Endoscopic submucosal dissection of gastric superficial neoplastic lesions: a single Western center experience

Abstract

Background

Endoscopic submucosal dissection (ESD) allows “en bloc” resection of superficial gastric lesions.

Objective

The aim of this study is to report “en bloc" resection rate, procedure results, post resection features, analysis of complications, and relapses at mid–long term follow-up on an Italian center case series of patients with superficial gastric neoplastic lesions treated with ESD.

Methods

This is a retrospective evaluation of a prospectively collected series of patients diagnosed with gastric superficial lesions who underwent ESD between November 2007 and May 2015. The analysis included demographic, clinical, endoscopic, and histological data. In detail, the following features were analyzed: (1) “en bloc” resection rate; (2) R0, R1, and RX rate; (3) complication rate; and (4) neoplastic recurrence during follow-up. A comparison was performed between lesions treated according to standard and to extended criteria.

Results

In the analysis, therefore, 70 superficial lesions and 70 endoscopic procedures were considered.

Overall, 53 (76%) procedures were performed according to the extended-ESD criteria. The remaining 17 (24%) cases were performed according to standard criteria. In 68/70 (97%) procedures “en bloc” resection was successful. Forty-four (65.6%) patients had a R0 resection, while 15 (22%) patients had a R1 and the remaining 8 (12.4%) patients were RX. Two patients (2.8%) experienced a major complication. Overall, 7 patients (10.4%) required surgery: in two cases because of local recurrence, in one case for an intraprocedural complication not amenable to endoscopic treatment, in three patients with R1 vertical margins, and in the last patient because of an aggressive pathology (i.e. signet ring cell adenocarcinoma). The mean follow-up of patients with neoplastic lesions was 36.6 ± 13.1 months.

Conclusion

ESD for early gastric cancer is a safe and effective technique when performed by experienced Endoscopists also in Western setting. When comparing the outcomes of the ESD in patients with standard and extended indication there is no significant difference in terms of complete resection, complications and local recurrence rates. Larger Western studies are needed to clearly define the role and the outcomes of ESD in regions at low incidence for gastric neoplasia.

Keywords

Early gastric cancer endoscopic submucosal dissection

Introduction

Gastric cancer is the third highest cancer-related cause of death worldwide.¹ Early detection of pre-neoplastic lesions, early neoplastic lesions and their appropriate removal are of paramount relevance. Both endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD) may allow a curative treatment for early gastric cancers (EGC) with low risk of lymph node metastasis. When lesions are larger than 20 mm in diameter, EMR often requires a “piecemeal” resection which impairs R0 resection rate. ESD allows “en bloc” resection of the lesions, allowing an adequate histopathological evaluation and a radical treatment with lower risk of relapse.^2,3 The criteria for the endoscopic treatment of EGC proposed by the Japanese Gastric Cancer Association (JGCA) were successively extended by Gotoda et al. (Figure 1).^4,5 Extended criteria are widely accepted, so that several EGC patients may be safely managed endoscopically avoiding a surgical resection. A recent meta-analysis reported that the ESD approach was superior to EMR in patients with extended criteria, in term of “en bloc” resection rate, histologically complete resection rate, and local recurrence, although associated with a higher perforation rate.⁶ In Western countries, EGC is still largely treated with a surgical approach, being EMR/ESD procedures still marginal and performed in few, highly specialized endoscopic centers.^7–13 This results in lack of data coming from Western countries.

Figure 1.

Standard criteria for gastric ESD according guideline and extended criteria for gastric ESD (SM: submucosal, UL: ulceration). Modified by Gotoda et al.⁵

The aim of this study is to report on a case series of patients with superficial gastric neoplastic lesions treated with ESD in a single Italian center. “En bloc” resection rate, procedure results, post resection features, analysis of complications, and relapses at mid–long term follow-up were evaluated.

Patients and methods

Patients

All patients were informed about the endoscopic technique, including advantages and disadvantages, and possible complications. All patients signed an informed consent.

This study was conducted in accordance to the Declaration of Helsinki and was approved by the Ethical Committee of our Hospital (Prot n 0040189/16).

