University of Groningen
Endoscopic imaging in inflammatory bowel disease
van der Laan, Jouke J H; van der Waaij, Anne M; Gabriëls, Ruben Y; Festen, Eleonora A M;
Dijkstra, Gerard; Nagengast, Wouter B
Published in:
Expert Review of Gastroenterology & Hepatology DOI:
10.1080/17474124.2021.1840352
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2020
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
van der Laan, J. J. H., van der Waaij, A. M., Gabriëls, R. Y., Festen, E. A. M., Dijkstra, G., & Nagengast, W. B. (2020). Endoscopic imaging in inflammatory bowel disease: current developments and emerging
strategies. Expert Review of Gastroenterology & Hepatology, 1-12. https://doi.org/10.1080/17474124.2021.1840352
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=ierh20
Expert Review of Gastroenterology & Hepatology
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/ierh20
Endoscopic imaging in inflammatory bowel
disease: current developments and emerging
strategies
Jouke J.H. van der Laan , Anne M. van der Waaij , Ruben Y. Gabriëls ,
Eleonora A.M. Festen , Gerard Dijkstra & Wouter B. Nagengast
To cite this article: Jouke J.H. van der Laan , Anne M. van der Waaij , Ruben Y. Gabriëls , Eleonora A.M. Festen , Gerard Dijkstra & Wouter B. Nagengast (2020): Endoscopic imaging in inflammatory bowel disease: current developments and emerging strategies, Expert Review of Gastroenterology & Hepatology, DOI: 10.1080/17474124.2021.1840352
To link to this article: https://doi.org/10.1080/17474124.2021.1840352
© 2020 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.
View supplementary material
Published online: 31 Oct 2020. Submit your article to this journal
Article views: 343 View related articles
REVIEW
Endoscopic imaging in inflammatory bowel disease: current developments and
emerging strategies
Jouke J.H. van der Laan *, Anne M. van der Waaij*, Ruben Y. Gabriëls, Eleonora A.M. Festen, Gerard Dijkstra
and Wouter B. Nagengast
Department of Gastroenterology and Hepatology, University Medical Centre Groningen, Groningen, The Netherlands
ABSTRACT
Introduction: Developments in enhanced and magnified endoscopy have signified major advances in endoscopic imaging of ileocolonic pathology in inflammatory bowel disease (IBD). Artificial intelligence is increasingly being used to augment the benefits of these advanced techniques. Nevertheless, treatment of IBD patients is frustrated by high rates of non-response to therapy, while delayed detection and failures to detect neoplastic lesions impede successful surveillance. A possible solution is offered by molecular imaging, which adds functional imaging data to mucosal morphology assess-ment through visualizing biological parameters. Other label-free modalities enable visualization beyond the mucosal surface without the need of tracers.
Areas covered: A literature search up to May 2020 was conducted in PubMed/MEDLINE in order to find relevant articles that involve the (pre-)clinical application of high-definition white light endoscopy, chromoendoscopy, artificial intelligence, confocal laser endomicroscopy, endocytoscopy, molecular imaging, optical coherence tomography, and Raman spectroscopy in IBD.
Expert opinion: Enhanced and magnified endoscopy have enabled an improved assessment of the ileocolonic mucosa. Implementing molecular imaging in endoscopy could overcome the remaining clinical challenges by giving practitioners a real-time in vivo view of targeted biomarkers. Label-free modalities could help optimize the endoscopic assessment of mucosal healing and dysplasia detection in IBD patients. ARTICLE HISTORY Received 25 August 2020 Accepted 19 October 2020 KEYWORDS Artificial intelligence; chromoendoscopy; confocal laser endomicroscopy; endoscopy; endocytoscopy; inflammatory bowel disease; molecular imaging; optical coherence tomography; raman spectroscopy; surveillance
1. Introduction
Endoscopic imaging is essential in diagnosis and clinical man-agement of patients with inflammatory bowel disease (IBD). Through assessment of the location, the extent and severity of the inflammation, a differentiation between Crohn’s disease (CD) and ulcerative colitis (UC) can be made. Pathohistological examination of the biopsies obtained during endoscopy can substantiate the optical endoscopic diagnosis. Furthermore, endoscopic remission is considered as the main therapeutic goal in the treat-to-target approach as this is associated with various favorable outcomes [1].
Over the past decades, multiple antibodies targeting var-ious molecules (tumor necrosis factor α (TNFa), interleukin-12/ 23 (IL-12/23), Janus kinase 1/3 (JAK-1/3) and α4β7 integrin) have been developed and have shown significant efficacy in inducing and maintaining remission in IBD patients [2–8]. Unfortunately, responsiveness to these selective biological agents varies per individual IBD patient which potentially exposes the patient to an ineffective treatment and its possi-ble side effects [9,10]. The operational mechanism of biologi-cals in IBD remains largely unknown; crucial questions regarding prediction of response to treatment, appropriate
dose regimens and loss of response to treatment are still unanswered. An increasing amount of data associates mucosal drug concentration as a key contributor to response to treat-ment in IBD, rather than serum drug concentration [11–13]. Surprisingly, the distribution and concentration of antibodies in the target – the inflamed gut – are unknown because no methods are available to measure these values directly. Therefore, it is unknown whether the drug reaches its target at all – or in sufficient amounts – and how local drug concen-trations are related to response.
Chronic colonic inflammation and other mechanisms can give rise to colitis-associated adenoma and carcinoma (CAC) in IBD patients with longstanding and extensive colitis [14–16]. Thus, these patients have an increased risk of developing colorectal carcinoma and are therefore subject to various time-consuming surveillance protocols [17–20]. However, the diagnosis of CAC is delayed or missed in 17% to 28% of these patients [21,22]. Dysplasia in IBD frequently manifests as non- pedunculated lesions that present with only subtle visible changes or are even invisible due to being surrounded by inflammation, scarring, pseudopolyps, or hyperplasia [23,24].
Recent developments in endoscopy have enabled a more accurate endoscopic evaluation of ileocolonic pathology in
CONTACT Wouter B. Nagengast w.b.nagengast@umcg.nl Department of Gastroenterology and Hepatology, University Medical Centre Groningen, Groningen, GZ, 9713, The Netherlands
*Both authors contributed equally to this manuscript and share first authorship
Supplemental data for this article can be accessed here.
EXPERT REVIEW OF GASTROENTEROLOGY & HEPATOLOGY https://doi.org/10.1080/17474124.2021.1840352
© 2020 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way.
IBD. In the next section, we review leading-edge studies on various endoscopic imaging techniques and we summarize the advantages and limitations of currently available enhanced and magnified endoscopy techniques for inflam-mation assessment and dysplasia detection in IBD. Based on the results of molecular imaging in several (pre-)clinical trials, we then envision future prospects for molecular imaging to overcome the remaining clinical challenges in IBD, such as pretreatment patient stratification for optimal therapy, mon-itoring treatment response, assessing mucosal drug distribu-tion, and detecting dysplasia. Lastly, we describe two label- free modalities that could help optimize endoscopic assess-ment in IBD patients without the need of tracers. Supplementary to these invasive endoscopic examinations, novel approaches in cross-sectional imaging techniques pro-vide an improved noninvasive assessment of disease activity and response to medical therapy in the bowel wall and mesentery, as thoroughly described in recent literature [25].
