Original article

19

Authors

Mitsuru Kaise1, Yasuo Ohkura2, Toshiro Iizuka1, Ryusuke Kimura1, Kosuke Nomura1, Yasutaka Kuribayashi1, Akihiro Yamada1, Satoshi Yamashita1, Tsukasa Furuhata1, Daisuke Kikuchi1, Osamu Ogawa1, Akira Matsui1, Toshifumi Mitani1, Shu Hoteya1

Institutions

1 2

submitted 25. December 2013 accepted after revision 7. July 2014

Bibliography DOI http://dx.doi.org/ 10.1055/s-0034-1377965 Published online: 15.9.2014 Endoscopy 2015; 47: 19–25 © Georg Thieme Verlag KG Stuttgart · New York ISSN 0013-726X Corresponding author Mitsuru Kaise, MD, PhD Department of Gastroenterology Toranomon Hospital 2-2-2 Toranomon Minato-ku Tokyo 105-8470 Japan Fax: +81-3-35827068 [email protected]

Department of Gastroenterology, Toranomon Hospital, Tokyo, Japan Department of Pathology, Kyorin University School of Medicine, Tokyo, Japan

Background and study aim: Endocytoscopy (ECS) enables in vivo microscopic imaging, which allows analysis of mucosal structures at the cellular level; however, limited data are available on the validity of ECS in the stomach. The aim of this study was to evaluate the feasibility of ECS in the diagnosis of early gastric cancer. Patients and methods: Gastric lesions that were the targets of histopathological diagnosis by endoscopic submucosal dissection or biopsy specimen were prospectively enrolled and evaluated using a single charge-coupled device-integrated endocytoscope, following double staining with crystal violet and methylene blue. High grade ECS atypia was defined according to specific irregularities in gland structure and cell nuclei. The primary end point was the accuracy of ECS diagnosis for gastric cancer, using histopathological diagnosis as the gold standard.

Results: A total of 82 lesions were investigated, including 23 early gastric cancers, 10 gastric adenomas, and 49 non-neoplastic lesions. Ten lesions could not be clearly observed by ECS because of poor staining due to viscous mucus or plaque; thus, assessability rates with ECS were 88 % in total and 91 % for gastric cancer. High grade ECS atypia was observed in 86 % of assessable gastric cancers, but not in any cases of gastric adenomas or non-neoplastic lesions. The sensitivity, specificity, positive and negative predictive values of high grade ECS atypia as the criterion for the diagnosis of gastric cancer were 86 %, 100 %, 100 %, and 94 %, respectively. No serious complications occurred during or after the examinations. Conclusion: ECS is a clinically feasible modality to obtain in vivo histology, with high diagnostic accuracy in gastric cancer.

Introduction

ynx [6], and colon [7 – 9]. Limited data are available on the validity of ECS in the diagnosis of gastric disorders [10, 11]. The aim of this study, therefore, was to evaluate the feasibility of ECS in diagnosing early gastric cancer.

!

Real-time endoscopic assessment of histology (endoscopic optical biopsy) is a major goal in diagnostic innovation that would enable accurate endoscopic diagnosis comparable to histopathological diagnosis. Attaining this goal requires endoscopic imaging data that correspond accurately with the in vivo histology and that visualize crucial histopathological findings for cancer diagnoses in terms of structural and cellular atypia. Endoscopic magnifications of × 80 – 100 could produce optical images that detect structural atypia [1], and ultra-high magnification of × 400 – 1000 could produce images that show cellular atypia [2]. Endocytoscopy (ECS) is based on ultra-high magnification endoscopy, and visualizes gastrointestinal cells using intraprocedural staining, making it possible to differentiate between cancerous and non-cancerous tissue. To date, ECS observation has been applied in the esophagus [3 – 5], phar-

Patients and methods !

Patients and procedures The study was approved by the Ethics Committees of Toranomon Hospital, and was conducted in accordance with the Declaration of Helsinki (as revised in 1989). Patients who were scheduled for magnifying endoscopic work-up before or after gastric endoscopic submucosal dissection (ESD) were consecutively enrolled between 2010 and 2012 after providing written informed consent. ECS examination was conducted using a single charge-coupled device (CCD)-integrated endocytoscope (GIF-Y0002; Olympus Corp., Tokyo, Ja-

Kaise Mitsuru et al. Endocytoscopy for gastric cancer … Endoscopy 2015; 47: 19–25

This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.