Endoscopic procedures

Initial diagnosis was performed by esophagogastroduodenoscopy (GIF H-180 or GIF-H180J, Olympus, Japan, Tokyo) with chromoendoscopy using indigo carmine and narrow band imaging (NBI). Endoscopic ultrasound (EUS) was not routinely performed. Lesions were classified according to the Paris classification.⁷ All procedures were performed either under deep sedation with propofol or general anesthesia. All the ESD procedures were performed by two expert endoscopists (GC and LP) using an Olympus GIF H-180 or GIF-H180J video endoscope system (EVIS EXERA II; Olympus, Tokyo, Japan). In all cases the electrosurgical unit was the VIO 300 (ERBE Elektromedizin GmbH, Tübingen, Germany) set at Endocut I, Dry Cut effect 2, 30W, Swift Coag 30 W.

A soft, straight, and transparent distal attachment (D-201-11304; Olympus, Tokyo, Japan) to facilitate the evaluation of the submucosal (SM) layer was used during ESD. Different cutting tools as Dual-Knife, IT-Knife and IT-Knife2 (KD-611L; Olympus) were used to mark the lesion, to perform the mucosal incision, and then the submucosal dissection. A grasper forceps (Coagrasper; Olympus) was used for bleeding control and hemostasis (ERBE VIO 300 D, Soft Coag 60 W), whereas endoscopic clips (EZ Clip HX-610-135L; Olympus) were used for bleeding control and closure of perforations. Figure 2 shows the procedure steps. In the first phase the lesion was observed in white light (a) and subsequently, to define the side edges, traditional indigo carmine 0.5% chromoendoscopy and high definition virtual chromoendoscopy (NBI) was used (b). Border marking was performed at 2–5 mm from the side margins of the lesions using closed toe cutting device Dual-Knife (c). Submucosal injection was done with a 10% glycerol solution (Chugai Pharmaceutical Co, Tokyo, Japan) with indigo carmine 0.5% and adrenaline at different dilutions or hydroxypropyl methyl cellulose (HMPC) with addition of indigo carmine and adrenaline at different dilutions (d). Circumferential incision to access the submucosa was performed with Dual-Knife externally to the border markings. Finally, ESD was performed using different cutting tools (Dual-Knife, Knife-IT, IT-Knife 2). When necessary, monopolar coagulation was performed by hemostatic forceps (Coagrasper).

Figure 2.

Endoscopic submucosal dissection of an angulus lesion (extended criteria): (a) white light lesion evaluation; (b) NBI lesion analysis; (c) border marking; (d) submucosal injection; (e) circumferential incision; (f) endoscopic submucosal dissection; (g) post “en bloc” resection eschar; (h) gastric lesion removed and staged oriented on cork.

Histopathological assessment

All resected lesions were placed and oriented on cork immediately after the procedure. Complete resection was confirmed by the presence of the markings on the side edges of the lesion. Prior to cutting, specimen bases were marked with India ink; complete resection was confirmed by the absence of neoplastic tissue on the marked edges of the resected specimen. The “piecemeal” removed lesions were reconstructed trying to respect the topography of the lesion and fixed on a cork. Lesions sizes were recorded. Samples were fixed in formalin for histopathological evaluation. Two millimeters cuts were performed and lateral margins, depth of invasion and the presence of tumor budding and lymphovascular embolization were evaluated.⁸ Lauren classification for histological diagnosis was used.⁹

Endpoints

Feasibility was defined as the success in removing the entire endoscopic lesion. The ESD was defined complete if dissection was completed, while incomplete if the lesion removal was completed by using EMR snares (Oval plurifilament type, Olympus, or monofilament type, Riab Medical s.r.l., Cesena, Italy). The JGCA definitions were adopted to classify the completeness of dissection. In detail, “en bloc” resection (ER) was defined as resection in a single piece as opposed to piecemeal resection (in multiple segments). Complete en bloc resection (R0) was defined when en bloc resection was achieved, with lateral and deep tumor-negative margins. A free margin was defined as a 2 mm distance from the lesion to the margin in the resection specimen. Curative resection (CR) was defined when R0 was achieved, with absence of lymphovascular invasion and submucosal infiltration <500 µm. Unclear horizontal and/or vertical margins were analyzed as RX.¹⁰ A resection with histopathological confirmation of residual lesion of deep and/or lateral margins was defined R1. Recurrence was defined as presence of neoplastic tissue (intraepithelial neoplasia or high and/or low-grade dysplasia) at the scar during follow-up endoscopy. Follow-up was performed by upper endoscopy and biopsies every 3 months for the first year, and thereafter once a year.

