Diagnostic ability of magnifying endoscopy compared to biopsy examination for early gastric cancer prior to endoscopic submucosal dissection

doi:10.21203/rs.3.rs-3984922/v1

Background Biopsy-specimen examination is the gold standard for the preoperative histological diagnosis of early gastric cancer (EGC). However, few studies have compared the diagnostic accuracies of biopsy and magnifying endoscopy with narrow-band imaging or blue laser imaging (ME-NBI/BLI). Thus, we compared the diagnostic accuracy of biopsy specimens and ME-NBI/BLI to evaluate whether ME-NBI/BLI is a feasible preoperative diagnostic tool for EGC.

Methods This retrospective single-center study enrolled 185 patients who underwent gastric endoscopic submucosal dissection (ESD) between January and December 2018. The sensitivity and positive predictive value (PPV) of the histological diagnosis of ME-NBI/BLI and biopsy were evaluated. Logistic regression analysis was used to assess the risk factors for the misdiagnosis of biopsy specimens and ME-NBI/BLI.

Results In total, 158 patients with EGC were analyzed. Sensitivities of biopsy and ME-NBI/BLI were 1 and 0 for adenomas (P = 0.333), 0.693 and 0.971 for differentiated adenocarcinomas (P < 0.001), and 0.688 and 0.625 for undifferentiated adenocarcinomas (P > 0.999), respectively. PPVs of biopsy and ME-NBI/BLI were 0.077 and 0 for adenomas (P> 0.999), 0.960 and 0.958 for differentiated adenocarcinomas (P > 0.999), and 0.750 and 0.750 for undifferentiated adenocarcinomas (P > 0.999), respectively. The underdiagnosis rate for differentiated adenocarcinomas was significantly higher in biopsy examination than in ME-NBI/BLI (27.9% vs. 0%, respectively, P < 0.001).

Conclusion ME-NBI/BLI had a higher sensitivity than biopsy examination for the preoperative diagnosis of differentiated adenocarcinomas. Therefore, performing ME-NBI/BLI for these lesions may be preferable regardless of their diagnosis as non-cancerous lesions from biopsy specimens.

Endoscopic submucosal dissection (ESD) has been widely employed for treating early gastric cancer (EGC) due to its high rate of curative resection. The indications for ESD have been recently expanded to include EGC without lymph node metastasis [1]; however, these indications differ for differentiated and undifferentiated adenocarcinomas. Therefore, preoperative histological diagnosis is a crucial factor in determining the indication for ESD. Currently, evaluation of biopsy specimens is the gold standard for preoperative EGC histological diagnosis [1]. However, the diagnostic accuracy of biopsy examination is approximately 70% [2–7], and the final histopathology identified after endoscopic treatment often differs from the preoperative biopsy diagnosis, including non-tumor atypical glands or adenoma to adenocarcinoma.

Recently, magnifying endoscopy with narrow-band imaging (NBI) and blue laser imaging (BLI) (ME-NBI/BLI) has been shown to accurately diagnose EGC by identifying the microstructure and microvessels of the malignancy [8–12]. However, few studies have compared the diagnostic accuracies of biopsy and ME-NBI/BLI for the same specimen in EGC. In this study, we evaluated the diagnostic ability of both biopsy specimens and ME-NBI/BLI to elucidate whether ME-NBI/BLI is a feasible preoperative diagnostic tool for EGC.

Patients

This retrospective study was conducted at the University Hospital of the Kyoto Prefectural University of Medicine. Consecutive patients who underwent gastric ESD at the hospital between January and December 2018 were enrolled in this study. The study protocol was approved by the University Ethical Review Committee (No. ERB-C-1600) and was conducted in accordance with the Declaration of Helsinki of the World Medical Association. Informed consent to perform ME and gastric ESD was obtained from all patients enrolled in this study. Informed consent from patients enrolled in the study was obtained in the form of an opt-out option on the study website.

Patients who underwent both preoperative biopsy examination and ME prior to gastric ESD were eligible for enrollment. ESD was performed for adenomatous and cancerous lesions histologically diagnosed by biopsy and for lesions diagnosed as a non-tumor lesion by biopsy but as adenoma or adenocarcinoma by ME. Indication of ESD for adenocarcinoma was decided according to the Japanese gastric cancer treatment guidelines 2014 [13]. The exclusion criteria were lesions < 5 mm in size on optical inspection during preoperative endoscopy owing to a risk of disappearance after biopsy, non-cancerous lesions diagnosed using both biopsy and ME, Barrett’s esophageal adenocarcinomas on the esophagogastric junction, and special types of adenocarcinomas.