Endocytoscopy is a promising modality with high diagnostic accuracy for gastric cancer

Original article

Fig. 1 Representative endocytoscopy (ECS) findings of non-neoplastic gastric mucosa, in which the gland lumen is well preserved and mucosal cells are regularly arranged as shown in the histological image (c) and schema (d). The ECS findings correspond to these histopathological features: wide gland lumen (a, blue arrows), or regular linear lumen (b, blue arrows) between glands lined with regularly arranged cells (green arrows). Unstained circular areas may also be observed (white arrows), which are compatible with goblet cells of intestinal metaplasia. a

b

c

d

a

d

b

Fig. 2 Representative pictures of high grade endocytoscopy (ECS) atypia. ECS atypia was defined by specific irregularities in gland structure and cell nuclei. In adenocarcinoma (b, c), cancer cells lose polarity and are irregularly arranged, which correspond with the findings of high grade ECS atypia: absence or fusion of lumen (a, d). If polarity and lining are further disordered, the gland structure disappears (e). Histopathologically, cancer cells show irregular nuclei (b), which are defined by three typical features: swelling, disarrangement, and heterogeneity of shape. These features of irregular nuclei clearly correspond with findings of high grade ECS atypia (a, c).

c

e

pan). An eligible gastric lesion destined for ESD or histopathological evaluation by biopsy was examined initially with whitelight endoscopy without magnification, followed by ultra-magnification by ECS using double staining with 0.05 % crystal violet and 0.5 % methylene blue. Because viscous mucus reduces the quality of the staining, mucus on a lesion of interest was carefully removed by flushing with a solution of water plus pronase and dimethicone before staining. The staining solution was applied to a lesion of interest using a spray catheter, and excess dye was

Kaise Mitsuru et al. Endocytoscopy for gastric cancer … Endoscopy 2015; 47: 19–25

removed by water flushing. The rinsing and staining procedure was repeated until satisfactory staining levels were obtained.

ECS findings ECS findings of non-neoplastic gastric mucosa showed either a " Fig. 1 a, indicated by blue arrows) or a regular linwide lumen (● " Fig. 1 b, blue arrows), between glands lined with ear lumen (● " Fig. 1 a, b, green arrows). In non-neoregular-appearing cells (● plastic gastric mucosa, the gland lumen is well preserved and

This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.

20

Fig. 3 Representative pictures of low grade endocytoscopy atypia, defined by mild irregularities in gland structure, lumen shortening and narrowing (a), or absence or fusion of part of the lumen (b).

" Fig. 1 c, d), and cells lining the mucosa are regularly arranged (● this morphology can be observed using ECS. Unstained circular " Fig. 1 a, white arrows), which areas are sometimes observed (● are compatible with those recently reported as goblet cells of intestinal metaplasia [11]. However, based on our experience, these findings are not always observed in gastric mucosa with goblet cells. ECS atypia was defined by specific irregularities in gland structure and cell nuclei. High grade ECS atypia was defined by any of the following features being consistently observed: lumen absence " Fig. 2 a, b), disappearance of gland structure (● " Fig. or fusion (● 2 c), or irregular nuclei showing three typical features – swelling, " Fig. 2 a, c). In adedisarrangement, and heterogeneity of shape (● " nocarcinoma (● Fig. 2 b, c), cancer cells lose polarity and are irregularly arranged, which are features that correspond with the findings of high grade ECS atypia; absence or fusion of the lumen is also evident in adenocarcinoma. If polarity and lining are further disordered, the gland structure disappears. Low grade ECS atypia was defined by mild irregularities in gland structure, lu" Fig. 3 a), or absence or fusion men shortening and narrowing (● " of part of the lumen (● Fig. 3 b). All biopsy and ESD specimens were fixed in 10 % formalin and embedded in paraffin, serially sectioned, and stained with hematoxylin and eosin. Lesions diagnosed as Category 3 low grade neoplasias, according to the revised Vienna classification [12], were designated as gastric adenomas. Lesions diagnosed as Category 4 neoplasias were designated as gastric cancers. Pathologists were blinded to the clinical information, including ECS findings. Absence or presence of ECS atypia was evaluated in each tested lesion immediately following the endoscopic examination. Correlations between ECS atypia and pathological results (gold standard) were evaluated. The primary end point was the accuracy of ECS diagnosis for gastric cancer using high grade ECS atypia as the criterion for cancer diagnosis. The 95 % confidence intervals (CIs) were calculated for each of the diagnostic indices.