2. Methods
The PubMed and MEDLINE databases were searched for articles relevant to inflammatory bowel disease in relation to high-definition white light endoscopy, artificial intelli-gence, dye-based/virtual chromoendoscopy, confocal laser endomicroscopy, endocytoscopy, molecular imaging, optical coherence tomography, and Raman spectroscopy. The aim of our literature search was to identify original studies con-ducted in IBD patients, involving endoscopic assessment of inflammatory activity or remission, predicting, or monitoring response to treatment, and dysplasia detection through the utilization of various techniques. Titles and abstracts of arti-cles published in English were reviewed. Eventually, we selected 51 articles that were published up to May 2020 based upon their relevance to the scope of this review. An overview of all studies included for the purpose of this manuscript is provided as supplementary material (Supplementary tables 1 and 2).
3. Enhanced endoscopic imaging
3.1. High-definition white light endoscopy (HD WLE)
Endoscopic evaluation of the gastrointestinal (GI) tract with high-definition white light endoscopy (HD WLE) enables visua-lization of the area of interest with a spatial resolution of up to two million pixels per image. Consequently, the operator can thoroughly examine the ileocolonic surface for subtle mucosal aberrances.
3.1.1. Assessment of endoscopic remission
HD WLE is widely available and generally considered as the standard technique for endoscopic examination. However, Iacucci et al. recently pointed out that an optimal definition of endoscopic remission has yet to be established. Current endoscopic indices that grade the severity of inflammation in IBD were not designed to assess endoscopic features of remission. Those indices were developed for previous genera-tions of endoscopy and lack parameters to deal with the increased complexity of the evaluation [26]. Additionally, there is a substantial inter-observer and intra-observer varia-bility in grading endoscopically observed severity of inflam-mation, even among experts [27].
3.1.2. Detection of dysplasia
Furthermore, although the introduction of HD WLE has yielded an improved dysplasia detection rate compared to standard definition WLE [28], the collection of random biop-sies during WLE surveillance examination is traditionally recommended, especially if appropriate expertise for a targeted approach is not available [29]. However, the value of segmental random biopsies is increasingly being ques-tioned due to its limited yield and time-consuming character compared to a targeted approach [30].
3.2. Artificial intelligence (AI)
In order to incorporate a more objective and independent assessment of mucosal imaging, the field of GI endoscopy is exploring the use of artificial intelligence (AI). Machine learn-ing is a form of AI that uses unique algorithms for data analysis; it includes various subtypes, such as deep learning. Convoluted neural network (CNN) algorithms are a specific form of deep learning, offering huge potential to improve the quality of endoscopic examination [31–33]. The first AI- based computer-aided detection (CAD) systems using CNN algorithms were developed for identifying endoscopic activ-ities of UC images according to the Mayo classification. These CAD systems showed promising results when staging inflam-mation activity, especially remission (Mayo 0–1) from moder-ate-to-severe activity (Mayo 2–3) [34,35]. Moreover, its performance was similar to experienced human reviewers [35]. In addition, two algorithms have been constructed that were able to provide an accurate and objective assessment of inflammation severity based upon endoscopic and histological disease activity. Thus, AI could identify patients in remission without the need for tissue specimen acquisition and exam-ination [36,37]. Although AI has been used extensively in
Article highlights
● The introduction of highdefinition white light endoscopy and chro-moendoscopy have signified major improvements in both the assess-ment of mucosal activity and the detection of dysplasia in IBD patients.
● Artificial intelligence is evolving as a topic of interest in the field of
gastrointestinal endoscopy as it could assist user interpretation of the progressively complex assessment of endoscopic images.
● Confocal laser endomicroscopy and endocytoscopy allow microscopic visualization of the ileocolonic mucosa, potentially enabling a detailed assessment of endoscopic remission and malignant lesions.
● Fluorescent molecular endoscopy can overcome remaining clinical
challenges in pretreatment patient stratification for optimal therapy, monitoring response to treatment, assessment of mucosal drug distri-bution and detection of dysplasia as solely endoscopic mucosal ima-ging appears to be insufficient.
● Label-free techniques provide information on mucosal healing and carcinogenesis in the bowel through the evaluation of mucosal and submucosal biomarkers.
computer-aided endoscopic polyp detection, no studies on CAC detection in IBD have been published so far.
Evaluation of current AI-inspired systems is hampered by relatively small datasets that are retrospectively selected by endoscopists that are not representative and overestimate quality of images [38]. In addition, diagnosis of individual images is most often based on a single expert which makes it subjective and less accurate [39,40]. Another limiting factor is that these systems are dependent on the quality of the colonoscopy, which can vary extensively dependent on opera-tor performance [41]. For computer-aided lesion detection and characterization, an adequate visualization is required in which the entire ileocolonic surface can be inspected and any poten-tial lesion is in the field of view [42]. To achieve this, an AI system was developed that can aid endoscopists in assessing their own performance, enabling them to re-inspect areas that were missed or not visualized sufficiently due to their own execution or poor bowel cleanliness [42]. Consequently, AI also provides opportunities for improving endoscopy indir-ectly by enhancing procedural quality. However, important ethical questions need to be discussed prior to worldwide implementation. These topics include whether the CAD- system will fulfill a role as second, concurrent, or independent observer and how to deal with misdiagnosis by AI. These questions are specifically relevant for CNN as this is considered as a complex ‘black box’ that is characterized by a lack of transparency. Therefore, understanding of its analysis is ham-pered which makes justification of unreliable predictions diffi-cult [38].
3.3. Chromoendoscopy (CE)
Chromoendoscopy (CE) can be differentiated into dye-based chromoendoscopy (DCE) and virtual chromoendoscopy (VCE). In DCE a contrast dye, indigo carmine or methylene blue, is sprayed topically, enabling visualization of mucosal lesions that disrupt the normal surface topography by highlighting inflamed or dysplastic areas Figure 1A [43].
VCE includes pre-processing optical imaging techniques such as narrow band imaging (NBI) (Olympus, Tokyo, Japan) and autofluorescence imaging (Olympus, Tokyo, Japan), but also the post-processing techniques iSCAN (Pentax, Tokyo, Japan), and Fujinon intelligent chromoen-doscopy (FICE) (Fujifilm, Tokyo, Japan). Of all VCE
techniques, NBI has been studied most frequently for assessing disease activity in IBD patients. This pre- processing technique uses light of specific green (540 nm) and blue (415 nm) wavelengths to enhance detail on the mucosal surface Figure 1B. Because the peak light absorption of hemoglobin occurs at these wavelengths, the capillary network and blood vessels on the mucosal surface appear dark and are more easily recognized [44]. Contrastingly, the post-processing iSCAN offers the poten-tial of assessing the large colonic areas by incorporating a multimodal approach in enhancing the structures of the mucosal surface, while retaining its natural color [45].