Study design

This study aimed to compare the diagnostic ability of biopsy examination and ME for the preoperative histological diagnosis of EGC. Histological diagnoses using biopsy specimens and magnifying endoscopic images obtained before gastric ESD were collected retrospectively. ME was performed by endoscopists to estimate the histological diagnosis. Based on the final pathological results after gastric ESD, we evaluated the sensitivity and positive predictive value (PPV) for non-tumor lesions, adenoma, differentiated adenocarcinoma, and undifferentiated adenocarcinoma.

Endoscopic systems and endoscopes

Magnified endoscopic images were obtained using a combination of LASEREO endoscopic systems and optical magnifying EG-L600ZW or EG-L600ZW7 endoscopes (Fujifilm Co., Tokyo, Japan), and a combination of EVIS LUCERA ELITE system and optical magnifying GIF-H260Z or GIF-H290Z endoscopes (Olympus Medical Co., Ltd., Tokyo, Japan). The structural enhancement function was maintained at the B8 and B6 levels in ME-NBI and ME-BLI, respectively. The color mode was fixed at level 1 for both ME-NBI and ME-BLI.

Evaluation of endoscopic findings

Endoscopic findings were independently evaluated from recorded images by each endoscopist without access to clinical and pathological data. The criteria for histological diagnosis using ME-NBI or ME-BLI were defined based on the microarchitecture of the lesions, including the microstructure and microvessels with a demarcation line (DL) [14]. Histological diagnosis was made by referring to the characteristic findings of microsurface (MS) and microvascular (MV) patterns as follows [15]. The presence of a MS pattern was endoscopically diagnosed as a differentiated adenocarcinoma. In the absence of a MS pattern, the presence of a polygonal or closed-loop type MV pattern was endoscopically diagnosed as a differentiated adenocarcinoma, whereas the presence of a opened-loop type MV pattern was endoscopically diagnosed as an undifferentiated adenocarcinoma (Fig. 1). The presence of a regular MS and MV pattern with a clear DL was endoscopically diagnosed as an adenoma. When a mixed-type adenocarcinoma was suspected using ME-NLI and ME-BLI, the histological type that occupied more than half of the entire lesion was determined as the predominant histological type on evaluation using images wherein the entire extent of the lesion was imaged. Some lesions that were diagnosed as noncancerous at the initial examination were reexamined multiple times for endoscopy when cancer was strongly suspected. The images used to make the histological typing diagnosis were the results of the initial examination. Any contradictions in the diagnosis of the same lesion among the two endoscopists were resolved on discussion to reach a consensus.

Participating endoscopists

Two endoscopists (O. D. and T. Y.), having performed more than 200 magnifying endoscopies at our institution, evaluated the recorded images for the histological diagnosis of the target lesions. None of the endoscopists had histological information on the lesions when evaluating the images.

Pathological diagnosis

Preoperative biopsy specimens were obtained in other hospitals or our institution. Histological diagnoses of the biopsy specimens were evaluated at each institution. ESD specimens were fixed with 10% formalin, sectioned into 2-mm-thick slices, and the histological types were defined by two expert clinical pathologists (M.K. and Y.M.), according to the Japanese Classification of Gastric Carcinoma proposed by the Japanese Gastric Cancer Association, 15th edition¹³. When the biopsy was classified as Group 1 or 2, the diagnosis was defined as a non-cancerous lesion; Groups 3 and 4/5 were defined as adenomas and adenocarcinomas, respectively. Moreover, the histological diagnosis of differentiated or undifferentiated adenocarcinomas was defined based on the dominant histology present in the biopsy or ESD specimens. Therefore, differentiated adenocarcinoma included pure differentiated-type and differentiated predominant mixed-type adenocarcinoma. Similarly, undifferentiated adenocarcinoma included pure undifferentiated-type and undifferentiated predominant mixed-type adenocarcinoma.

When the biopsy specimen was diagnosed as a non-tumor lesion, the requirement for re-biopsy was judged by the endoscopists. ESD was performed without re-biopsy when the lesion was diagnosed as cancer or adenoma using ME with high confidence. However, re-biopsy was performed when the ME diagnosis was of low confidence. In this study, biopsy diagnosis was evaluated using initially obtained specimens.