Adverse events

Major and minor adverse events were routinely recorded. Perforation was considered a major complication. Perforation was defined as the presence of mesenteric fat or intra-abdominal space visualization through the gastric wall during the procedure. Procedural bleeding was considered a major adverse event only when not manageable endoscopically or requiring red blood cells transfusion. A delayed bleeding was defined a major adverse event in case of hematemesis and melena appearing >24 h post-procedure and/or a reduction of more than 2 g/L of hemoglobin. Microperforations or bleeding episodes that occurred during the procedure and that could be successfully managed by endoscopy during the procedure were considered as minor complications.

Statistical analysis

Statistical Package for Social Sciences (SPSS 17.0) software was used for statistical analysis of the data. Comparisons between groups of extended criteria (Ext-G) and the standard criteria (St-G) was performed using the Fisher test. The difference was defined as statistically significant for p < 0.05. The comparison of the size and characteristics of the lesions was performed with the t-test. The difference was defined as statistically significant when p < 0.001.

Results

A total of 70 patients (34, 49% males, 36, 51% females, mean age 67.7 years, range 34–85 years) were eligible for ESD, for a total of 71 superficial gastric lesions. One patient presented with two lesions, while in one case the lifting sign was absent. In this latter patient the endoscopic resection was not attempted and he underwent surgical resection. In the analysis, therefore, 70 superficial lesions and 70 endoscopic procedures were considered.

Overall, 53 (76%) procedures were performed according to the extended-ESD criteria. The remaining 17 (24%) cases were performed according to standard criteria. Fifty-five lesions (79%) were localized in the antrum, 7 (10%) in the gastric body, and 8 (11% ) in the fundus. Thirty-two (46%) lesions were elevated with depressed area (type IIa + IIc of the Paris classification). The Paris classification of the remaining lesions is reported in Table 1. The mean diameter of the lesions was 25.4 ± 10.15 mm. Reason for extended criteria was size (≥20 mm) and ulcerated morphology. In details, 46 lesions were ≥20 mm in size while 7 lesions were <30 mm in size showing an ulcerated morphology.

Table 1.

Demography, classification of lesions, and pathology results.

	Total 70 (%)	Extended (total 53) (%)	Standard (total 17) (%)	p (Ex-St)
“En bloc” resections	68 (97)	52 (98)	16 (94)	n.s.
Age, median (years)	67.69 ± 12.55	70 ± 11.2	62 ± 10.6	n.s.
Male (n)	34 (49)	23 (43)	10 (59)	n.s.
Female (n)	36 (51)	30 (57)	7 (41)	n.s.
Early gastric cancer (ECG)	44 (63)	37 (70)	7 (41)	<0.05
Adenomas HGD/LGD - NET	23 (33)	13 (24)	10 (59)	<0.05
Benign lesions (hyperplasia)	3 (4)	3 (6)	0	n.s.
Diameter lesions, mean (mm)	25.4 ± 10.15	28.29 ± 8,4	17.29 ± 5,6	<0.001
Site of lesions (n)
Fundus	8 (11)	8 (15)	0	n.s.
Body	7 (10)	5 (9)	2 (12)	n.s.
Antrum	55 (79)	40 (76)	15 (88)	n.s.
Type of lesions (Paris class)
0-Is	13 (18)	6 (11)	7 (41)	n.s.
0-IIa	16 (23)	9 (17)	7 (41)	<0.05
0-IIb	3 (4)	1 (2)	2 (12)	n.s.
0-IIc/III	6 (9)	6 (11)	0	n.s.
0-IIa + II c	32 (46)	31 (59)	1 (6)	<0.05
Pathology post resection
Early gastric cancer
Intestinal type Lauren G1	23 (33)	20 (38)	3 (18)	n.s.
Intestinal type Lauren G2	19 (27)	15 (28)	4 (23)	n.s.
Diffuse (Signet Ring Cell)	2 (3)	2 (4)	0	n.s.
Undifferentiated	0	0	0	n.s.
Intraepithelial neoplasia
Adenoma HGD	13 (19)	9 (17)	4 (23)	n.s.
Adenoma LGD	7 (10)	4 (8)	3 (18)	n.s.
NET	3 (4)	0	3 (18)	n.s.
Hyperplastic lesions	3 (4)	3 (5)	0	n.s
				n.s.

