Beyond the Lens: Advancing Clinical Practice with Image-Enhanced Endoscopy Part 2: Colon

Vol. 30, Issue 3 (September 2025)

 Kiran Josy Kanjamala, MD, DrNB
 Consultant, Department of Gastroenterology
 Lisie Hospital
 Kochi, India
 

 Shibi Mathew, MD, DM
 Senior Consultant, Department of Gastroenterology
 Lisie Hospital
 Kochi, India
 

 Mathew Philip, MD, DM
 Senior Consultant, Department of Gastroenterology
 Lisie Hospital
 Kochi, India
 

 Prakash Zacharias, MD, DM
 Senior Consultant, Department of Gastroenterology
 Lisie Hospital
 Kochi, India

 

Introduction

The following is the second installment of "Beyond the Lens: Advancing Clinical Practice with Image-Enhanced Endoscopy." To read part one, which covers the upper gastrointestinal tract, you may access the June issue here.

Clinical Applications of IEE: Colon

Worldwide, colorectal cancer (CRC) is one of the leading causes of cancer deaths. To reduce mortality from CRC, screening and early detection of adenomatous lesions or detection of CRC in their early stages and subsequent treatment by endoscopy plays a significant role. Neoplastic lesions such as adenomas, sessile serrated lesions (SSL), and intramucosal CRC are managed endoscopically as there is no extra-local metastasis. In case of submucosal invasive cancer, as there is a possibility of lymph node metastasis up to 10%, surgery is usually recommended. Submucosal invasive CRCs without the combined findings of (i) poorly differentiated adenocarcinoma, signet-ring cell carcinoma, or mucinous carcinoma, (ii) lymphovascular invasion positive, (iii) depth of submucosal invasion ≥1000 μm, and (iv) budding grade 2 or 3 at the site of deepest invasion, have a very low risk for lymph node metastasis based on pathological findings; and hence, can be considered for endoscopic treatment.¹ Therefore, it is imperative that intramucosal neoplasm or shallow submucosal invasive (T1a) CRC should be differentiated from deep submucosal invasive (T1b) CRC endoscopically, which is referred to as “depth diagnosis” of early CRC. The pit pattern classification proposed by Kudo et al.² is an endoscopic diagnostic method for colorectal lesions. The pit is the shape of the gland orifice of a colorectal lesion. It is observed using indigo carmine as a contrast stain or crystal violet staining and magnifying endoscopy. The pit pattern is classified into the following types, including subcategories and thus, a qualitative and depth diagnoses can be inferred.

Table 1.1 Kudo's pit pattern (Adapted from @drkeithsiau0

Type VI is further divided into two subcategories: VI low-grade and VI high-grade. Type VI low-grade pits appear as irregular arrangements and sizes of IIIL, IIIS, and IV pits; they are considered to histologically represent adenoma-cT1a. Type VI high-grade pits are characterized by a narrowed or irregular gland orifice, an unclear outline of the gland, and the appearance of non-structural pits with complete destruction of each pit orifice. Type VI high-grade and type VN are important indices for diagnosing cT1b CRC and are managed surgically.

In view of development of a number of classification systems based on the surface and vascular pattern of colonic lesions, Japan NBI Expert Team (JNET) introduced a unified JNET classification in 2015,³ which adopts both vascular and surface pattern findings and is classified into four types: type 1, type 2A, type 2B, and type 3.

Table 1.2 JNET classification⁴

Fig 1.1 A lesion the rectum with most of the areas appearing JNET 2B (blue arrow) with focal areas appearing JNET 3 (green arrow) suggestive of deep submucosal invasion

Lesions diagnosed as type 2A usually are T1a cancer and hence can be managed endoscopically. Lesions diagnosed as type 3 should be treated by radical surgery. Meanwhile, the positive predictive value of type 2B is not as high as that of type 2A and 3. Around 10%–20% of lesions diagnosed as type 2B turn out to be pathological T1b cancer after resection. Hence in type 2B lesions, a combined use of pit pattern diagnosis is necessary for more accurate invasion depth diagnosis.