Statistical analysis

The PPVs of histological diagnosis by endoscopy and biopsy were compared using the Mann–Whitney U test for continuous variables and Fisher’s exact test for categorical variables. Furthermore, we employed univariate and multivariate logistic regression models to identify risk factors for misdiagnosis. Factors in the multivariate analysis were selected from factors with P values < 0.1 in the univariate analysis. A P < 0.05 was considered statistically significant. All statistical analyses were performed using SPSS software (version 26.0; IBM, Armonk, NY, USA).

A total of 185 consecutive patients who underwent ESD for lesions initially diagnosed as EGC were enrolled in this study, as shown in Fig. 2. Twenty-seven patients were excluded because of the following reasons: nine did not have preoperative biopsy, 10 presented lesions < 5 mm in size, one was diagnosed with a non-cancerous lesion one with Barrett’s esophageal adenocarcinoma, and six with a special type of gastric cancer including three gastric carcinomas with lymphoid stroma and three gastric adenocarcinomas of the fundic gland type. Ultimately, 158 patients were analyzed. Patient characteristics are summarized in Table 1. The male ratio was 74.7%, and the average age of the patients was 73 years (range: 26–73 years). The dominant locations and morphologies were the lower third (53.2%) and depressed type (61.4%), respectively. The final histological diagnoses comprised two adenomas (1.3%), 140 differentiated adenocarcinomas (88.6%), and 16 undifferentiated adenocarcinomas (10.1%). The differentiated adenocarcinomas comprised 119 pure differentiated adenocarcinomas (75.3%) and 21 mixed-type differentiated adenocarcinomas (13.3%). The undifferentiated adenocarcinomas comprised 10 pure undifferentiated adenocarcinomas (6.3%) and 6 mixed-type undifferentiated adenocarcinomas (3.8%).

Table 1

Characteristics of patients and lesions SD, standard deviation.
Characteristics		N = 158
Sex, n (%)
	Male	118 (74.7)
	Female	40 (25.3)
Age, median, years (range)		73 (26–73)
Tumor size, mean ± SD, mm		17.1 ± 13.8
Location, n (%)
	Upper third	16 (10.1)
	Middle third	58 (36.7)
	Lower third	84 (53.2)
Macroscopic type, n (%)
	Elevated	49 (31.0)
	Flat	12 (7.6)
	Depressed	97 (61.4)
Number of biopsy specimens, mean (range)		1.39 (1–4)
Final histology, n (%)
Adenoma			2 (1.3)
Differentiated adenocarcinoma			140 (88.6)
Pure differentiated			119 (75.3)
Mixed differentiated			21 (13.3)
Undifferentiated adenocarcinoma			16 (10.1)
Pure undifferentiated			10 (6.3)
Mixed undifferentiated			6 (3.8)

The relationship between histological diagnosis by biopsy and ME is summarized in Table 2. Sensitivities of biopsy and ME-NBI/BLI were 1 and 0 for adenomas (P = 0.333), 0.693 and 0.971 for differentiated adenocarcinomas (P < 0.001), and 0.750 and 0.750 for undifferentiated adenocarcinomas (P > 0.999), respectively. PPVs of biopsy and ME-NBI/BLI were 0.077 and 0 for adenomas (P > 0.999), 0.960 and 0.958 for differentiated adenocarcinomas (P > 0.999), and 0.750 and 0.750 for undifferentiated adenocarcinomas (P > 0.999), respectively. When differentiated and undifferentiated types were grouped into pure type and mixed type, the sensitivities of biopsy and ME-NBI/BLI were 0.689 and 0.983 for pure differentiated adenocarcinomas (P < 0.001), 0.714 and 0.904 for mixed-type differentiated adenocarcinomas (P = 0.238), 1 and 0.800 for pure undifferentiated adenocarcinomas (P = 0.474), and 0.333 and 0.667 for mixed-type undifferentiated adenocarcinomas (P = 0.567), respectively. PPVs of biopsy and ME-NBI/BLI were 0.976 and 0.962 for pure differentiated adenocarcinomas (P > 0.708), 0 and 0.909 for mixed-type differentiated adenocarcinomas (P = 0.444), 0.733 and 0.833 for pure undifferentiated adenocarcinomas (P > 0.999), and 1 and 0.500 for mixed-type undifferentiated adenocarcinomas (P > 0.999), respectively.

