow insights into the physiological conditions experienced by non-disintegrating dosage forms on their way through the gastrointestinal tract. In this work, the different ingestible sensor systems which can be used for this purpose are described and their potential applications as well as difficulties and pitfalls with respect to their use are presented. It is also highlighted how the data on transit times, pH, temperature and pressure as well as the data from different animal models commonly used in drug product development such as dogs and pigs have contributed to a deeper mechanistic understanding of oral drug delivery.

Direct visualization of drug behaviors in the upper GI tract via magnetically controlled capsule endoscopy

Background and Aims

Actual behaviors of drugs in the upper GI tract are not well elucidated. We assess the feasibility of magnetically controlled capsule endoscopy (MCE) in direct and real-time visualization of oral drug behaviors in the stomach.

Methods

From November 2019 to December 2019, 9 patients with a recent history of upper GI symptoms and 10 healthy volunteers were enrolled in this study. Participants swallowed magnetically controlled capsules to examine the whole stomach. After baseline examination, participants ingested dyed sucralfate gel, and MCE recorded the adhesion time, retention time, and distribution area of sucralfate gel. Outcomes included behaviors of sucralfate gel, safety, and satisfaction assessment of the procedures.

Results

Adhesion time of sucralfate gel in the abdominal symptoms group was significantly shorter than in the healthy control group (23.76 ± 1.37 minutes vs 31.96 ± 3.09 minutes; P = .032), whereas retention time was longer (98.85 ± 13.94 minutes vs 63.93 ± 8.57 minutes; P = .043). The distribution area of sucralfate gel in the abdominal symptoms group was significantly larger than in healthy control group in cardia (24.29 ± 7.39 vs 9.18 ± 4.06; P < .0001), fundus (18.90 ± 7.08 vs 8.49 ± 4.10; P = .0015), and pylorus (4.64 ± 2.72 vs 0.94 ± 0.90; P = .0019). No adverse events were observed. All participants had a high degree of satisfaction.

Conclusions

MCE is a feasible and noninvasive tool for direct and real-time visualization of drug behaviors (eg, sucralfate gel) in the stomach. (ClinicalTrials.gov. ID: NCT04327869.)

Dissolution testing of modified release products with biorelevant media: An OrBiTo ring study using the USP apparatus III and IV

During the OrBiTo project, our knowledge on the gastrointestinal environment has improved substantially and biorelevant media composition have been refined. The aim of this study was to propose optimized biorelevant testing conditions for modified release products, to evaluate the reproducibility of the optimized compendial apparatus III (USP apparatus III) and compendial apparatus IV (USP apparatus IV, open-loop mode) dissolution methods and to evaluate the usefulness of these methods to forecast the direction of food effects, if any, based on the results of two «ring» studies and by using two model modified release (MR) products, Ciproxin / Cipro XR and COREG CR. Six OrBiTo partners participated in each of the ring studies. All laboratories were provided with standard protocols, pure drug substance, and dose units. For the USP apparatus III, the dissolution methods applied to Ciproxin / Cipro XR, a monolithic MR product of an active pharmaceutical ingredient (API) with moderate aqueous solubility, were robust with low intra- and inter-laboratory data variability. Data from all partners were in line on a qualitative basis with food effect data in humans. For the USP apparatus IV, the dissolution methods applied to COREG CR, a multiparticulate, pH dependent, MR product of an API with low and pH dependent solubility led to high intra- and inter- laboratory data variability. Data from all partners were in line, on a qualitative basis, with the previously observed food effects in humans.

The Use of Capsule Endoscopy to Determine Tablet Disintegration In Vivo

The preferred delivery route for drugs targeted for systemic effect is by oral administration. Following oral administration, a solid dosage form must disintegrate and the drug dissolve, thereafter permeating the intestinal mucosa. Several different in vitro methods are used to investigate these processes, i.e., disintegration tests, dissolution tests, and permeability models. However, the actual behavior of oral dosage forms in the environment of the gastro-intestinal tract is not very well elucidated using these conventional methods. In this study, the use of capsule endoscopy to determine tablet disintegration in vivo was assessed. Panadol and Panadol Rapid (acetaminophen/paracetamol) were used as the test material. The in vivo tablet disintegration behavior in beagle dogs was assessed by the use of capsule endoscopy. The in vitro tablet disintegration behavior was assessed using the European Pharmacopeia (Ph. Eur.) disintegration test. The study showed that the in vivo disintegration times of Panadol and Panadol Rapid were 24.7 and 16.5 min, respectively, when determined by capsule endoscopy, which corresponded to the pharmacokinetic data. By contrast, the in vitro disintegration times of the same formulations were 5.5 and 4.0 min, respectively, when determined by the Ph. Eur. disintegration test. In conclusion, capsule endoscopy can be used to determine the in vivo tablet disintegration behavior. By contrast, the in vitro methods appear to not be predictive of the disintegration behavior in vivo but may be used to rank the order the formulations with respect to disintegration time.

In Vitro and In Vivo Test Methods for the Evaluation of Gastroretentive Dosage Forms

More than 50 years ago, the first concepts for gastroretentive drug delivery systems were developed. Despite extensive research in this field, there is no single formulation concept for which reliable gastroretention has been demonstrated under different prandial conditions. Thus, gastroretention remains the holy grail of oral drug delivery. One of the major reasons for the various setbacks in this field is the lack of predictive in vitro and in vivo test methods used during preclinical development. In most cases, human gastrointestinal physiology is not properly considered, which leads to the application of inappropriate in vitro and animal models. Moreover, conditions in the stomach are often not fully understood. Important aspects such as the kinetics of fluid volumes, gastric pH or mechanical stresses have to be considered in a realistic manner, otherwise, the gastroretentive potential as well as drug release of novel formulations cannot be assessed correctly in preclinical studies. This review, therefore, highlights the most important aspects of human gastrointestinal physiology and discusses their potential implications for the evaluation of gastroretentive drug delivery systems.

