Endoscopic Full-Thickness Resection vs Submucosal Tunneling Endoscopic Resection for Gastric Submucosal Tumors

Abstract

INTRODUCTION:

The new working submucosal tunnel space allows entry to deeper layers of the luminal wall or even entirely outside the gastrointestinal tract for the treatment of submucosal tumors. Based on this concept, we developed submucosal tunneling endoscopic resection (STER). Here, we compared the clinical outcomes between exposed endoscopic full-thickness resection (EFTR) and STER (nonexposed EFTR) and analyzed the efficacy and safety of Natural Orifice Transluminal Endoscopic Surgery (NOTES) based on STER for extra-gastrointestinal stromal tumors (EGISTs).

METHODS:

Sixty consecutive patients with tumors in the lesser curvature of the stomach corpus were enrolled from July 2019 to December 2023. Data on clinicopathologic features, treatment results, and follow-up outcomes were collected and analyzed retrospectively.

RESULTS:

Among the 60 patients, 31 patients underwent EFTR and 29 patients underwent STER. The EFTR group had a shorter procedure time (P = 0.016) but a longer postoperative hospital stay (P = 0.004) than the STER group. Tumor size > 2 cm and endoloop-clips suture were significantly associated with long-time procedure. NOTES based on STER was successful for EGISTs. Follow-up data from 6 to 60 months was collected with no loss. All patients were free from local recurrence and distant metastasis during the study period.

CONCLUSIONS:

Although the procedure time of STER is longer than that of EFTR, the postoperative hospital stay is shorter. Tumor size > 2 cm and use of endoloop-clips suture are significantly associated with long-time procedure. In addition, STER-based NOTES is a promising and safe methodology for the resection of EGISTs.

Download video file ^{(15.3MB, mp4)}

Download video file ^{(23.1MB, mp4)}

INTRODUCTION

In recent years, the incidence of submucosal tumors (SMTs), including gastrointestinal stromal tumors (GISTs), has increased dramatically worldwide (1). Of note, gastric GISTs have a predominant extraluminal growth type, and some may occur outside of the gastrointestinal tract (2). Known as extra-gastrointestinal stromal tumors (EGISTs), these tumors have several important differences from GISTs, including relative aggressiveness, large size, poor prognosis, and low mitotic index (3).

Compared with the traditional surgical approach, endoscopic treatment has the advantages of minimal invasiveness, less pain, and faster recovery of digestive tract function. For lesions that originated from the deeper muscularis propria layer with a predominant extraluminal growth, exposed endoscopic full-thickness resection (EFTR) is an effective and feasible endoscopic treatment option. More recently, access to the submucosal layer and the subserosal layer, also known as the third space and the fourth space, respectively, has highlighted the potential of interventional endoscopy in mitigating the need for invasive surgical interventions for the management of luminal lesions (4,5). Accessing the submucosa endoscopically provides additional working tunnel space, allowing entry to deeper layers of the luminal wall or even entirely outside the gastrointestinal tract (5). Notably, the mucosal entry point above this submucosal working tunnel could be effectively clipped with standard endoscopic devices, thereby maintaining luminal integrity (6,7). Accordingly, we developed a new technique, submucosal tunneling endoscopic resection (STER), also known as nonexposed EFTR, to use the submucosal space for resection of the upper gastrointestinal SMTs (8). Since its clinical introduction in 2011, STER has become widely established for over a decade. In the process, we have continuously performed a series of research on therapeutic indications, treatment effect, and safety of STER, as well as the prevention of adverse events, and compared our findings with those of traditional surgeries (6,7,9,10). Theoretically, compared with exposed EFTR, STER has advantages in preserving the integrity of the gastrointestinal tract that could promote wound healing and decrease the risk of gastrointestinal tract leakage and secondary infection. In this study, we compared the clinical outcomes of EFTR and STER in the treatment of SMTs in the lesser curvature of the stomach corpus and evaluated the efficacy and safety of Natural Orifice Transluminal Endoscopic Surgery (NOTES) based on STER technique for EGISTs. The data and conclusions provided in this study may become a useful reference for doctors and researchers in this gastroenterology.