Table 2

Sensitivity and positive predictive value of biopsy examination and magnifying endoscopy
		Sensitivity (95% CI)		P	PPV (95% CI)		P
		Biopsy examination	Magnifying endoscopy	P	Biopsy examination	Magnifying endoscopy	P
Final histopathology
	Adenoma	1 (0.224-1.000)	0 (0-0.776)	0.333	0.077 (0.009–0.251)	0	> 0.999
	Differentiated adenocarcinoma	0.693 (0.609–0.768)	0.971 (0.928–0.992)	< 0.001	0.960 (0.609–0.768)	0.958 (0.928–0.992)	> 0.999
	Pure-diff.	0.689 (0.598–0.771)	0.983 (0.941–0.998)	< 0.001	0.976 (0.917–0.996)	0.962 (0.913–0.987)	0.708
	Mixed-diff.	0.714 (0.478–0.887)	0.904 (0.696–0.988)	0.238	0 (0-0.776)	0.909 (0.587–0.998)	0.444
	Undifferentiated adenocarcinoma	0.750 (0.505–0.898)	0.750 (0.505–0.898)	> 0.999	0.750 (0.505–0.898)	0.750 (0.505–0.898)	> 0.999
	Pure-undiff.	1 (0.741-1.000)	0.800 (0.444–0.975)	0.474	0.733 (0.449–0.922)	0.833 (0.516–0.979)	> 0.999
	Mixed-undiff.	0.333 (0.043–0.777)	0.667 (0.223–0.957)	0.567	1 (0.050-1)	0.500 (0.068–0.932)	> 0.999
CI, confidence interval; diff., differentiated.

Table 3 shows the concordance between the final diagnosis and the diagnosis by biopsy examination or ME. Among the 140 differentiated adenocarcinomas identified in the final diagnosis, biopsy examination diagnosed 15 of those as non-cancerous lesions, 24 as adenomas, 97 as differentiated and 4 as undifferentiated adenocarcinomas. On the other hand, ME diagnosed 2 differentiated adenocarcinomas among the two adenomas identified in the final diagnosis and 136 differentiated and four undifferentiated adenocarcinomas among the 140 adenocarcinomas identified in the final diagnosis. Therefore, the underdiagnosis rates of biopsy examination and ME were 27.9% (39/140) and 0% in differentiated adenocarcinomas, respectively, with a significant difference (P < 0.001).

Table 3

Concordance between final diagnosis and the diagnosis of biopsy examination or magnifying endoscopy.
		Diagnosis of biopsy examination/ magnifying endoscopy
		Non-tumor	Adenoma	Differentiated adenocarcinoma	Undifferentiated adenocarcinoma
Final histopathology
	Adenoma	0 / 0	2 / 0	0 / 2	0 / 0
	Differentiated adenocarcinoma	15 / 0	24 / 0	97 / 136	4 / 4
	Undifferentiated adenocarcinoma	0 / 0	0 / 0	4 / 4	12 / 12

Table 4 presents the number of correctly and incorrectly diagnosed lesions by biopsy examination and ME. The misdiagnosis rates for elevated type and differentiated adenocarcinoma were significantly higher for biopsy examinations, but the misdiagnosis rate for differentiated adenocarcinoma was significantly lower in ME.

Table 4

Number of correctly and incorrectly diagnosed lesions by biopsy examination and magnifying endoscopy IEE, image-enhanced endoscopy; NBI, narrow-band imaging; BLI, blue laser imaging.
		Biopsy examination			Magnifying endoscopy
		Correct	Incorrect	P	Correct	Incorrect	P
Tumor size				0.701			0.912
	≤ 20 mm	78	35		106	7
	> 20 mm	33	12		42	3
Tumor color				0.633			0.190
	Reddish	71	27		94	4
	Normal	19	11		28	2
	Discolored	21	9		26	4
Location				0.274			0.501
	Upper third	14	2		15	1
	Middle third	39	19		56	2
	Lower third	58	26		77	7
Macroscopic type				0.046			0.400
	Elevated	28	21		47	2
	Flat	10	2		12	0
	Depressed	73	24		89	8
Number of biopsies				0.579			> 0.999
	1	75	34		102	7
	2–4	36	13		46	3
IEE				0.728			0.194
	NBI	54	21		68	7
	BLI	57	26		80	3
Final histopathology				< 0.001			< 0.001
	Adenoma	2	0		0	2
	Differentiated adenocarcinoma	97	43		136	4
	Undifferentiated adenocarcinoma	12	4		12	4

Logistic regression analysis was performed to identify the risk of misdiagnosis using biopsy examination and ME. Univariate analysis revealed that elevated lesions and differentiated adenocarcinomas were risk factors for misdiagnosis in biopsy examinations (odds ratio [OR] 2.394 and 4.850, respectively, Table 5a). However, no independent risk factors were identified in multivariate analyses. Univariate analysis revealed that differentiated adenocarcinoma was a risk factor for misdiagnosis in ME (OR 0.059, Table 5b).