3.3.1. Assessment of endoscopic remission
A study from 1997 concluded that DCE was not useful in assessing mucosal inflammation severity and monitoring treatment responses in IBD [46]. Several studies reported a significant correlation between vessel density in both normal and inflamed areas and NBI findings. However, all studies had small sample sizes of up to 52 patients [47,48]. Studies of comparable sizes that combine the iSCAN tech-nology with zoom endoscopy indicate that iSCAN shows promise in the assessment of mucosal healing, perhaps prompting further validation in larger clinical studies [49,50].
3.3.2. Dysplasia detection
In the SCENIC guidelines, DCE with targeted biopsies is recom-mended as the method for dysplasia identification in surveil-lance of IBD patients [51]. However, the relevant evidence is contradictory. For example, a meta-analysis of three (total n = 538) randomized controlled clinical trials (RCTs) (2015–-2018) did not demonstrate a significant benefit of DCE over HD WLE (RR 1.36; 95% CI: 0.84–2.18) [52], while a more recent systematic review that included only one additional RCT con-cluded that DCE is superior (RR: 1.38; 95% CI: 1.02–1.88) [53]. This latter conclusion is supported by a recent RCT [54]. Nevertheless, this technique is hampered by several limita-tions such as the need for operator training, bowel cleanliness and diminished usefulness when active inflammation is pre-sent [55]. Multiple systematic reviews have assessed the role of aforementioned techniques of VCE in detection of dysplasia in IBD patients; all stating that VCE has no additional value to HD WLE [56–58].
Figure 1. Surveillance examinations using chromoendoscopy in patients with longstanding, quiescent UC. Two flat lesions in colon transversum are visualized with dye-based chromoendoscopy using indigo carmine (A) and a small, slightly elevated lesion in sigmoid is observed with Narrow Band Imaging (B).
4. Magnified endoscopic imaging
4.1. Confocal laser endomicroscopy (CLE)
Confocal laser endomicroscopy (CLE) is a probe-based modal-ity (Mauna Kea Technologies, Paris, France) that enables for an up to thousand-fold magnified visualization of the targeted mucosa. This allows clinicians to make an in vivo assessment of histology. To illuminate the tissue, CLE uses a low-powered blue laser that is focused at a specific depth; it detects only the reflected light that passes through the pinhole confocal aperture. Prior to the procedure, fluorescent agents are admi-nistered topically (cresyl violet or acriflavine) or intravenously (fluorescein) [59].
Despite its potential benefits, CLE imaging also has several disadvantages. Firstly, confocal imaging is still indicated only after mucosal aberrances are seen during WLE, and only the aberrant mucosal area is visualized even though the colon may be extensively inflamed. Secondly, CLE is a relatively costly modality compared to currently used endoscopic tech-niques while it is not reimbursed in most countries. Lastly, it requires training the operator in image interpretation. It is therefore not widely adopted in clinical practice despite the availability of a commercial system.
4.1.1. Assessment of endoscopic remission
CLE can probe to a depth of up to 250 μm in the mucosa and is, therefore, able to detect multiple aberrations on a microscopic scale. Thus, CLE can visualize microscopic fea-tures which can otherwise only be observed during histo-pathological examination and contribute to the differentiation between CD and UC. However, CLE is unable to assess inflammation beyond the mucosal layer due to its limited maximum penetration depth [60]. Impaired intestinal permeability as observed by CLE is increasingly associated with ongoing bowel symptoms in symptomatic IBD patients with macroscopic remission [61]. Moreover, patients with UC in clinical and endoscopic remission may still express histolo-gical inflammatory activity that can be detected by CLE and evaluated by image-analysis systems [62]. This activity has also been shown to be predictive for a clinical relapse within 12 months in IBD patients with either clinically or endoscopi-cally inactive disease [63,64]. Furthermore, it has been observed that CLE findings are superior to clinical and endo-scopic assessment alone in predicting disease progression [65]. However, a prospective trial in UC patients concluded that additional CLE assessment did not yield a more accurate prediction of histological healing than careful examination with both HD WLE and iSCAN [66]. Lastly, CLE is less feasible for a quick assessment of treatment response after treatment escalation in IBD [67].
4.1.2. Detection of dysplasia
CLE has also been used as part of the surveillance strategy for CAC in IBD, yielding mixed results. Kiesslich et al. reported improved detection of dysplasia by combining chromoendo-scopy with CLE, but Freire et al. found no improvement in detection rates compared to WLE [68,69]. A more targeted strategy was enabled by additional endomicroscopic analysis
after identification of macroscopically suspect lesions, overall requiring fewer biopsies due to the implementation of CLE [68–70]. In one study, CLE assessment led to an average prolongation of examination time of 20 minutes, while Van den Broek et al. calculated an additional 30 to 40 minutes [69,70]. In a meta-analysis of nine studies, CLE had a pooled sensitivity of 87% and pooled specificity of 94% for differen-tiating neoplastic from non-neoplastic lesions in IBD [71].
4.2. Endocytoscopy (EC)
Endocytoscopy (EC) (Olympus, Tokyo, Japan) is based on the principle of contact light microscopy. It uses a visible light source and a high-power, fixed-focus objective lens that pro-jects images onto a charge-coupled device. It enables real time in vivo visualization of various features of the superficial mucosa at magnifications ranging from 520 to 1100 [72]. To assess cytological features, staining the colonic mucosa with a mixture of 0.05% crystal violet and 1% methylene blue is recommended [73]. Data have shown that EC is reliable for assessing the presence of histological inflammatory activity in patients with IBD [74,75]. Furthermore, a recently developed CAD system has enabled the fully automatic identification of persistent histologic inflammation on EC images associated with UC [76]. Subsequent research has demonstrated that endocytoscopic stratification can be predictive for relapse in patients with mildly to moderately active UC [77]. In case of reports, EC was successfully used for adenoma identification during the surveillance of an IBD patient [78,79]. Although these data suggest a promising role for EC in IBD, caution is warranted because this technique has essentially the same disadvantages as CLE. Moreover, less clinical experience in IBD has been acquired with EC compared to CLE Table 1.
5. Molecular imaging
In molecular imaging, pathological tissue is visualized by using various kinds of substrates – such as antibodies, small mole-cules, or peptides – to selectively label proteins that are over-expressed in cells [80]. By linking fluorescent dyes to the substrate, a tracer can be created with a benign character that allows repeated high-resolution imaging [81].