In 68/70 (97%) procedures, “en bloc” resection was successful; in one case a circumferential incision followed by lesion removal in two pieces by diathermy snare was performed (Hybrid-ESD), while in the remaining patient a partial ESD was done and completed with “en bloc” EMR. The average procedure time was 70 ± 42 min (Table 1).

The histology of the resected lesions is reported in Table 1. Sixty-seven out of the 70 resected lesions were neoplastic, while the remaining three were hyperplastic lesions.

Of the resected neoplastic lesions (n = 67), 32 (48%) were intramucosal lesions, 28 (42%) were infiltrating the first third of the submucosa (up to 500 μm), while in the remaining 7 cases (10%) the submucosal involvement exceeded the first third. Details of the grade of lesion infiltration are reported in Table 2.

Table 2.

Post resection features and follow-up.

	Total 67 (%)	Extended (total 50) (%)	Standard (total 17) (%)	p (Ex-St)
Vertical margins
R0	59 (89)	45 (90)	15 (88)	n.s.
R1	5 (7)	4 (8)	1 (6)	n.s.
RX	2 (3)	1 (2)	1 (6)	n.s.
Lateral margins
R0	47 (70)	32 (64)	15 (88)	<0.05
R1	13 (19)	12 (24)	1 (6)	n.s.
RX	7 (11)	6 (12)	1 (6)	n.s.
Depth of invasion
m1-m2-m3	32 (48)	22 (44)	10 (59)	n.s.
sm1	28 (42)	23 (46)	5 (30)	n.s.
sm2-sm3	7 (10)	5 (10)	2 (11)	n.s.
Vascular lymphatic invasion
Yes	4 (6)	4 (8)	0	n.s.
No	63 (94)	46 (92)	17 (100)	n.s.
Recurrence (HGD-LGD)
Yes	6(9)	6 (12)	0	n.s.
No	61 (91)	44 (88)	17 (100)	n.s.
Surgery
Total	7 (10)	7 (14)	0	n.s.
Recurrence	2
Diffuse (signet ring cell)	1
R1 vertical margins	3
Major complication	1
Second treatment
ESD	1
APC	2

sm: submucosa; m: mucosa; APC: argon plasma coagulation; HGD: high-grade dysplasia; LGD: low-grade dysplasia.

Forty-four (65.6%) patients had both a complete en bloc (R0) and curative resection (i.e. lateral and deep tumor-negative margins with absence of lymphovascular invasion and submucosal infiltration <500 µm), while 15 (22%) patients had a R1 and the remaining 8 (12.4%) patients were RX. Details of the margin resection are shown in Table 3. All but one patient with R1 and 6 out of 8 patients with RX were in the extended group. Details of patients with R1 and/or RX resections are listed in Table 3.

Table 3.

Details of patients with R1 and/or RX.

St-G/Ex-G S/E	Histology	Vertical margin (R1/RX)	Lateral margin (R1/RX)	Depth of invasion	Vascular lymphatic invasion (Yes/No)	Outcome
S	NET G1	R1	R1	sm3	No	Follow-up
E	Intestinal type Lauren G2	R1	R1	sm1	No	Surgery
E	Intestinal type Lauren G2	R1	R1	sm2	Yes	Surgery
E	Intestinal type Lauren G1	R1	R0	sm1	No	Follow-up
E	Intestinal type Lauren G2	R1	R0	sm3	Yes	Surgery
E	Intestinal type Lauren G1	R0	R1	sm1	No	Follow-up
E	Intestinal type Lauren G1	R0	R1	m3	No	Follow-up
E	Intestinal type Lauren G1	R0	R1	m3	No	Follow-up
E	Intestinal type Lauren G1	R0	R1	m3	No	Follow-up
E	Adenoma HGD	Rx	R1	m3	No	Recurrence-surgery
E	Adenoma HGD	R0	R1	m3	No	Follow-up
E	Adenoma HGD	R0	R1	m3	No	Follow-up
E	Diffuse (signet ring cell)	R0	R1	sm1	No	Surgery
E	Diffuse (signet ring cell)	R0	R1	sm1	No	Unfit for surgery
E	Intestinal type Lauren G2	R0	R1	sm1	No	Follow-up
S	NET G1	Rx	R0	sm3	No	Follow-up
E	Adenoma LGD	R0	RX	m3	No	APC
E	Adenoma LGD	R0	RX	m3	No	Follow-up
E	Adenoma HGD	R0	RX	m3	No	Follow-up
E	Intestinal type Lauren G2	R0	RX	sm1	No	Follow-up
S	Intestinal type Lauren G2	R0	RX	sm1	No	Follow-up
E	Intestinal type Lauren G2	R0	RX	sm1	No	Follow-up
E	Intestinal type Lauren G2	R0	RX	sm1	No	Follow-up