Table 5a. Logistic regression analysis related to misdiagnosis in biopsy examination.

		Univariate			Multivariate
		Odds ratio	95% CI	P	Odds ratio	95% CI	P
Lesion size
	> 20 mm	0.810	0.369–1.792	0.701
	≤ 20 mm	Ref
Lesion color
	Normal, pale	1.315	0.641–2.616	0.476
	Reddish	Ref
Morphology
	Elevated	2.394	1.134–4.804	0.023	2.039	0.987–4.211	0.054
	Flat, depressive	Ref			Ref
Location
	Upper or Middle third	0.884	0.442–1.724	0.731
	Lower third	Ref
Number of biopsies
	1	1.255	0.586–2.607	0.579
	2 ~ 4	Ref
IEE
	NBI	0.853	0.602–2.344	0.728
	BLI	Ref
Final histopathology
	Differentiated adenocarcinoma	4.850	2.311–9.680	< 0.001	1.313	0.400–4.308	0.654
	Others	Ref			Ref
CI, confidence interval; IEE, image-enhanced endoscopy; NBI, narrow-band imaging; BLI, blue laser imaging.

Table 5

b. Univariate analysis related to misdiagnosis in magnifying endoscopy.
		Univariate
		Odds ratio	95% CI	P
Lesion size, mm
	> 20	1.082	0.267–4.381	0.912
	≤ 20	Ref
Lesion color
	Normal, pale	2.611	0.705–9.664	0.151
	Reddish	Ref
Morphology
	Elevated	0.537	0.110–2.628	0.443
	Flat, depressive	Ref
Location
	Upper or Middle third	0.465	0.116–1.867	0.280
	Lower third	Ref
Number of biopsies
	1	1.052	0.272–3.876	> 0.999
	2–4	Ref
IEE
	NBI	2.745	0.683–11.028	0.155
	BLI	Ref
Final histopathology
	Differentiated adenocarcinoma	0.059	0.018–0.220	< 0.001
	Others	Ref
CI, confidence interval; IEE, image-enhanced endoscopy; NBI, narrow-band imaging; BLI, blue laser imaging.

In this study, we compared the sensitivity and PPV of biopsy and ME-NBI/BLI for the preoperative histological diagnosis of EGCs. This study revealed that ME-NBI/BLI had a higher sensitivity for differentiated adenocarcinomas than biopsies, whereas the PPV of ME-NBI/BLI was comparable to that of biopsy. Moreover, no cancerous lesions were underdiagnosed by ME-NBI/BLI. These results suggest that ME-NBI/BLI is an alternative to biopsy examination for the preoperative diagnosis of differentiated adenocarcinomas.

Even in cases in which lesions were diagnosed as Group 2 or 3 (atypical glands or adenoma) by biopsy examination, ME-NBI/BLI revealed an irregular MS or MV pattern with a DL, which may be diagnosed as a malignancy according to the vessel-plus-surface classification and magnifying endoscopy simple diagnostic algorithm for early gastric cancer (MESDA-G) [7] [16]. In our study, most of these lesions were finally diagnosed as differentiated adenocarcinomas. Therefore, ME-NBI/BLI could be preferable for differentiated adenocarcinomas, regardless of whether group 2 or 3 tumors are identified in biopsy examinations. Moreover, the preoperative diagnosis of ME for undifferentiated adenocarcinoma was as reliable as that of a biopsy examination. Further prospective studies are necessary to identify whether ME-NBI/BLI is feasible for application in preoperative histological diagnosis compared to biopsy examination [2].