Tracers using dyes that fluoresce in the near-infrared (NIR) spectrum (wavelength: 700 nm to 900 nm) are less influenced by absorption and scattering compared to the visible light spectrum (wavelength: 300 nm to 700 nm). NIR imaging can visualize GI tissue up to a depth of 1 cm. In molecular imaging, dyes that fluoresce in the NIR spectrum are therefore favored over those that fluoresce in the visible spectrum [82–84]. However, since the human eye is unable to perceive light in the NIR spectrum, special camera systems are required to detect the emitted fluorescence signal [85]. Tracers can be administered systemically or applied topically onto the tissue of the digestive tract. For intravenous administration of the tracer, additional knowledge is required for timing of the administration and optimal dosing of the probe to maximize target uptake and minimize background uptake [80]. Subsequently, intravenous administration can closely approx-imate distribution throughout the ileocolonic tissue and reveal
drug engagement with target tissue, whereas topical admin-istration provides information only for a small area that is sprayed based upon its macroscopic appearance. Therefore, the fluorescent signal could enable a targeted biopsy strategy even if no macroscopic abnormalities are visible at the colonic surface, thereby avoiding sampling error.
Since molecular imaging is still under research, financial investment, and cooperation between various disciplines (technicians, pharmacists, clinicians) are necessary for further development. Prior to clinical implementation of tracers, extensive preclinical studies and large-scale clinical trials need to be conducted which are costly and time-consuming. Secondly, consensus regarding standardization of imaging procedures, image interpretation, and quantification of the fluorescent signal needs to be achieved.
5.1. Fluorescent molecular endoscopy (FME)
Incorporation of molecular imaging in wide-field endoscopes enables fluorescent molecular endoscopy (FME) [81]. As demonstrated by several preclinical and clinical studies, FME is an interesting approach for real-time in vivo identification and visualization of the molecular composition of inflamma-tion and dysplasia in IBD Figure 2.
5.1.1. Patient stratification and treatment monitoring
The first clinical trial using fluorescent molecular imaging to predict response to treatment in patients with IBD was per-formed in 2014 by Atreya et al. In 25 CD patients, fluorescein isothiocyanate (FITC) labeled adalimumab was sprayed onto the macroscopically most inflamed region of the bowel during colonoscopy. In vivo confocal imaging of the tracer identified membrane-bound TNF-expressing (m-TNF) cells in the mucosa. Patients with a minimum of 20 m-TNF-expressing cells per confocal image had a significantly higher response rate to adalimumab therapy than patients with fewer m-TNF- expressing cells (92% vs. 15%). Furthermore, they exhibited a sustained clinical and endoscopic remission, while four patients in the other group had to undergo surgery [86]. Comparably, CLE imaging of FITC-labeled vedolizumab pre-dicted response to vedolizumab therapy in five CD patients with active mucosal inflammation who were not responding
to anti-TNFa therapy. Prior to vedolizumab therapy, two responders were identified as positive for α4β7 integrin- expressing mucosal cells, whereas in three non-responders no α4β7 integrin expression was observed [87].
The aforementioned studies indicate that FME could sub-stantially improve upfront patient stratification to therapy with biologicals and subsequent dose optimization in IBD Figure 3. In addition, FME could transform treatment monitoring, as the molecules that are targeted and visualized by the fluorescent tracer can act as ‘effect sensors’ in assessing response to treatment (e.g. macrophages and T cells). A molecular imaging study in two groups of mice with UC (treated and untreated) investigated a folate-receptor-targeted NIR dye that accumu-lates in activated macrophages at sites of inflammation (OTL0038). Treated mice showed lower OTL0038 uptake in their colons than the diseased control group both prior to and during the improvement in their clinical condition at the end of therapy [88]. These findings could assist in drug devel-opment by determining the optimal dose during phase I/II studies and could improve understanding of drug activity in IBD patients [80,81]. This information could enhance persona-lized therapeutic management in the future. Subsequently, FME could contribute to achieving the newly proposed ther-apeutic endpoint of ‘disease clearance.’ This concept refers to the combined state of symptomatic remission (patient- reported outcomes) and mucosal healing (endoscopic and histological healing). Although the definition needs further validation and standardization, treat-to-clear is hypothesized to be the ultimate goal in achieving improved outcomes [89].
5.1.2. Detection of dysplasia
Despite the complexity of the genetic background and the expected confounding interference of gut microbiota that cause CAC, its first reliable biomarkers for FME have been identified [90]. Several specific probes labeled with a fluorescent NIR dye to detect CAC were developed and used successfully in murine models for targeting overex-pressed proteins, such as γ-glutamyltranspeptidase (GGT), cathepsin, matrix metalloproteinases-2/-9 (MMP-2/-9) and endothelin-A-receptors (ETAR) [91–95]. Furthermore, one ex vivo pilot study has been performed in which the fluorescein- conjugated VRPMPLQ peptide was applied topically and
Table 1. Overview of various aspects of magnified endoscopy.
Features Confocal laser endomicroscopy Endocytoscopy Operational
mechanism
● Reflection of a low-powered blue laser light through a confocal pinhole.
● Magnified visualization of a factor of 1000x.
● Plane of depth of up to 250 μm.
● Application of topical or intravenous staining.
● Principle of contact light microscopy using a high-power fixed focus lens.
● Magnified visualization of a factor of 520 up to 1100x.
● Plane of depth of up to 50 μm. ● Application of topical staining. Current position ● Commercial system is available.
● Not widely adopted in clinical practice in IBD.
● Commercial system is available.
● Clinical experience in IBD is less compared to CLE.
Opportunities ● In vivo assessment of histological healing and CAC identification. ● In vivo assessment of histological healing and CAC identification. Limitations ● Operator training for procedural competence and image
interpretation required.
● Small field of imaging.
● Imaging indicated by WLE macroscopic appearance.
● Persistent sampling error.
● Operator training for procedural competence and image interpretation required.
● Small field of imaging.
● Imaging indicated by WLE macroscopic appearance.
● Persistent sampling error.
examined with CLE, resulting in demarcation of dysplastic lesions in the excised colon of patients [96]. Moreover, FME has been used successfully as a ‘red flag’ technique for polyp detection in patients with Lynch syndrome and for dysplasia detection in patients with Barrett’s esophagus [97,98]. These results suggest a promising role for FME in surveillance for dysplasia in IBD patients with longstanding colitis.
5.2. MDSFR/SFF spectroscopy
To optimize the potential benefits of FME in IBD, the implemen-tation of multiple-diameter single-fiber reflectance (MDSFR) and single-fiber fluorescence (SFF) spectroscopy has shown promis-ing results. This approach can accurately quantify the intrinsic fluorescent signal by correcting for tissue properties such as the
degree of tissue absorption and scattering [99]. Consequently, this probe-based modality is able to measure the local drug concentration in the tissue, as this is proportionally correlated to the amplitude of the fluorescence signal. Hence, quantitative FME allows for the improved assessment of response to therapy, as established in patients with locally advanced rectal carcinomas that were treated with neoadjuvant chemoradiotherapy [100].
6. Label-free modalities
Moving beyond surface-only morphology and tissue discolora-tions, imaging techniques that visualize surface and subsur-face molecules in IBD pathophysiology have become increasingly interesting to researchers as well as clinicians. In contrast to molecular imaging, optical coherence tomography
Figure 2. Conceptual overview of previously investigated and potential targets for fluorescent molecular endoscopy in the microenvironment of inflammatory and dysplastic tissue in IBD.