Two patients (2.8%) experienced a major complication. One patient presented a delayed (24 hours post procedure) bleeding with hematemesis and reduction of 4 g/L of hemoglobin. An urgent endoscopy was performed and a combined hemostasis (monopolar coagulation with Coagrasper and clipping (Type EZ LONG, Olympus) was successfully attempted. Another patient had an intraprocedural perforation that was unsuccessfully treated by clips (OTSC System, Ovesco Endoscopy AG, Tübingen, Germany). This patient was, therefore, referred for surgical resection. Two intraprocedural microperforations occurred. These patients were treated successfully by clipping, without sequelae.

Overall, 7 patients (10.4%) required surgery: in 2 cases because of local recurrence, in 1 case for an intraprocedural complication not amenable to endoscopic treatment, in 3 patients with R1 vertical margins and in the last patient because of an aggressive pathology (i.e. signet ring cell adenocarcinoma). The mean follow-up of patients with neoplastic lesions was 36.6 ± 13.1 months. The details of patients with recurrence, the therapeutic approaches, and the outcomes are listed in Table 4. At a telephone contact in December 2016 all patients were alive.

Table 4.

Details of patients who presented recurrence.

St-G/Ex-G S/E	Histology	Vertical margin (R1/Rx)	Lateral margin (R1/Rx)	Depth of invasion	Vascular lymphatic invasion (yes/no)	Timing of recurrence (months)	Therapeutic action	Type of recurrence (LGD/HGD)	Outcome/ months follow up
E	Intestinal type Lauren G1	R0	R0	sm1	no	12	ESD	HGD	Free of disease/ 30 months
E	Intestinal type Lauren G2	R0	R0	sm1	no	3	Surgery	HGD	Free of disease/ 24 months
E	Adenoma HGD	RX	R1	m3	no	1	APC/surgery (24 months)	HGD	Free of disease/ 32 months
E	Intestinal type Lauren G2	R0	RX	sm1	no	1	na	HGD	Lost at follow-up
E	Intestinal type Lauren G1	R0	R0	sm1	no	24	APC	HGD	Free of disease/ 28 months
E	Adenoma LGD	R0	RX	m3	No	3	APC	LGD	Free of disease/ 12 months

Tables 1 and 2 report the details of a comparison between the extended criteria group (Ex-G) and standard criteria group (St-G). To note, a statistical significant difference was observed when considering the histology, diameter, type (Paris classification) and completeness of resection when comparing lesions with extended and standard criteria. In details, lesions with extended criteria were more frequently EGC (70% vs. 41%), were bigger in diameter (28.3 mm vs. 17.3 mm) and they were more frequently complex lesions (i.e. 0-IIa + IIc) (59% vs 6%). Lesions with standard criteria had a statistically higher R0 resection for lateral margins when compared to lesions with extended criteria. No differences were observed between the two groups in terms of en-bloc resection, R0 resection for deep margin, lympho-vascular invasion, recurrence and adverse events.