The underdiagnosis rate for lesions diagnosed preoperatively as adenomas was reported to be 18.7–47.7%, with a pathologic diagnosis of adenocarcinoma after endoscopic resection [2] [3]; the underdiagnosis rate of biopsy examinations in our study was comparable to that in these reports. In general, the differential diagnosis between adenoma and adenocarcinoma is based on the atypia of tumor cells, including nuclear and structural atypia. We speculate that two reasons exist for the underdiagnosis rate of biopsy examinations. First, the tissue collected by the biopsy forceps may be selected as having mild-to-moderate atypia only from the adenocarcinoma lesion. Second, underdiagnosis may be caused by the misdiagnosis of structural atypia when the entire mucosal glandular cell cannot be evaluated owing to the small amount of surface tissue from the biopsy specimens. Nonaka et al. reported that tumor typing based on the NBI classification was useful for the endoscopic differentiation of gastric adenomas and well-differentiated adenocarcinomas [17]. Therefore, for an accurate biopsy diagnosis of an elevated lesion suspected to be an adenoma or adenocarcinoma, it is necessary to obtain a sufficient amount of biopsy specimens from the site of the irregular MS or MV pattern detected by ME.

This study had some limitations. First, this was a retrospective, single-center study that used collected images; therefore, selection bias could not be avoided. Second, we excluded non-tumor lesions diagnosed using both biopsy and ME. Therefore, it is difficult to analyze the specificity of biopsy examinations and ME. Third, the pathological diagnosis using biopsy specimens was performed by multiple pathologists. Different pathologists may use different diagnostic criteria for gastric neoplasms. Fourth, only two expert endoscopists evaluated the diagnoses of ME-NBI/BLI. Fifth, the number of undifferentiated adenocarcinoma cases was small, and the effectiveness of ME-NBI/BLI has not yet been verified for undifferentiated adenocarcinomas. Sixth, ME-NBI and ME-BLI images were evaluated together in our study. There are a few reports comparing the diagnostic ability of ME-NBI and ME-BLI [12]. When comparing the misdiagnosis risk with univariate analysis in this study, no significant difference was found between ME-NBI and ME-BLI. Thus, evaluating ME-NBI and ME-BLI images together was considered acceptable.

In conclusion, ME-NBI/BLI could have a higher sensitivity than biopsy examination for the preoperative diagnosis of differentiated adenocarcinomas. Therefore, ME-NBI/BLI may be preferable for these lesions regardless of their diagnosis as non-cancerous lesions from biopsy examinations.

Author contribution

Conception and design: Takuma Yoshida, Osamu Dohi.

Analysis and interpretation of date: Takuma Yoshida, Osamu Dohi, Mayuko Seya, Katsuma Yamauchi, Hayato Fukui, Hajime Miyazaki, Takeshi Yasuda, Tsugitaka Ishida, Naoto Iwai, Toshifumi Doi, Ryohei Hirose, Ken Inoue, Naohisa Yoshida, Kazuhiko Uchiyama, Takeshi Ishikawa, Tomohisa Takagi, Hideyuki Konishi, Yukiko Morinaga, Mitsuo Kishimoto.

Drafting of the article: Takuma Yoshida.

Critical revision of the article for important intellectual content: Takuma Yoshida, Osamu Dohi, Naoto Iwai.

Final approval of the article: Takuma Yoshida, Osamu Dohi, Mayuko Seya, Katsuma Yamauchi, Hayato Fukui, Hajime Miyazaki, Takeshi Yasuda, Tsugitaka Ishida, Naoto Iwai, Toshifumi Doi, Ryohei Hirose, Ken Inoue, Naohisa Yoshida, Kazuhiko Uchiyama, Takeshi Ishikawa, Tomohisa Takagi, Hideyuki Konishi, Yukiko Morinaga, Mitsuo Kishimoto, Yoshito Itoh.

Acknowledgement

We thank all members of Molecular Gastroenterology and Hepatology, Kyoto prefectural University of Medicine Graduate School of Medical Science for helping conduct this study.

Conflict of interest

Osamu Dohi received a collaborating research fund from Fujifilm Co., Ltd. Fujifilm Co., Ltd. had no role in the design, conduct, data collection, data interpretation, or reporting of this study. The other authors declare no conflicts of interest in association with this study.