Figure legend: Endothelin A Receptor (ETAR), Folate beta receptor (Fβ R), γ-Glutamyltranspeptidase (GGT), Interleukin 12/23 receptor (IL12/23 R), Interleukin 12/23 (IL12/23), Myeloid Derived Suppressor Cell (MDSC), Matrix Metalloproteinas-2/-9 (MMP-2/-9), membrane-bound Tumor Necrosis Factor alpha (mTNF), soluble Tumor Necrosis Factor alpha (sTNF).
Figure 3. Molecular imaging in IBD activity assessment.
Figure legend: Various pro-inflammatory agents result in IBD immunopathology, leading to heterogeneity between patients. These agents can be targeted by specific biologicals, which can also be linked to fluorescent dyes (I). The fluorescence enables the colon in IBD to be screened for its immunological phenotype. This could guide treatment decisions, as fluorescent signals can be quantified, and the presence of molecular targets can be measured over time as ‘effect sensors’ for response to therapy (II). Ultimately, a more tailored therapy in IBD can be achieved when tools for prediction of therapy response and treatment monitoring become available (III).
(OCT) and Raman spectroscopy (RS) are label-free modalities that do not require tracers Table 2.
6.1. Optical coherence tomography (OCT)
In optical coherence tomography (OCT) an image of the mucosal structure is generated with a system that is analo-gous to ultrasonography, but uses near-infrared light (750–-1300 nm) instead of sound waves. With OCT, tissue can be visualized to depths between 2 mm and 3 mm [101]. This modality has been studied in a few prospective cohorts with up to 70 IBD patients. These studies indicated that OCT could potentially be used to distinguish between inflamed and nor-mal tissue and to differentiate between hyperplastic polyps and adenomas [101–103]. Most studies conducted with OCT in the GI system have been performed in patients with Barrett’s esophagus, and they reported several challenges with this modality. Firstly, OCT does not allow co-registration with white light imaging, which causes difficulties with obtaining biopsies from the imaged locations [101]. Secondly, training practitioners to interpret OCT images is very time-consuming, and no standardized methods are available. Thirdly, real-time volumetric scans from inside the GI tract cover only 6 cm in length per image; if this modality is used for the colon, it would prolong the duration of the endoscopy substan-tially [104].
6.2. Raman spectroscopy (RS)
Raman spectroscopy (RS) is based upon the inelastic scattering of photons that results from the Raman effect, in which the vibrational, rotational, or electronic energy of a molecule decreases or increases after excitation [105]. As each molecule has a characteristic Raman signal, Raman spectra can be used to identify the molecular ‘fingerprint’ of a tissue [106]. Because RS can accurately discriminate Crohn’s disease from ulcerative colitis ex vivo, the technique has been used in patients with IBD [107,108]. Several studies have also shown the potential of
probe-based endoscopic RS as a diagnostic tool for assessing the presence of mucosal healing or inflammation in IBD, both ex vivo and in vivo [109–111]. Due to the weakness of its signal, there are various disadvantages ascribed to RS, such as low sensitivity, long acquisition time and slow imaging by point scanning. Together, these make it so that video rate imaging is almost impossible [112]. By using stimulated Raman scattering (SRS), the Raman signal can be detected in real-time at faster acquisition rates and with better resolution and sensitivity [105]. One study incorporated SRS in a label- free multimodal approach and identified a set of quantitative features that accurately predict disease activity [113]. Despite the promising results of (S)RS in detecting premalignant or fully malignant tumors of the GI tract, no studies investigating the feasibility of this technique for CAC detection have yet been published.
7. Conclusion
Endoscopic imaging of the ileocolonic mucosa constitutes the cornerstone in the management of IBD patients. Firstly, ileo-colonoscopy is performed in order to assess severity, extent, and localization of inflammation and monitor its activity dur-ing evaluation of response to treatment. Endoscopic remission has been favored as therapeutic endpoint in the treat-to- target approach. Secondly, patients are stratified to certain intervals for surveillance endoscopy to detect colitis- associated carcinoma in an early stage. In current clinical practice, high-definition white light endoscopy and dye- based chromoendoscopy are mainly used to perform endo-scopic imaging of the ileocolonic mucosa. Commercial sys-tems of confocal laser endomicroscopy and endocytoscopy are also available although these magnification techniques are not widely clinically adopted yet. Lastly, techniques that allow submucosal visualization in addition to mucosal assess-ment are emerging. However, fluorescent molecular endo-scopy and label-free modalities are still in an experimental phase.
Table 2. Overview of various aspects of fluorescent molecular endoscopy and label-free modalities.
Features Fluorescence molecular endoscopy Optical coherence tomography Raman spectroscopy Operational
mechanism
● Principle of molecular imaging using a NIR wide-field endoscope.
● Visualization of mucosal and submucosal biomarkers.
● Application of intravenous or topical administration of
fluorescent tracers.
● Principle analogs to
ultrasonography, using NIR light instead of sound waves.
● Visualization of mucosal and
submucosal structures.
● No tracer administration needed.
● Detection of the Raman signal of biomarkers after excitation of photons by incident NIR light.
● Visualization of the ‘molecular fingerprint’
of the tissue.
● No tracer administration needed. Current
position
● No commercial endoscopic system is available; financial and technical support are required.
● First (pre-) clinical trials completed.
● Commercial endoscopic system is available.
● First studies for clinical
applica-tion in IBD completed.
● No commercial endoscopic system is available.
● First studies for clinical application in IBD
completed. Opportunities ● Real time in vivo assessment of targeted biomarkers
allowing prediction and monitoring of response to treatment and CAC detection.
● In vivo assessment of histological healing and CAC identification.
● In vivo assessment of histological healing and CAC identification.
Limitations ● Limited reliable biomarkers are yet identified.
● Operator training for interpretation of the fluorescent sig-nal required.
● No co-registration with white light imaging.
● Operator training for interpreta-tion of the OCT images required.
● Substantial prolongation of endoscopy.
● Various disadvantages related to the weakness of the Raman signal.
● WLE required for anatomical imaging.
● Advanced data analysis.
8. Expert opinion
These days, a significant scientific effort is being made to develop novel endoscopic mucosal imaging techniques that can improve both the assessment of inflammation activity and the detection of dysplastic lesions. Recent advances in endo-scopic evaluation of the ileocolonic surface have improved mucosal assessment for inflammatory activity and dysplasia detection. In particular, the introduction of HD WLE has improved endoscopic examination in IBD. Enhanced endo-scopic imaging for improved visualization of the vascular pat-tern and magnified endoscopy for visualization of architectural and cytological features has led to the introduction of the ‘optical biopsy.’ However, due in part to the increasing clinical use of costly biologicals with moderate response rates, there is an urgent need for reliable techniques that can visualize drug distribution throughout the inflamed tissue, help stratify patients before treatment initiation and monitor their response to treatment.