The NBI International Colorectal Endoscopic (NICE) classification was proposed in 2009 and is based on the color, vessels, and surface pattern of the colorectal tumors observed on non-magnifying endoscopy.⁵

Table 1.3 NICE classification⁶

Fig 1.2 NICE type 1 polyp

Fig 1.3 NICE type 2 polyp

Fig 1.4 Polypoidal lesion in sigmoid colon on WLE and NBI which appeared predominantly JNET 2b with focal areas appearing JNET 3

Endocytoscopy

Endocytoscopy is a type of ultra-magnifying endoscopy that helps to visualize the cellular image and hence provides an accurate pathological prediction based on a histologic-equivalent image. The clinical applications of endocytoscopy are multifold. In the esophagus, it can be used for the early detection of squamous cell carcinoma. Endocytoscopic atypia (ECA) classification, reported by Inoue et al. is
based on the irregularity of cell nuclei. ECA is classified into five categories (ECA1– ECA5).⁷ Endocytoscopic diagnoses in accordance with the ECA classification correlated with histopathological diagnoses based on the biopsy of endoscopically resected specimens. In differentiating between non-malignant tissue and squamous cell carcinoma, the overall accuracy of EC was 82%. Compared with esophagus, the number of studies on EC of the stomach is limited since the stomach is a mucous-rich organ, and it is difficult to obtain a clear endocytoscopic image. Some studies have shown that EC can differentiate cancerous tissue from non-neoplastic lesions and can also predict Helicobacter pylori infection.

The most extensively investigated area in EC is for the detection of colorectal neoplasms. EC can visualize the cellular atypia of colorectal lesions. Subsequently, Kudo’s group established an endoscopic image classification of colorectal lesions (EC classification). The EC classification has three tiers: EC1 corresponds to non-neoplasia, EC2 corresponds to adenoma, and EC3 corresponds to cancerous tissue.⁸ Studies have shown that EC had an accuracy of 96.8% for diagnosing adenoma. EC is also useful in the detection of sessile serrated lesion, identification of tumor pathological grade, prediction of the invasive depth of colorectal cancer and distinguishing low-grade from high-grade adenoma polyps.

Fig 1.5 Endocytoscopy of the corresponding areas of the above polyp

In inflammatory bowel disease (IBD), particularly ulcerative colitis (UC), EC can be used to study the histological inflammation grade as well as colitis-associated neoplasia. Studies have also investigated the relationship between the EC findings and Matts’ histological grading for patients with UC and established the EC system score (ECSS).⁹ The ECSS includes three factors: the shape of the crypts, distance between the crypts, and vessels. Studies have shown that the ECSS showed a strong correlation with Matts’ histological grade. EC can also be used to identify the various infiltrated inflammatory cells such as neutrophils, eosinophils, and basophils. The intramucosal capillary/crypt (ICC) index is a scoring system based on the findings of EC with NBI, developed to evaluate the severity of inflammation in patients with UC. It was found that the ICC index could stratify the risk of clinical relapse of UC.

Fig 1.6 A case of ulcerative colitis with Mayo endoscopic score 3, on endocytoscopy showing crypt architectural distortion, reduction in crypt number and infiltration with inflammatory cells

Artificial Intelligence in IEE

In recent years, deep learning-based artificial intelligence (AI) has been applied across various medical fields. In gastrointestinal endoscopy, computer-aided detection (CADe) and diagnosis (CADx) assist endoscopists in identifying potential neoplastic lesions and distinguishing between non-cancerous and cancerous lesions. GI-Genius by Medtronic was the first commercially available CADe. In a recent
multicenter randomized controlled trial, COLO-DETECT by Seager et al., the ADR was 8·3% higher in CADe-assisted colonoscopy compared to standard colonoscopy (adjusted odds ratio 1·47 [95% CI 1·21–1·78], p<0·0001).10 EndoBRAIN (Cybernet System Corp.; Olympus Corp., Tokyo) is a CADx system that allows real-time identification of colorectal polyps based on crypt structure, cell nuclei, and micro vessels in endoscopic images. In a study by Barua et al., the sensitivity for detection of neoplastic polyps was 90.4% with EndoBRAIN compared to 88.4% with standard method (P=0.33).¹¹

Fig 1.7 AI in Endocytoscopy using EndoBRAIN for characterizing the polyp

The Clinical Context of IEE Techniques

Utilizing the appropriate techniques for different clinical settings is crucial due to the varying levels of specialized training and expertise required, as well as considerations of time and cost. Community gastroenterologists benefit from the ease of use of virtual chromoendoscopy techniques such as NBI and chromoendoscopy. These techniques are well established with extensive literature that defines
their use and limitations in different settings. They are user-friendly, widely available, require less intensive training, and are more time efficient. In contrast, advanced techniques such as CLE and endocytoscopy demand significant expertise and specialized training, and are more time-consuming and costly, making them better suited for academic tertiary care centers and research units where they can establish their role in surveillance and guidance for complex interventions. This strategic allocation ensures optimal patient care and resource utilization across different clinical environments.