Japanese Gastric Cancer Association. Japanese gastric cancer treatment guidelines 2018 (5th edition). Gastric Cancer 2021; 24: 1–21. doi:10.1007/s10120-020-01042-y
Kim YJ, Park JC, Kim JH et al. Histologic diagnosis based on forceps biopsy is not adequate for determining endoscopic treatment of gastric adenomatous lesions. Endoscopy 2010; 42: 620–626. doi:10.1055/s-0030-1255524
Takao M, Kakushima N, Takizawa K et al. Discrepancies in histologic diagnoses of early gastric cancer between biopsy and endoscopic mucosal resection specimens. Gastric Cancer 2012; 15: 91–96. doi:10.1007/s10120-011-0075-8
Komatsu S, Ichikawa D, Miyamae M et al. Discrepancies in the histologic type between biopsy and resected specimens: a cautionary note for mixed-type gastric carcinoma. World Journal of Gastroenterology 2015; 21: 4673–4679. doi:10.3748/wjg.v21.i15.4673
Kim Y, Yoon HJ, Kim JH et al. Effect of histologic differences between biopsy and final resection on treatment outcomes in early gastric cancer. Surgical Endoscopy 2020; 34: 5046–5054. doi:10.1007/s00464-019-07301-z
Muehldorfer SM, Stolte M, Martus P et al. Diagnostic accuracy of forceps biopsy versus polypectomy for gastric polyps: a prospective multicentre study. Gut 2002; 50: 465–470. doi:10.1136/gut.50.4.465
Miwa K, Doyama H, Ito R et al. Can magnifying endoscopy with narrow band imaging be useful for low grade adenomas in preoperative biopsy specimens? Gastric Cancer 2012; 15: 170–178. doi:10.1007/s10120-011-0093-6
Nakayoshi T, Tajiri H, Matsuda K et al. Magnifying endoscopy combined with narrow band imaging system for early gastric cancer: correlation of vascular pattern with histopathology (including video). Endoscopy 2004; 36: 1080–1084. doi:10.1055/s-2004-825961
Yagi K, Nozawa Y, Endou S, Nakamura A. Diagnosis of early gastric cancer by magnifying endoscopy with NBI from viewpoint of histological imaging: mucosal patterning in terms of White zone visibility and its relationship to histology. Diagnostic and Therapeutic Endoscopy 2012; 2012: 954809. doi:10.1155/2012/954809
Yao K, Nagahama T, Matsui T, Iwashita A. Detection and characterization of early gastric cancer for curative endoscopic submucosal dissection. Digestive Endoscopy 2013; 25; Suppl 1: 44–54. doi:10.1111/den.12004
Dohi O, Yagi N, Majima A et al. Diagnostic ability of magnifying endoscopy with blue laser imaging for early gastric cancer: a prospective study. Gastric Cancer 2017; 20: 297–303. doi:10.1007/s10120-016-0620-6
Dohi O, Yagi N, Yoshida S et al. Magnifying blue Laser imaging versus magnifying narrow-band imaging for the diagnosis of early gastric cancer: A prospective, multicenter, comparative study. Digestion 2017; 96: 127–134. doi:10.1159/000479553
Japanese Gastric Cancer Association. A. Japanese gastric cancer treatment guidelines 2014 (ver. 4). Gastric Cancer 2017; 20: 1–19. doi:10.1007/s10120-016-0622-4
Yao K, Anagnostopoulos GK, Ragunath K. Magnifying endoscopy for diagnosing and delineating early gastric cancer. Endoscopy 2009; 41: 462–467. doi:10.1055/s-0029-1214594
Kanesaka T, Uedo N, Doyama H et al. Diagnosis of histological type of early gastric cancer by magnifying narrow-band imaging: a multicenter prospective study. DEN Open 2022; 2: e61
Muto M, Yao K, Kaise M et al. Magnifying endoscopy simple diagnostic algorithm for early gastric cancer (MESDA-G). Digestive Endoscopy 2016; 28: 379–393. doi:10.1111/den.12638
Nonaka K, Arai S, Ban S et al. Prospective study of the evaluation of the usefulness of tumor typing by narrow band imaging for the differential diagnosis of gastric adenoma and well-differentiated adenocarcinoma. Digestive Endoscopy 2011; 23: 146–152. doi:10.1111/j.1443-1661.2010.01070.x

No competing interests reported.

Diagnostic ability of magnifying endoscopy compared to biopsy examination for early gastric cancer prior to endoscopic submucosal dissection

Status:

Version 1

Abstract

Figures

Introduction

Materials and Methods

Patients

Study design

Endoscopic systems and endoscopes

Evaluation of endoscopic findings

Participating endoscopists

Pathological diagnosis

Statistical analysis

Results

Discussion

Declarations

References

Additional Declarations

Status:

Version 1