Discussion

The results of the present study show that there was no difference between standard and extended criteria group in terms of complete resection rate (98% vs. 94%; p = ns). Complication rate (0% vs. 2.8%; p = ns) and local recurrence rate (0% vs. 12%; p = ns) were lower in the standard indication group, nevertheless the difference was not statistically significant. Endoscopic resection has become the standard treatment for pre-neoplastic or early malignant lesions in the stomach. Conventional EMR, although very useful for the removal of small lesions, has the limitation at not allowing a safe and potentially curative “en bloc” resection when the lesions are bigger than 20 mm in diameter. ESD has been developed in Japan to overcome this limitation.¹¹ The advantage of ESD is the ability to perform “en bloc” resections without any dimensional limitation. This advantage leads to an adequate histopathological evaluation of R0 resections and reduces the risk of local recurrence.¹² Large studies have reported “en bloc” resection rates with no cancer margins in more than 90% of EGC and large flat gastric adenomas.¹² Traditionally, only mucosal lesions smaller than 20 mm were considered suitable for endoscopic treatment. The study by Gotoda et al. on a large series of 5265 patients who underwent gastrectomy with lymph node resection for EGC proposed new criteria suggesting a low risk of lymph node metastasis even in large and complex lesions.⁵ Based on the results of this trial an extended indication for endoscopic resections was proposed for EGC. In adjunction, the advance of instruments as well the development of new techniques allowed to perform safe endoscopic resections also of complex gastric lesions, avoiding unnecessary surgery. Therefore, the number of patients receiving ESD for EGC has continuously increased during the last years.

Several studies showed that ESD in extended indication has similar recurrence rate and disease-free survival rate when compared to standard indication. However, the vast majority of the published data comes from Japan or Eastern countries. This is due to a lower incidence of gastric cancer and especially of EGC in the Western world in comparison to Asian regions,¹³ where there are widely diffused mass screening programs.

The results of the present study bring out some considerations that deserve to be commented. First, about 74% of the lesions included in this trial were selected according to extended criteria, therefore in a more advanced disease stage. This might be the result of the lower incidence in the Western Countries of gastric cancer and of the lack of screening programs aimed at intercepting lesions at early stages. Second, the 97% of “en bloc” resection is comparable to the rate of complete resection described in other bigger series (Table 5). The remaining patients who could not benefit from an “en bloc” resection were treated by a hybrid endoscopic treatment (i.e. circumferential incision and resection with a snare or partial ESD completed by piece-meal mucosal resection). To note, there is no difference in terms of “en bloc” resection between the standard and extended groups in the present series. Third, there is no difference in terms of R0 resection for vertical margins while, when considering the lateral margins, R0 is statistically significant lower in lesions with extended indication. Although a clear explanation of such underperformance is not completely clear, a possible reason could be an improper definition of the lateral margin of the lesions and/or lateral border marking performed at a distance lower than 5 mm. When compared to the Eastern series (Table 5), the low R0 resection for lateral margins observed in this study is certainly relevant and needs to be better assessed and understood in order to identify corrective actions to improve the final result of endoscopic treatment. Fourth, when compared to previous studies, the higher overall recurrence (9%) of the present study is due to the high recurrence reported in the extended group (12%), while no recurrence was observed in the standard group over a period of a mean of 36 months of follow up. Such high recurrence rate is probably directly connected to the low R0 resection for lateral margin described above. Fifth, the overall major adverse event occurrence was 2.8% (i.e. 1 perforation and 1 delayed bleeding) and it is comparable to that of previous series (Table 5).^14–17

Table 5.

Comparison with previous major series.

Author	Year	Country	Patients n.	Lesions n.	StG/ExtG (n. lesions)	En bloc resection (StG/ExG) %	Tumor size >20 mm (StG/ExG) %	RO latera margin (StG/ExG) %	RO vertical margin (StG/ExG) %	Complication Perforation (StG/ExG) %	Complication Bleeding (StG/ExG) %	Follow up Months	Recurrence %	Time min
Jeon et al.²¹	2012	Korea	321	348	100/248	100/94.4	0/47.2	100/90.3	100/99.2	na	na	26	0	na
Ohnita et al.²²	2014	Japan	1209	1332	617/507	96.6/91.5	46.3/38.1	na	na	1.8/4.3 2.9 overall	1.1 /1.8	66	na	na
Park et al.²³	2014	Korea	1152	1175	366/565	92.9/89.9	0/100	89.1/83.2	89.1/83.2	2.7/3.7	3.8/6.4	StG24.1/ExG30	2.2/1.3	na
Kosaka et al.²⁴	2014	Japan	472	438	297/107	98/88 UL-95.8UL+	na	99.3/89.2 UL- 100UL+	100/96.4 UL- 100UL+	2.4/6 UL- 4.2UL+	3.0/3.6 UL- 8.3 UL+	32.4	0.3/4.8 UL- 0 UL+	47
Probst et al.¹⁰	2010	Germany	83	85	10/51	100/88.2	0/100	90/68.3	na	0	8.3 overall	27	1.5 % (ESD) 50%(PM-EMR)	142
Present series	2017	Italy	70	70	17/53	94/98	7/39	88/64	88/90	0/1.4	0/1.4	36	8.9	70