PATIENTS AND METHODS

Patients

A total of 60 consecutive patients with tumors in the lesser curvature of the stomach corpus were enrolled in this study from July 2019 to December 2023 (Figure 1). Lesions with an indication for EFTR and STER were determined by endoscopic features using endoscopic ultrasonography (EUS) or computed tomography (CT). According to our previous experience, effective tunneling can be successfully performed on the lesser curvature of gastric corpus, so tumors at this site are very suitable for treatment with STER (6). The inclusion criteria for enrollment were as follows: single tumor in the lesser curvature of the stomach corpus; patients-reported typical or atypical upper gastrointestinal symptoms, suffered from severe anxiety about a potentially malignant tumor, and specifically requested aggressive treatment; patients consent to undergo a STER or EFTR procedure. The exclusion criteria were as follows: confirmed lymphatic or distant metastasis; failure to complete a planned procedure or receive concomitant tissue resection; and for patients with severe cardiopulmonary dysfunction, inability to tolerate anesthesia. All procedures were performed by multiple advanced endoscopists who had more than 10 years of experience in performing endoscopic resection. Written informed consent was obtained from each patient, and the study was approved by the Institutional Review Board of Shanghai East Hospital, Tongji University School of Medicine, Shanghai, China.

Figure 1.

Flow chart of the study profile. EFTR, endoscopic full-thickness resection; STER, submucosal tunneling endoscopic resection.

Submucosal tunneling endoscopic resection

STER was performed with the patients under general anesthesia with endotracheal intubation. Endoscopic equipment and accessories were described in our previous study (6,9,10). The application device for cutting or coagulation was the high-frequency electrosurgical unit (VIO200D, EndoCutQ effect 3- duration 2-interval 4; forced coagulation mode effect 2, 50W; soft coagulation mode effect 4, 80W; ERBE, Germany). The procedure began by making a fluid cushion by injecting mixed solution at 3–5 cm orally to the proximal margin of the tumor and a 2-cm longitudinal mucosal incision was made by using a hook knife or hybrid knife at the top of the fluid cushion to provide mucosal entry to the submucosal space (Figures 2b and 3a). The submucosal tunnel was created between the mucosal and muscular layers and ended at 1–2 cm distal to the tumor or the tunnel exit on the serous side of the stomach (Figures 2c and 3c), which ensured a satisfactory endoscopic view of the tumor and enough working space for the resection. STER of the tumor was performed by using a Hook knife, golden knife (Nanwei Medical Technology Co., Ltd.), or hybrid knife (Nanwei Medical Technology Co., Ltd.). Extraction of the mobilized tumor was achieved by using a string bag or an snare to grasp the tumor, and then taking it out through the mucosal entry (Figures 2e and 3e). The resection was terminated when the lesions were endoscopically removed completely without residual tumor tissue. After tumor resection, argon plasma coagulation or coagrasper was used for hemostasis in the tunnel (Figure 2g) and metallic clips (ROCC-D-26–195-C, Micro-Tech, Nanjing, China) or endoloop-clips was used to close the mucosal entry (Figures 2h and 3f, Video 1 and Supplementary Video 1).

Figure 2.

STER for an EGIST. (a) Endoscopic view of the tumor. (b) The mucosal entry established. (c) The submucosal tunnel established and the large blood vessel seen (yellow arrow). (d) The submucosal tumor exposed using the submucosal tunnel technique (yellow arrow). (e) An intraprocedural view showing the traction of the tumor. (f) The submucosal tumor dissected endoscopically and the tumor bed seen (yellow arrow). (g) Endoscopic view of the submucosal tunnel and tumor bed after the tumor was removed. (h) The mucosal entry closed using endoscopic clips. (i) The tumor. EGIST, extra-gastrointestinal stromal tumors; STER, submucosal tunneling endoscopic resection.

Figure 3.

Schematic diagram of STER. (a) The STER started at less curvature of gastric body. (b) Submucosal tunnel established. (c) STER of the EGIST by submucosal tunnel. (d) The tumor resected completely. (e) Extraction of the tumor by using string bag. (f) The closure of the mucosal entry. EGIST, extra-gastrointestinal stromal tumors; STER, submucosal tunneling endoscopic resection.