Summary

IEE uses dyes, electronic, and optical methods to enhance contrast and delineate normal, non-neoplastic and neoplastic mucosa, thus aiding the detection, diagnosis, and treatment endoscopically, of precancerous lesions or early cancers of the esophagus, stomach, colon, and rectum. The use of Lugol's solution for patients at high risk for squamous cell carcinoma of the esophagus and the use of indigo carmine solution to diagnose early gastric cancer have made diagnosis at a very early stage possible. The use of NBI and other optically enhanced methods for the detection of gastrointestinal neoplasms, particularly in the colon, is well established and likely to expand as the technology innovations are added and AI is further refined. The role of the more advanced IEE techniques requires more robust clinical trials that include cost effectiveness to define their role for screening and surveillance.

References

1. Tanaka S, Kashida H, Saito Y, Yahagi N, Yamano H, Saito S, et al. Japan Gastroenterological Endoscopy Society guidelines for colorectal endoscopic submucosal dissection/endoscopic mucosal resection. Dig Endosc Off J Jpn Gastroenterol Endosc Soc. 2020 Jan;32(2):219–39.
2. Kudo S, Hirota S, Nakajima T, Hosobe S, Kusaka H, Kobayashi T, et al. Colorectal tumours and pit pattern. J Clin Pathol. 1994;47(10):880–5.
3. Sano Y, Tanaka S, Kudo SE, et al. Narrow-band imaging (NBI) magnifying endoscopic classification of colorectal tumors proposed by the Japan NBI Expert Team. Dig Endosc Off J Jpn Gastroenterol Endosc Soc. 2016;28(5):526–33.
4. Inoue H,  Kaga M,  Ikeda H, et al.  Magnification endoscopy in esophageal squamous cell carcinoma: A review of the intrapapillary capillary loop classification. Annals of Gastroenterology : quarterly publication of the Hellenic Society of Gastroenterology. 2015;28:41-48.
5. Hayashi N, Tanaka S, Hewett DG, et al. Endoscopic prediction of deep submucosal invasive carcinoma: validation of the narrow-band imaging international colorectal endoscopic (NICE) classification. Gastrointest Endosc. 2013;78(4):625–32.
6. Endoscopic Classification Review Group. Update on the Paris classification of superficial neoplastic lesions in the digestive tract. Endoscopy. 2005;37(6):570–8.
7. Inoue H, Sasajima K, Kaga M, et al. Endoscopic in vivo evaluation of tissue atypia in the esophagus using a newly designed integrated endocytoscope: a pilot trial. Endoscopy. 2006;38(9):891–5.
8. Kudo SE, Wakamura K, Ikehara N,et al. Diagnosis of colorectal lesions with a novel endocytoscopic classification - a pilot study. Endoscopy. 2011;43(10):869–75.
9. Maeda Y, Kudo S ei, Ogata N, et al. Endocytoscopic intramucosal capillary network changes and crypt architecture abnormalities can predict relapse in patients with an ulcerative colitis Mayo endoscopic score of 1. Dig Endosc. 2020;32(7):1082–91.
10. Seager A, Sharp L, Neilson LJ, et al. Polyp detection with colonoscopy assisted by the GI Genius artificial intelligence endoscopy module compared with standard colonoscopy in routine colonoscopy practice (COLO-DETECT): a multicentre, open-label, parallel-arm, pragmatic randomised controlled trial. Lancet Gastroenterol Hepatol. 2024;9(10):911–23.
11. Barua I, Wieszczy P, Kudo S ei, et al. Real-Time Artificial Intelligence–Based Optical Diagnosis of Neoplastic Polyps during Colonoscopy. NEJM Evid 2022;1(6):EVIDoa2200003. DOI: 10.1056/EVIDoa2200003