Results !

A total of 82 lesions in 54 patients were evaluated, and of these, 23 were assigned as early gastric cancers (21 well differentiated, 2 poorly differentiated), 10 as gastric adenomas, and 49 as nonneoplastic lesions. The non-neoplastic lesions comprised 10 gastric ulcers, 10 gastric erosions, and 18 gastric protruded lesions including hyperplastic polyps and intestinal metaplasia. In addition, 11 areas of gastric mucosa surrounding gastric neoplasia were also examined in order to evaluate the extent of neoplasia " Table 1). The mean for precise endoscopic resection by ESD (● time of total endoscopic examination including ECS was 26.0 minutes. Approximately half of the examination time was for ECS observation. Although all enrolled lesions were stained and examined according to the study protocol, 10 lesions were poorly visible by ESC due to the poor staining of viscous mucus or dense ulcer plaques. As a result, 88 % of all lesions (72 /82) were assessable by ECS, with rates of 91 % (21 /23) for gastric cancer lesions and only 50 " Table 1). Gastric hyper% (2 /4) for gastric hyperplastic polyps (● plastic polyps secrete abundant viscous mucus, which is not easily removed by water flushing, and well-stained areas were very limited, even in assessable hyperplastic polyps. High grade ECS atypia was observed in 86 % of assessable gastric cancers (18 /21), but in none of the gastric adenomas or other non-neoplastic lesions. Low grade ECS atypia was observed in 10 % of gastric cancers, 78 % of gastric adenomas, and 2 % of non" Table 2 shows ECS findings categorized acneoplastic lesions. ● cording to whether lesions were cancerous (gastric cancer) or noncancerous (gastric adenoma and non-neoplasia) on pathological cancer diagnosis. Using high grade ECS atypia as a diagnostic criterion for gastric cancer, ECS diagnosis had sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of 86 % (95 %CI 75.9 % – 93.1 %), 100 % (95 %CI 95 % – 100 %), 96 % (95 %CI 88.3 % – 99.1 %), 100 % (95 %CI 95 % – 100 %), and 94 % (95 %CI 86.4 % – 98.5 %), respectively. No remarkable complications occurred during or after the ECS examinations. " Fig. 4. A representative case of small gastric cancer is shown in ● A tiny gastric depression was found in the antrum, but conven-

Kaise Mitsuru et al. Endocytoscopy for gastric cancer … Endoscopy 2015; 47: 19–25

21

This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.

Original article

Original article

n

Assessable by

Table 1 Findings from endocytoscopy of gastric lesions.

ECS atypia

ECS n

%

None

Low

High

Gastric cancer

23

21

91.3

1

2

Gastric adenoma

10

9

90.0

2

7

0

Non-neoplastic lesion

49

42

85.7

41

1

0

Hyperplastic polyp

18

4

2

50.0

2

0

0

14

12

85.7

12

0

0

Gastric erosion

10

8

80.0

8

0

0

Gastric ulcer

10

9

90.0

9

0

0

Non-neoplastic mucosa surrounding neoplasia

11

11

100.0

10

1

0

Total

82

72

87.8

44

10

18

Intestinal metaplasia

ECS, endocytoscopy.

Gastric cancer

No cancer (adenoma,

Total

non-neoplasia) High grade ECS atypia None or low grade ECS atypia Total

18

0

3

51

54

21

51

72

ECS, endocytoscopy.

Fig. 4

A small early gastric cancer clearly detected by endocytoscopy (ECS).

Kaise Mitsuru et al. Endocytoscopy for gastric cancer … Endoscopy 2015; 47: 19–25

18

Table 2 Findings or endocytoscopy for gastric cancer compared with noncancerous lesions.

This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.

22

23

This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.

Original article

Fig. 5 Comparison between histopathological findings and endocytoscopic images in the case of small early gastric cancer resected by endoscopic submucosal dissection.

Kaise Mitsuru et al. Endocytoscopy for gastric cancer … Endoscopy 2015; 47: 19–25

Original article

Fig. 6 A representative case of intramucosal gastric carcinoma. A red depressed in the antrum (a, c) was endocytoscopically observed as having high grade dysplasia (b), compared with the surrounding mucosa, which had no atypia (d). Histopathological findings from the resected specimen (e) were comparable to those of endocytoscopy images (f, g).