Capsule Design for Blue Light Therapy against Helicobacter pylori

A photo-medical capsule that emits blue light for Helicobacter pylori treatment was described in this paper. The system consists of modules for pH sensing and measuring, light-emitting diode driver circuit, radio communication and microcontroller, and power management. The system can differentiate locations by monitoring the pH values of the gastrointestinal tract, and turn on and off the blue light according to the preset range of pH values. Our experimental tests show that the capsule can operate in the effective light therapy mode for more than 32 minutes and the wireless communication module can reliably transmit the measured pH value to a receiver located outside the body.

Visualization of delayed release of compounds from pH-sensitive capsules in vitro and in vivo in a hamster model

Delayed controlled release is an innovative strategy to locally administer therapeutic compounds (e.g. chemotherapeutics, antibodies etc.). This would improve efficiency and reduce side effects compared with systemic administration. To enable the evaluation of the efficacy of controlled release strategies both in vitro and in vivo, we investigated the release of contrast agents ((19)F-FDG and BaSO4) to the intestinal tract from capsules coated with pH-sensitive polymers (EUDRAGIT L-100) by using two complementary techniques, i.e. (19)F magnetic resonance imaging (MRI) and computed tomography (CT). Using in vitro (19)F-MRI, we were able to non-destructively and dynamically establish a time window of 2 h during which the capsules are resistant to low pH. With (19)F-MRI, we could establish the exact time point when the capsules became water permeable, before physical degradation of the capsule. This was complemented by CT imaging, which provided longitudinal information on physical degradation of the capsule at low pH that was only seen after 230 min. After oral administration to hamsters, (19)F-MRI visualized the early event whereby the capsule becomes water permeable after 2 h. Additionally, using CT, the integrity and location (stomach and small intestines) of the capsule after administration could be monitored. In conclusion, we propose combined (19)F-MRI and CT to non-invasively visualize the different temporal and spatial events regarding the release of compounds, both in an in vitro setting and in the gastrointestinal tract of small animal models. This multimodal imaging approach will enable the in vitro and in vivo evaluation of further technical improvements to controlled release strategies.

Software for enhanced video capsule endoscopy: challenges for essential progress

Video capsule endoscopy (VCE) has revolutionized the diagnostic work-up in the field of small bowel diseases. Furthermore, VCE has the potential to become the leading screening technique for the entire gastrointestinal tract. Computational methods that can be implemented in software can enhance the diagnostic yield of VCE both in terms of efficiency and diagnostic accuracy. Since the appearance of the first capsule endoscope in clinical practice in 2001, information technology (IT) research groups have proposed a variety of such methods, including algorithms for detecting haemorrhage and lesions, reducing the reviewing time, localizing the capsule or lesion, assessing intestinal motility, enhancing the video quality and managing the data. Even though research is prolific (as measured by publication activity), the progress made during the past 5 years can only be considered as marginal with respect to clinically significant outcomes. One thing is clear-parallel pathways of medical and IT scientists exist, each publishing in their own area, but where do these research pathways meet? Could the proposed IT plans have any clinical effect and do clinicians really understand the limitations of VCE software? In this Review, we present an in-depth critical analysis that aims to inspire and align the agendas of the two scientific groups.

Identification of key factors affecting the oral absorption of salts of lipophilic weak acids: a case example

Objectives

Evaluate the ability of biorelevant media to adequately predict solubility in human gastrointestinal aspirates collected in the fasted state for the sodium salt of a highly dosed, Biopharmaceutics Classification System II (BCS II) compound with weakly acidic properties (L-870,810, pKa 7.3, HA (5-(1,1-dioxothiazinan-2-yl)-N-((4-fluorophenyl)methyl)-8-hydroxy-1,6-naphthyridine-7-carboxamide)). Identify key luminal processes that dictate the behaviour of sodium salt of HA (NaA), after single-dose administrations of high (relatively to solubility limit) doses corresponding to 400 and 800 mg of HA in the fasted state.

Methods

Aspirates from stomach and upper small intestine were collected from eight healthy fasted adults, after administration of 240 ml of water. Solubilities of NaA and HA were measured in aspirated samples and biorelevant media. Dissolution experiments of NaA granules were performed in biorelevant media. Prediction of oral pharmacokinetics was evaluated in silico using Stella software.

Key findings

Equilibrium solubility of NaA in fluids aspirated from the upper gastrointestinal tract is more transient than of HA. Solubility in upper gastrointestinal lumen was adequately estimated by data in biorelevant media. Supersaturation, followed by precipitation, which did not fully revert to the equilibrium solubility of HA, was observed during the dissolution of NaA granules in biorelevant media. Physiologically based pharmacokinetic modelling indicated that while intragastric processes had no significant impact on absorption kinetics, dissolution kinetics, kinetic solubility, radial transport rates and, for the 800-mg dose, precipitation kinetics in the small intestine had the greatest impact on absorption profiles.

Conclusions

Adequate prediction of the average plasma profile, after administration of NaA, required consideration of region-dependent dissolution rates and/or solubilisation.