Both complications in the present series occurred in patients with extended indications. Although there is no statistically difference in terms of recurrence rate and adverse event occurrence between the two groups, the higher recurrence and adverse event rates observed in the extended group should be considered with caution suggesting the higher complexity of ESD in patients with bigger and/or more advanced lesions. Sixth, EUS was not routinely performed in the present series. Several studies have compared the accuracy of invasion depth measurement between conventional endoscopy and EUS. Two recent meta-analyses concluded that EUS shows low accuracy for staging the depth of EGC invasion limiting the role of EUS in the staging algorithm and for the selection of patients with early gastric neoplasia suitable to ESD treatment.^19,20

Finally, to note, the 70 patients included in the present study were collected during a 8 years period of enrollment while, for example, Kosaka et al. collected 472 cases of EGC during a 5 years period of enrollment.¹⁷ The low incidence of gastric lesions with an indication for ESD in Western Countries makes the learning curve long and difficult to be completed in a reasonable period of time, even in referral centers. This reflects the different skills and it might be in part responsible of the differences of the performances when comparing the present series with larger, Eastern series.

This study presents some limitations. The main limitation of the present study is the small sample size. Although the sample size is similar to that of previous European studies, this study seems to be hardly comparable when looking at East. On the other hand, only few studies report Western experiences on the endoscopic treatment of gastric lesions. Therefore, the present study gives a new, additional picture of the gastric ESD in regions where the incidence of gastric neoplasia is low and where, probably, the endoscopic skill for ESD needs to be improved. The small number of the two groups (i.e. standard and extended indications) makes the comparison difficult. As a consequence of the small sample size the difference of most of the evaluated parameters was not significant. Finally, although patients were prospectively collected, the retrospective design and the lack of a formal power calculation are a clear limitation. About this, it needs to be underlined that although the differences between standard and extended criteria group failed to reach statistical significance a formal power calculation was not performed and with just 24% of cases meeting standard criteria it is likely that the study is underpowered to show any difference.

In conclusion, the results of the present study suggest that ESD for EGC is a safe and effective technique when performed by experienced Endoscopists also in Western setting. When comparing the outcomes of the ESD in patients with standard and extended indication there is no significant difference in terms of complete resection, complications and local recurrence rates. The small sample size of the present trial limits the generalizability of the results. Larger Western studies are needed to clearly define the role and the outcomes of ESD in regions at low incidence for gastric neoplasia.

Footnotes

Declaration of conflicting interests

None declared.

Ethics approval

This study was approved by the Ethical Committee of our Hospital (Prot n 0040189/16).

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Informed consent

All patients were informed about the endoscopic technique, including advantages and disadvantages, and possible complications. All patients signed an informed consent.

References

Jemal

Siegel

Ward

et al.

Cancer statistics, 2009. CA Cancer J Clin 2009; 59: 225–249.

Miyamoto

Muto

Hamamoto

et al.

A new technique for endoscopic mucosal resection with an insulated-tip electrosurgical knife improves the completeness of resection of intramucosal gastric neoplasms. Gastrointest Endosc 2002; 55: 576–581.

Yamamoto

Kawata

Sunada

et al.

Successful en-bloc-resection of large superficial tumors in the stomach and colon using sodium hyaluronate and small caliber-tip-transparent hood. Endoscopy 2003; 35: 690–694.

Japanese Gastric Cancer Association. Japanese classification of gastric carcinoma: 2nd English edition. Gastric Cancer 1998; 1: 10–24.

Gotoda

Yanagisawa

Sasako

et al.

Incidence of lymph node metastasis from early gastric cancer: estimation with a large number of cases at two large centers. Gastric Cancer 2000; 3: 219–225.

Lian

Chen

Zhang

et al.

A meta-analysis of endoscopic submucosal dissection and EMR for early gastric cancer. Gastrointest Endosc 2012; 76: 763–770.