This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.

24

tional white-light observation did not show any definite findings " Fig. 4 a). After double staining with 0.05 % of gastric cancer (● crystal violet and 0.1 % methylene blue, magnifying endoscopic observation was performed using ECS. Magnification of approxi" Fig.4 b) demonstrated a depressed area covered mately × 100 (● with small mucosal pits, but could not produce a definitive diagnosis of cancer. Ultra-magnification at × 400 of the area surround" Fig. 4 c, green aring the depression showed linear lumens (● rows) between ridge-like glands composed of regularly arranged " Fig. 4 c, white arcell linings and small vessels in the center (● rows); the lesion was graded as no ECS atypia. However, in the tiny depression, absence or fusion of lumens was consistently ob" Fig. 4 d), compatible with high grade ECS atypia. Subseserved (● Kaise Mitsuru et al. Endocytoscopy for gastric cancer … Endoscopy 2015; 47: 19–25

quent ESD and pathological examination demonstrated category " Fig. 5). The 4.2 – well-differentiated adenocarcinoma in situ (● area assessed as non-atypic by ECS showed regularly lined glands " Fig. 4 c. In contrast, cancerous and lumens, corresponding to ● glands and irregularly-lined lumen were also observed, cor" Fig. 4 d that was assessed as high responding to the area in ● grade ECS atypia. A lesion diagnosed as intramucosal carcinoma is shown in " Fig. 6. The ECS findings of carcinoma (● " Fig. 6 c) were com● " Fig. 6 f). parable to those of histopathology (●

Original article

!

ECS has been reported as a safe and effective new endoscopic imaging technique that enables in vivo histology imaging and guided biopsies with high diagnostic accuracy in the esophagus and colon [13]. However, studies in the stomach have been limited to one report using the probe-type ECS [12]. Therefore, the current study is the first to show the feasibility of ECS using a single CCD-integrated endocytoscope for the diagnosis of gastric cancer. Good staining of the lesion of interest is essential for adequate ECS observation. Squamous mucosa and columnar mucosa, which are composed mainly of absorptive cells, show good staining with crystal violet and methylene blue. In contrast, gastric mucosa, which is mostly composed of secretory cells, is poorly stained because of the abundant viscous mucosa and inflammatory products. Indeed, a previous study reported that the accuracy of ECS diagnosis for gastric neoplasia was significantly low due to gastric mucus secretion [12]. To overcome this issue, complete flushing of the tissue and dye staining were performed at least twice in the current study. If the ECS image was still unsatisfactory, direct staining was done during ECS observation at full magnification by applying dyes through the endoscope channel. As a result, the overall assessability rate for gastric lesions of interest, including gastric cancers, was approximately 90 %. Adenocarcinomas typically show cancerous glands lined with compact and irregular cells, and characterized by irregularities in the gland structure including absence, narrowing, or fusion of lumens, or disappearance of the gland structure. Adenocarcinomas also show irregularities in cell nuclei, such as swelling, disarrangement, and heterogeneity of shape. Therefore, in the current study ECS atypia was defined using these microscopic features; results showed that high grade ECS atypia might be a feasible and valuable criterion for the diagnosis of gastric cancer, providing satisfactory overall accuracy. Therefore, ECS also has the potential to facilitate both diagnosis and patient management of early gastric cancer. Cumulative evidence has shown that magnifying endoscopy combined with narrow-band imaging (NBI) achieves significantly accurate diagnosis of early gastric cancer compared with conventional endoscopy [1, 14, 15]. Diagnosis using NBI magnification is based on the irregularity of gland structure, but not on the cellular irregularity due to the limitation of the magnification level. Thus, ultra-magnifying endoscopy, such as ECS and confocal endomicroscopy, may have an advantage as a diagnostic strategy. " Fig. 4 and ● " Fig. 5 was diagIndeed, a tiny depression as shown● nosed as non-cancer with NBI magnification but as cancer by ECS, suggesting a superior value of ECS diagnosis in a high risk population. However, ultra-magnifying endoscopy requires additional intervention, such as dye staining or injection, and is a longer procedure, as shown in this study. Therefore, further study is necessary to validate a true practical advantage or disadvantage of ECS compared with conventional endoscopy and NBI magnification. This pilot study has some limitations. First, one expert endoscopist performed the ECS observation, introducing subjectivity and bias based on personal experience and competence in recognizing morphological features and making a diagnosis. Therefore, studies are needed to verify whether other endoscopists, includ-