Endoscopic full-thickness resection

The EFTR procedures were performed with the patients under general anesthesia with endotracheal intubation (Supplementary Video 2). Endoscopic equipment and accessories have been described previously (6,9,11). EFTR consists of several major procedures. (i) Based on factors such as tumor size and growth pattern, the resection line was 0.5 cm distance from the SMT protrusion (Figure 4a). (ii) The submucosal positions were injected with enough solution of normal saline containing 0.4% indigo carmine and 0.025 mg/mL epinephrine (usually 3–5 mL per injection point) to distinguish the submucosal layer and improve operative safety. (iii) Layer by layer, a semicircumferential incision was performed around the marker. (iv) Sequential incision of the seromuscular layer around the lesion was performed to form a deliberate perforation into the abdomen (Figure 4b). (v) A half-completed full-thickness resection (from the muscularis propria layer to the serosal layer) with a minimal margin was performed. (vi) The procedure continued with dental floss traction and full-thickness resection from the serosal side to the mucosal side (Figure 4d). (vii) Closure of the gastric wall defect using the clips or endoloop–clips method was performed after retrieval of the specimen perorally (Figure 4f,b). When the tumor had been retrieved (Figure 4h), it was reassembled and assessed (6).

Figure 4.

EFTR for a tumor in the stomach corpus. (a) Endoscopic view of the tumor. (b) Sequential incision of the seromuscular layer around the lesion. (c) A half-completed full-thickness resection. (d) The procedure continued with dental floss traction and full-thickness resection from the serosal side to the mucosal side. (e) The defect of gastric wall. (f) Closure of the defect using the endoloop–clips method. (g) The gastric wall after closure. (h) The tumor after retrieval. EFTR, endoscopic full-thickness resection.

Histopathological evaluation

The resected specimens were fixed in 10.0% buffered formalin, embedded with paraffin, and sectioned for pathological examination. A regular shaped tumor was defined as a tumor had the capsule or globular shape. A complete en bloc resection was defined as a tumor removed in a single piece, with the capsule intact. A piecemeal resection was defined as a tumor removed as pieces. When the organization of the tumor was difficult to determine, immunohistochemical staining with antibodies S-100, smooth muscle antigen, vimentin, desmin, CD117, and CD34 was performed. Histological diagnoses were based on the 2010 World Health Organization Classification of Tumors.

Management and adverse events

All patients underwent preoperative fasting initiated at 20:00 on the preceding day, with prophylactic antibiotics administered intravenously 30 minutes before the operation. CO₂ insufflation was used in all of the patients to prevent the gas-related complications. During the procedure, intraoperative pneumoperitoneum was inevitable. Thus, a 20 mL syringe with 10 mL of normal saline was inserted into the right lower quadrant to release the gas. This procedure provided clear operation vision by maintaining a stable intra-abdominal pressure, similar to establishing a stable pneumoperitoneum during a laparoscopic operation. The 20 mL syringe was kept in place until the defect closure was completed. At the end of the procedure, gas injection was performed, and the bubbles in the syringe was observed to confirm the closure effect. Postoperatively, all patients fasted and sustained a semirecumbent position. The nasogastric tube was kept, and intravenous antibiotics and proton pump inhibitors were administered for 3 days as a part of the standard postoperative regimen. Tube removal and advancement to liquid intake were permitted only on resolution of febrile symptoms, vomit, or abdominal discomfort. After discharge, all patients continued to take pantoprazole for an additional 2 months.

Intraoperative minor bleeding was successfully treated by immediate endoscopic coagulation. A small amount of gas in the abdomen may be spontaneously absorbed soon after EFTR or STER that cannot be detected by radiological examination. As reported in other studies regarding this tunneling technique (6,7), these minor adverse technical events have minimal clinical impact and symptoms; therefore, they were not considered adverse events in this study. Major adverse events include intraoperative and postoperative major bleeding (>200 mL), postoperative perforation, perioperative infection, and obvious retroperitoneal emphysema. CT scan was used to detect adverse events. Besides radiography, important symptoms and signs (e.g. abdominal pain, distension, and fever) should be observed carefully.

Follow-up

All of the patients were followed up by standard endoscopy at 3, 6, and 12 months during the first year after the initial procedure to view the healing of the wound and to check for residual tumor or recurrence. EUS was also used for surveillance to detect small residual tumors. Thereafter, patients were followed-up yearly. For patients with GISTs, abdominal ultrasound every 6 months and contrast-enhanced CT scan every 12 months were recommended to evaluate distant metastasis.

Data collection and statistical analysis

Clinicopathological, demographic, and endoscopic data were collected and analyzed. Statistical analysis between groups was performed using the Student t test, Pearson χ² test, Continuity Correction χ² test, or Fisher exact test, as appropriate. All of the reported P values were two-sided, and P < 0.05 was considered statistically significant. IBM SPSS Statistics 20 (SPSS, Chicago, IL) was used for statistical analysis.