The first completed studies using molecular imaging with fluorescent-labeled antibodies and probes in IBD for stratifi-cation of patients and monitoring response to treatment have been promising. Moreover, several pre-clinical studies have shown that optical molecular imaging can accurately visualize colitis-associated cancer. First-in-human studies have recently established that dysplastic tissue in the GI tract can be visualized with fluorescent molecular endoscopy (FME). Based on these results, FME potentially enables personalized medicine in the upcoming 10 years. In this scenario, multiple antibodies can be conjugated to distinct fluorescent dyes with specific light spectra. These can be simultaneously visualized in the inflamed gut through the use of multispec-tral camera systems, and their concentration can be mea-sured through MDSFR/SFF spectroscopy. Therefore, a monoclonal antibody might be selected for a specific patient before treatment initiation, while its dosage can sub-sequently be titrated during response evaluation according to its mucosal concentration. In addition, this research might identify other biomarkers that are associated with sufficient drug penetration or response which can be more easily implemented in the clinic than FME. This would lead to a personalized strategy in treating IBD patients that increases therapeutic responsiveness among patients and therefore the likelihood of achieving mucosal healing. Moreover, other associated outcomes such as hospitalization rates and the number of surgical interventions could potentially decrease while the cost-effectiveness could improve. Additionally, sur-veillance strategies in IBD could be transformed into a more personalized approach as dysplastic cells can be targeted and reliably detected during colonoscopy, even in the inflamed mucosa. The use of FME in IBD needs to be tested and validated in further clinical studies. To do so, progress should be made in adaptation of fluorescence imaging within com-mercial endoscopes and fluorescent drugs should be made available.
In addition, we also expect an exciting future for new label-free imaging techniques in IBD. Optical coherence tomography and Raman spectroscopy have become interest-ing areas of research for current and future generations of
scientists. Both techniques can be complementary to HD WLE and reliably provide an in vivo assessment of histological healing as well as CAC detection. Eventually, the human factor will become the weakest link in incorporating all the additional parameters that are offered by these new techni-ques. In order to meet this complex interpretation, an impor-tant role is reserved for AI in perceiving all the information provided during examination. Next to the increasing number of monoclonal antibodies to treat the IBD patient, in the future, gastroenterologists will have access to a considerable variety of equipment from which an endo-scopic technique can be chosen. These techniques can be selected based upon the indication of lesion characterization in IBD patients such as the assessment of mucosal remission or the detection of dysplastic cells. Subsequently, both treat-ment and surveillance strategies transform from an empirical approach in personalized approaches tailored to the indivi-dual IBD patient.
Ultimately, successful clinical implementation of FME and label-free modalities necessitates investing in technical development, as well as a multidisciplinary approach between clinicians, technicians and pharmacists. Efforts to make these techniques clinically feasible will therefore require financial investments for technical refinement of the techniques, tracer development and the composing of training programs. To achieve proper operator training, such training modules can be time-consuming and are thus asso-ciated with additional costs. Even after successful fulfillment of the learning process, experience is necessitated before the interpretation of required images becomes instinctive and straightforward. Both of these challenges are still ser-ious hurdles that need to be overcome before widespread implementation of these novel endoscopic techniques into clinical practice can be achieved. Nevertheless, we expect that the clinical benefits and cost-effectiveness will out-weigh the expensive research and development phases of these techniques and that they will herald a new era in personalized IBD management.
List of abbreviations
AI: Artificial intelligence; Anti-TNFa: Anti-tumor necrosis factor α antibodies; CAC: Colitis-associated adenoma and carcinoma; CAD: Computer-aided detection; CD: Crohn’s disease; CE: Chromoendoscopy; CLE: Confocal laser endomicroscopy; CNN: Convoluted neural networks; DCE: Dye-based chromoen-doscopy; EC: Endocytoscopy; ETAR: Endothelin-A-receptor; FICE: Fujinon intelligent chromoendoscopy; FITC: Fluorescein isothiocynate; FMI/E:Fluorescent molecular imaging/endo-scopy; GGT: γ-glutamyltranspeptidase; GI: Gastrointestinal; HD: High definition; IBD: Inflammatory bowel disease; IL-12/ 23: Interleukin-12/23; JAK-1/3: Janus kinase 1/3; MDSFR/SFF: Multi-diameter single-fiber reflectance and single-fiber fluor-escence (spectroscopy); MMP-2/-9: Matrix metalloproteinases- 2/-9; mTNF:Membrane-bound TNF; NBI: Narrowband imaging; NIR: Near infrared; OCT: Optical coherence tomography; RCT: Randomized controlled trial; RS: Raman spectroscopy; SRS: Stimulated Raman Spectroscopy; TNFa: Tumor necrosis factor
α; UC: Ulcerative colitis; VCE: Virtual chromoendoscopy; WLE: White light endoscopy
Notes on contributions
JL, AW and WN prepared the draft of the manuscript. JL and AW per-formed the literature search, wrote the content of the manuscript and designed the figures. JL and AW contributed equally to this manuscript and share first authorship. RG contributed to the literature search and to the content of the manuscript. EF, GD and WN reviewed the manuscript and figures for important intellectual feedback. All authors read and approved the final version of the manuscript.
Declaration of interest
WB Nagengast served as advisory board member of the GlaxoSmithKline imaging hub. G Dijkstra received unrestricted research grants from Abbvie, Takeda and Ferring Pharmaceuticals and speakers’ fees from Takeda and Janssen Pharmaceuticals, and served as advisory board member for Mundipharma and Pharmacosmos. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.
Reviewer Disclosures
Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.
ORCID
Jouke J.H. van der Laan http://orcid.org/0000-0001-6727-715X
References
Papers of special note have been highlighted as either of interest (•) or of considerable interest (••) to readers.
1. Peyrin-Biroulet L, Sandborn W, Sands BE, et al. Selecting therapeu-tic targets in inflammatory bowel disease (STRIDE): determining therapeutic goals for treat-to-target. Am J Gastroenterol. 2015
Sep;110(9):1324–1338.
2. Bernstein CN. Treatment of IBD: where we are and where we are going. Am J gastroenterol. 2015 Jan; 110(1):114–126.
3. Danese S, Vuitton L, Peyrin-Biroulet L. Biologic agents for IBD: practical insights. Nat Clin Pract Gastroenterol Hepatol. 2015
Sep;12(9):537–545.
4. Targan SR, Hanauer SB, van Deventer SJ, et al. A short-term study of chimeric monoclonal antibody cA2 to tumor necrosis factor alpha for Crohn’s disease. Crohn’s Disease cA2 study group. N Engl J Med. 1997 Oct 9;337(15):1029–1035.
5. Reinisch W, Sandborn WJ, Hommes DW, et al. Adalimumab for induction of clinical remission in moderately to severely active ulcerative colitis: results of a randomised controlled trial. Gut.
2011 Jun;60(6):780–787.
6. Sandborn WJ, Gasink C, Gao LL, et al. Ustekinumab induction and maintenance therapy in refractory Crohn’s disease. N Engl J Med.
2012 Oct 18;367(16):1519–1528.
7. Sandborn WJ, Su C, Sands BE, et al. Tofacitinib as induction and maintenance therapy for ulcerative colitis. N Engl J Med. 2017 May 04;376(18):1723–1736.