Probst A, Golger D, Arnholdt H, et al. Endoscopic submucosal dissection of early cancers, flat adenomas and submucosal tumors in the gastrointestinal tract. Clin Gastroenterol Hepatol 2009; 7: 149–155.

Dinis-Ribeiro M, Pimentel-Nunes P, Afonso M, et al. A European case series of endoscopic submucosal dissection for gastric superficial lesions. Gastrointest Endosc 2009; 69: 350–355.

Repici A, Zullo A, Hassan C, et al. Endoscopic submucosal dissection of early gastric neoplastic lesions: a western series. Eur J Gastroenterol Hepatol 2013; 25 :1261–1264.

10.

Probst A, Pommer B, Golger D, et al. Endoscopic submucosal dissection in gastric neoplasia – experience from a European center. Endoscopy 2010; 42: 1037–1044.

11.

Schumacher B, Charton JP, Nordmann T, et al. Endoscopic submucosal dissection of early gastric neoplasia with a water jet-assisted knife: a Western, single-center experience. Gastrointest Endosc 2012; 75: 1166–1174.

12.

Pimentel-Nunes P, Mourão F, Veloso N et al. Long-term follow-up after endoscopic resection of gastric superficial neoplastic lesions in Portugal. Endoscopy 2014; 46: 933–940.

13.

Emura F, MejÚJ, Donneys A, et al. Therapeutic outcomes of endoscopic submucosal dissection of differentiated early gastric cancer in a Western endoscopy setting (with video). Gastrointest Endosc 2015; 82: 804–811.

14.

Participants in the Paris Workshop. The Paris endoscopic classification of superficial neoplastic lesions: esophagus, stomach, and colon: November 30 to December 1, 2002. Gastrointest Endosc 2003; 58: S3–S43.

15.

Mino-Kenudson M, Brugge WR, Puricelli WP, et al. Management of superfcial Barrett’s epithelium-related neoplasms by endoscopic mucosal resection: clinicopathologic analysis of 27 cases. Am J Surg Pathol 2005; 29: 680–686.

16.

Lauren P. The two histological main types of gastric carcinoma: diffuse and so-called intestinal-type carcinoma. Acta Pathol Microbiol Scand 1965; 64: 31–49.

17.

Japanese Gastric Cancer Association. Gastric cancer treatment guideline. 4th ed. Tokyo: Kanahara; 2014: 20–23 (in Japanese).

18.

Hosokawa K and Yoshida S. Recent advances in endoscopic mucosal resection for early gastric cancer (in Japanese with English abstract). J Gan To Kagaku Ryoho 1998; 25: 476–483.

19.

Onozato Y, Ishihara H, Iizuka H, et al. Endoscopic submucosal dissection for early gastric cancers and large flat adenomas. Endoscopy 2006; 38: 980–986.

20.

Forman D and Burley VJ. Gastric cancer: global pattern of the disease and an overview of environmental risk factors. Best Pract Res Clin Gastroenterol 2006; 20: 633–649.

21.

Jeon SR, Cho JY, Bok GH, et al. Does immunohistochemical staining have a clinical impact in early gastric cancer conducted endoscopic submucosal dissection? World J Gastroenterol 2012; 18: 4578–4584.

22.

Ohnita K, Isomoto H, Shikuwa S, et al. Early and long-term outcomes of endoscopic submucosal dissection for early gastric cancer in a large patient series. Exp Ther Med 2014; 7: 594–598.

23.

Park CH, Shin S, Park JC, et al. Long-term outcome of early gastric cancer after endoscopic submucosal dissection: expanded indication is comparable to absolute indication. Dig Liver Dis 2013; 45: 651–656.

24.

Kosaka T, Endo M, Toya Y, et al. Long-term outcomes of endoscopic submucosal dissection for early gastric cancer: a single-center retrospective study. Dig Endosc 2014; 26: 183–191.

25.

Pei Q, Wang L, Pan J, et al. Endoscopic ultrasonography for staging depth of invasion in early gastric cancer: A meta-analysis. J Gastroenterol Hepatol 2015; 30: 1566–1573.

26.

Mocellin S and Pasquali S. Diagnostic accuracy of endoscopic ultrasonography (EUS) for the preoperative locoregional staging of primary gastric cancer. Cochrane Database Syst Rev 2015. DOI: 10.1002/14651858.