ing trainees, could attain the same diagnostic accuracy as that achieved in the current study. Second, part of the pathological information on assigned lesions was not blinded during the ECS diagnosis in this study, and therefore information bias might affect the judgment of ECS atypia. Third, use of initial white-light endoscopy diagnosis may have introduced bias that affected the assessment of ECS atypia. Therefore, further prospective studies targeting newly diagnosed lesions are required to confirm the feasibility of ECS in the diagnosis of gastric cancer. In conclusion, this pilot study showed that ECS is a clinically feasible modality to obtain in vivo histology with high diagnostic accuracy in gastric cancer, although approximately 10 % of target lesions were not assessable because of poor dye staining. Competing interests: None

Acknowledgment !

The authors thank Ms. Keiko Hayakawa for her helpful assistance.

References 1 Kaise M, Kato M, Tajiri H. High definition endoscopy and magnifying endoscopy combined with narrow band imaging in gastric cancer. Gastroenterol Clin North Am 2010; 39: 771 – 784 2 Goetz M, Malek NP, Kiesslich R. Microscopic imaging in endoscopy: endomicroscopy and endocytoscopy. Nat Rev Gastroenterol Hepatol 2014; 11: 11 – 18 3 Kumagai Y, Monma K, Kawada K. Magnifying chromoendoscopy of the esophagus: in-vivo pathological diagnosis using an endocytoscopy system. Endoscopy 2004; 36: 590 – 594 4 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: 891 – 895 5 Kumagai Y, Kawada K, Yamazaki S et al. Endocytoscopic observation of esophageal squamous cell carcinoma. Dig Endosc 2010; 22: 10 – 16 6 Shimizu Y, Takahashi M, Yoshida T et al. Endoscopic in vivo cellular imaging of superficial squamous cell carcinoma of the head and neck by using an integrated endocytoscopy system (with video). Gastrointest Endosc 2013; 78: 351 – 358 7 Kudo S, Wakamura K, Ikehara N et al. Diagnosis of colorectal lesions with a novel endocytoscopic classification – a pilot study. Endoscopy 2011; 43: 869 – 875 8 Neumann H, Vieth M, Neurath MF et al. Endocytoscopy allows accurate in vivo differentiation of mucosal inflammatory cells in IBD: a pilot study. Inflamm Bowel Dis 2013; 19: 356 – 362 9 Cipolletta L, Bianco MA, Rotondano G et al. Endocytoscopy can identify dysplasia in aberrant crypt foci of the colorectum: a prospective in vivo study. Endoscopy 2009; 41: 129 – 132 10 Eberl T, Jechart G, Probst A et al. Can an endocytoscope system (ECS) predict histology in neoplastic lesions? Endoscopy 2007; 39: 497 – 501 11 Chiu PW, Ng EK, To KF et al. Recognition of goblet cells upon endocytoscopy indicates the presence of gastric intestinal metaplasia. Dig Endosc 2014; 26: 52 – 56 12 Dixon MF. Gastrointestinal epithelial neoplasia: Vienna revisited. Gut 2002; 51: 130 – 131 13 Neumann H, Fuchs FS, Vieth M et al. Review article: in vivo imaging by endocytoscopy. Aliment Pharmacol Ther 2011; 33: 1183 – 1193 14 Kato M, Kaise M, Yonezawa J et al. Magnifying endoscopy with narrow band imaging achieves superior accuracy in the differential diagnosis of superficial gastric lesions identified with white light endoscopy: a prospective study. Gastrointest Endosc 2010; 72: 523 – 529 15 Ezoe Y, Muto M, Uedo N et al. Magnifying narrowband imaging is more accurate than conventional white-light imaging in diagnosis of gastric mucosal cancer. Gastroenterology 2011; 141: 2017 – 2025

Kaise Mitsuru et al. Endocytoscopy for gastric cancer … Endoscopy 2015; 47: 19–25

This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.

Discussion

25

Copyright of Endoscopy is the property of Georg Thieme Verlag Stuttgart and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.

Endocytoscopy is a promising modality with high diagnostic accuracy for gastric cancer.

Endocytoscopy (ECS) enables in vivo microscopic imaging, which allows analysis of mucosal structures at the cellular level; however, limited data are ...
1MB Sizes 0 Downloads 7 Views