8. Sandborn WJ, Feagan BG, Rutgeerts P, et al. Vedolizumab as induc-tion and maintenance therapy for Crohn’s disease. N Engl J Med.
2013 Aug 22;369(8):711–721.
9. Yanai H, Hanauer SB. Assessing response and loss of response to biological therapies in IBD. Am J gastroenterol. 2011 Apr; 106 (4):685–698.
10. Paramsothy S, Rosenstein AK, Mehandru S, et al. The current state of the art for biological therapies and new small molecules in inflammatory bowel disease. Mucosal Immunol. 2018;11 (6):1558–1570.
11. Gaujoux R, Starosvetsky E, Maimon N, et al. Cell-centred meta-analysis reveals baseline predictors of anti-TNFalpha non-response in biopsy and blood of patients with IBD. Gut. 2019
Apr;68(4):604–614.
12. Ungar B, Kopylov U, Yavzori M, et al. Association of vedolizumab level, anti-drug antibodies, and alpha4beta7 occupancy with response in patients with inflammatory bowel diseases. Clin Gastroenterol Hepatol. 2018 May;16(5):697–705.e7.
13. Yarur AJ, Jain A, Sussman DA, et al. The association of tissue anti-TNF drug levels with serological and endoscopic disease activ-ity in inflammatory bowel disease: the ATLAS study. Gut. 2016
Feb;65(2):249–255.
14. Galandiuk S, Rodriguez-Justo M, Jeffery R, et al. Field cancerization in the intestinal epithelium of patients with Crohn’s ileocolitis. Gastroenterology. 2012 Apr;142(4):855–864.e8.
15. Itzkowitz SH, Yio X. Inflammation and cancer IV. Colorectal cancer in inflammatory bowel disease: the role of inflammation. Am J Physiol Gastrointest Liver Physiol. 2004 Jul;287(1):G7–17. 16. Risques RA, Lai LA, Himmetoglu C, et al. Ulcerative
colitis-associated colorectal cancer arises in a field of short telo-meres, senescence, and inflammation. Cancer Res. 2011 Mar 1;71 (5):1669–1679.
17. Jess T, Gamborg M, Matzen P, et al. Increased risk of intestinal cancer in Crohn’s disease: a meta-analysis of population-based cohort studies. Am J Gastroenterol. 2005 Dec;100(12):2724–2729. 18. Jess T, Rungoe C, Peyrin-Biroulet L. Risk of colorectal cancer in
patients with ulcerative colitis: a meta-analysis of population-based cohort studies. Clin Gastroenterol Hepatol. 2012
Jun;10(6):639–645.
19. Eaden JA, Abrams KR, Mayberry JF. The risk of colorectal cancer in ulcerative colitis: a meta-analysis. Gut. 2001 Apr;48(4):526–535. 20. Canavan C, Abrams KR, Mayberry J. Meta-analysis: colorectal and
small bowel cancer risk in patients with Crohn’s disease. Aliment Pharmacol Ther. 2006 Apr 15;23(8):1097–1104.
21. Lutgens MW, Vleggaar FP, Schipper ME, et al. High frequency of early colorectal cancer in inflammatory bowel disease. Gut. 2008
Sep;57(9):1246–1251.
22. van den Broek FJ, Stokkers PC, Reitsma JB, et al. Random biopsies taken during colonoscopic surveillance of patients with longstand-ing ulcerative colitis: low yield and absence of clinical consequences. Am J Gastroenterol. 2014 May;109(5):715–722. 23. Zisman TL, Bronner MP, Rulyak S, et al. Prospective study of the
progression of low-grade dysplasia in ulcerative colitis using cur-rent cancer surveillance guidelines. Inflamm Bowel Dis. 2012;18 (12):2240–2246.
24. Choi CH, Rutter MD, Askari A, et al. Forty-year analysis of colono-scopic surveillance program for neoplasia in ulcerative colitis: an updated overview. Am J Gastroenterol. 2015 Jul;110(7):1022–1034. 25. Deepak P, Fowler KJ, Fletcher JG, et al. Novel imaging approaches
in inflammatory bowel diseases. Inflamm Bowel Dis. 2019 Jan 10;25 (2):248–260.
26. Iacucci M, Furfaro F, Matsumoto T, et al. Advanced endoscopic techniques in the assessment of inflammatory bowel disease: new technology, new era. Gut. 2018 Dec 22;68(3):562-572. 27. Reinink AR, Lee TC, Higgins PD. Endoscopic mucosal healing
pre-dicts favorable clinical outcomes in inflammatory bowel disease: a meta-analysis. Inflamm Bowel Dis. 2016 Aug;22(8):1859–1869. 28. Subramanian V, Ramappa V, Telakis E, et al. Comparison of high
definition with standard white light endoscopy for detection of dysplastic lesions during surveillance colonoscopy in patients with colonic inflammatory bowel disease. Inflamm Bowel Dis. 2013
Feb;19(2):350–355.
29. Annese V, Daperno M, Rutter MD, et al. European evidence based consensus for endoscopy in inflammatory bowel disease. J Crohn’s Colitis. 2013 Dec 15;7(12):982–1018.
30. Rubin DT, Ananthakrishnan AN, Siegel CA, et al. ACG clinical guide-line: ulcerative colitis in adults. Am J Gastroenterol. 2019;114 (3):384–413.
31. Patel V, Khan MN, Shrivastava A, et al. Artificial intelligence applied to gastrointestinal diagnostics: A Review. J Pediatr Gastroenterol Nutr. 2020 Jan;70(1):4–11.
32. Sharma P, Pante A, Gross SA. Artificial intelligence in endoscopy. Gastrointest Endosc. 2020;91(4):925-931.
33. Le Berre C, Sandborn WJ, Aridhi S, et al. Application of artificial intelligence to gastroenterology and hepatology. Gastroenterology. 2020 Jan;158(1):76–94.e2.
34. Ozawa T, Ishihara S, Fujishiro M, et al. Novel computer-assisted diagnosis system for endoscopic disease activity in patients with ulcerative colitis. Gastrointest Endosc. 2019 Feb;89(2):416–421.e1. 35. Stidham RW, Liu W, Bishu S, et al. Performance of a deep learning
model vs human reviewers in grading endoscopic disease severity of patients with ulcerative colitis. JAMA Network Open. 2019 May 3;2(5):e193963.
36. Takenaka K, Ohtsuka K, Fujii T, et al. Development and validation of a deep neural network for accurate evaluation of endoscopic images from patients with ulcerative colitis. Gastroenterology.
2020 Feb 11;158(8):2150–2157.
37. Bossuyt P, Nakase H, Vermeire S, et al. Automatic, computer-aided determination of endoscopic and histological inflammation in patients with mild to moderate ulcerative colitis based on red density. Gut. 2020 Jan 8;69(10):1778–1786.
• Disease activitity can accurately be assessed in UC patients by a computer-based score from endoscopic images using the ‘red density’ algorithm
38. Ahmad OF, Soares AS, Mazomenos E, et al. Artificial intelligence and computer-aided diagnosis in colonoscopy: current evidence and future directions. Lancet Gastroenterol Hepatol. 2019 Jan;4 (1):71–80.
39. Iakovidis DK, Koulaouzidis A. Automatic lesion detection in capsule endoscopy based on color saliency: closer to an essential adjunct for reviewing software. Gastrointest Endosc. 2014 Nov;80 (5):877–883.
40. van der Sommen F, de Groof J, Struyvenberg M, et al. Machine learning in GI endoscopy: practical guidance in how to interpret a novel field. Gut. 2020 May 11;69(11):2035–2045.
41. Corley DA, Jensen CD, Marks AR, et al. Adenoma detection rate and risk of colorectal cancer and death. N Engl J Med. 2014 Apr 3;370 (14):1298–1306.
42. Thakkar S, Carleton NM, Rao B, et al. Use of artificial intelligence-based analytics from live colonoscopies to optimize the quality of the colonoscopy exam in real-time: proof of concept. Gastroenterology. 2020;S0016–5085(20):30081.
43. Flynn AD, Valentine JF. Chromoendoscopy for dysplasia surveil-lance in inflammatory bowel disease. Inflamm Bowel Dis. 2018
Jun 8;24(7):1440–1452.
44. Gono K. Narrow band imaging: technology basis and research and development history. Clin Endosc. 2015 Nov;48(6):476–480. 45. Kodashima S, Fujishiro M. Novel image-enhanced endoscopy with
i-scan technology. World J Gastroenterol. 2010;16(9):1043–1049. 46. Matsumoto T, Kuroki F, Mizuno M, et al. Application of magnifying
chromoscopy for the assessment of severity in patients with mild to moderate ulcerative colitis. Gastrointest Endosc. 1997 Nov;46 (5):400–405.
47. Sasanuma S, Ohtsuka K, Kudo SE, et al. Narrow band imaging efficiency in evaluation of mucosal healing/relapse of ulcerative colitis. Endosc Int Open. 2018 May;6(5):E518–e523.
48. Danese S, Fiorino G, Angelucci E, et al. Narrow-band imaging endoscopy to assess mucosal angiogenesis in inflammatory bowel disease: a pilot study. World J Gastroenterol. 2010;16 (19):2396–2400.
49. Klenske E, Atreya R, Hartmann A, et al. Magnification endoscopy with optical chromoendoscopy shows strong correlation with
histologic inflammation in patients with inflammatory bowel disease. Endosc Int Open. 2019 Aug;7(8):E1018–e1026.
50. Honzawa Y, Matsuura M, Higuchi H, et al. A novel endoscopic imaging system for quantitative evaluation of colonic mucosal inflammation in patients with quiescent ulcerative colitis. Endosc Int Open. 2020;8(1):E41–E49.
51. Laine L, Kaltenbach T, Barkun A, et al. SCENIC international con-sensus statement on surveillance and management of dysplasia in inflammatory bowel disease. Gastroenterology. 2015 Mar;148 (3):639–651.e28.
52. Feuerstein JD, Rakowsky S, Sattler L, et al. Meta-analysis of dye-based chromoendoscopy compared with standard- and high-definition white-light endoscopy in patients with inflamma-tory bowel disease at increased risk of colon cancer. Gastrointest Endosc. 2019 Aug;90(2):186–195.e1.
53. Wan J, Wang X, Yang ZP, et al. Systematic review with meta-analysis: chromoendoscopy versus white light endoscopy in detection of dysplasia in patients with inflammatory bowel disease. J Dig Dis. 2019 Apr;20(4):206–214.
54. Alexandersson B, Hamad Y, Andreasson A, et al. High-definition chromoendoscopy superior to high-definition white-light endo-scopy in surveillance of inflammatory bowel diseases in a randomized trial. Clin Gastroenterol Hepatol. 2020 Apr 27;18 (9):2101–2107.
55. Soetikno R, Kaltenbach T, McQuaid KR, et al. Paradigm shift in the surveillance and management of dysplasia in inflammatory bowel disease (West). Digestive Endosc. 2016 Apr;28(3):266–273. 56. Iannone A, Ruospo M, Palmer SC, et al. Systematic review with
network meta-analysis: endoscopic techniques for dysplasia sur-veillance in inflammatory bowel disease. Aliment Pharmacol Ther.
2019 Oct;50(8):858–871.
57. Lv XH, Wang BL, Cao GW. Narrow band imaging for surveillance in inflammatory bowel disease: a systematic review and meta-analysis. J Clin Gastroenterol. 2019 Sep;53(8):607–615. 58. El-Dallal M, Chen Y, Lin Q, et al. Meta-analysis of virtual-based
chromoendoscopy compared with dye-spraying chromoendoscopy standard and high-definition white light endoscopy in patients with inflammatory bowel disease at increased risk of colon cancer. Inflamm Bowel Dis. 2020 Aug;26(9):1319-1329.
59. Kiesslich R, Goetz M, Angus EM, et al. Identification of epithelial gaps in human small and large intestine by confocal endomicroscopy. Gastroenterology. 2007 Dec;133(6):1769–1778. 60. Tontini GE, Mudter J, Vieth M, et al. Confocal laser endomicroscopy
for the differential diagnosis of ulcerative colitis and Crohn’s dis-ease: a pilot study. Endoscopy. 2015 May;47(5):437–443.
61. Chang J, Leong RW, Wasinger VC, et al. Impaired intestinal perme-ability contributes to ongoing bowel symptoms in patients with inflammatory bowel disease and mucosal healing. Gastroenterology. 2017 Sep;153(3):723–731.e1.
•• Confocal laser endomicroscopy enables visualization of micro-scopic mucosal inflammation in IBD patient with persistent bowel symptoms and endoscopic remission accordint to HD WLE.
62. Mace V, Ahluwalia A, Coron E, et al. Confocal laser endomicroscopy: a new gold standard for the assessment of mucosal healing in ulcerative colitis. J Gastroenterol Hepatol. 2015 Mar;30(Suppl 1):85–92.
63. Kiesslich R, Duckworth CA, Moussata D, et al. Local barrier dysfunc-tion identified by confocal laser endomicroscopy predicts relapse in inflammatory bowel disease. Gut. 2012 Aug;61(8):1146–1153. 64. Karstensen JG, Saftoiu A, Brynskov J, et al. Confocal laser
endomi-croscopy: a novel method for prediction of relapse in Crohn’s disease. Endoscopy. 2016 Apr;48(4):364–372.
65. Tontini GE, Mudter J, Vieth M, et al. Prediction of clinical outcomes in Crohn’s disease by using confocal laser endomicroscopy: results from a prospective multicenter study. Gastrointest Endosc. 2018
Jun;87(6):1505–1514.e3.
66. Iacucci M, Cannatelli R, Gui XS, et al. Assessment of endoscopic healing by using advanced technologies reflects histologic healing in ulcerative colitis. J Crohn’s Colitis. 2020;14(9):1282–1289.