RESULTS

Clinicopathological characteristics

Among the 60 patients with tumors in the lesser curvature of the stomach corpus (Figure 1), 49 had gastric SMTs, and 11 had EGISTs. Overall, 29 patients received STER, whereas 31 patients received EFTR. The median age of all patients was 57 years (range 26–82 years) (Table 1), and 20 patients were men (33.3%) and 40 patients were women (66.7%). Among all the tumors, 58 had regular shapes (96.7%), 2 had irregular shapes (3.3%), and 22 tumors (36.7%) were > 2.0 cm. No significant difference was observed in age, sex, tumor size, tumor shape, or tumor histopathology between the STER and EFTR groups.

Table 1.

Characteristics of 60 patients and submucosal tumors

Characteristics	STER	EFTR	P
Age, y, median (range)	57 (26–82)	57 (34–76)	0.645
<60	17	17	0.768
≥60	12	14
Sex, male/female	10/19	10/21	0.855
Tumor size
1.0–2.0 cm	18	20	0.844
2.1–4.0 cm	11	11
Shape
Regular	27	31	0.229
Irregular	2	0
Histopathology
GIST	26	25	0.539
Leiomyoma	3	6

EFTR, endoscopic full-thickness resection; GIST, gastrointestinal stromal tumor; STER, submucosal tunneling endoscopic resection.

Technique and outcomes

In this study, 31 EFTR and 29 STER with longitudinal incision were applied. The tunnel length proximate to the tumor was 3.0–5.0 cm. Overall, en bloc resection was achieved in all 60 treated lesions (100.0%) (Table 2), and the median time of procedure was 50 minutes (range 15–130 minutes). Fifty-one (85%) gastric defects were sutured by metallic clips and 9 (15%) by endoloop-clips. According to the χ² analysis (Table 3), tumor size > 2 cm and endooop-clips suture were significantly associated with procedure time > 60 minutes (P < 0.001 and P = 0.004, respectively). In addition, the analysis of Mann-Whitney U test indicated that the procedure time of GIST was longer than that of leiomyoma overall (P = 0.027, Supplementary Table S1, http://links.lww.com/CTG/B322). Further χ² analysis indicated that the suture method significantly influenced the procedure time, regardless of whether it was STER (P = 0.03) or EFTR (P = 0.042) (Supplementary Table S2 and S3, http://links.lww.com/CTG/B322).

Table 2.

Outcomes after resection in patients with endoscopic full-thickness resection and submucosal tunneling endoscopic resection

Outcome	STER	EFTR	P
En bloc, n/N	29/29	31/31
Procedure time, median (range), min	55 (21–130)	40 (15–90)	0.016
Suture method
Metallic clips	25	26	>0.999
Endoloop-clips	4	5
Adverse events
yes	2	1	0.953
no	27	30
Median postoperative hospital stay, median (range), days	4 (2–6)	5 (3–8)	0.004
Recurrence	0	0
Distant metastasis	0	0

EFTR, endoscopic full-thickness resection; STER, submucosal tunneling endoscopic resection.

Table 3.

The characteristic endoscopic features associated with long operative time (≥60 min) and adverse events

	Operative time			Adverse events
	<60 min	≥60 min	P	Without	With	P
Age
<60 (34)	19	15	0.457	33	1	0.811
≥60 (26)	17	9		24	2
Sex
Male (20)	14	6	0.264	19	1	>0.999
Female (40)	22	18		38	2
Size
1.0–2.0 cm (38)	30	8	<0.001	38	0	0.085
2.1–4.0 cm (22)	6	16		19	3
Shape
Regular (58)	36	22	0.156	55	3	>0.999
Irregular (2)	0	2		2	0
Histopathology
GIST (51)	28	23	0.072	48	3	>0.999
Leiomyoma (9)	8	1		9	0
Suture method
Metallic clips(51)	35	16	0.004	49	2	0.391
Endoloop-clips(9)	1	8		8	1

GIST, gastrointestinal stromal tumor.

The incidence of adverse events, including intraoperative bleeding, postoperative bleeding, and obvious retroperitoneal emphysema, was 5% (Table 2). Intraoperative bleeding and postoperative bleeding were treated by immediate endoscopic coagulation. Retroperitoneal emphysema in a patient who had received CO2 insufflation resolved spontaneously within 1–2 hours. Adverse events occurred in patients with 3 tumor sizes (5%), i.e., 3, 3.5, and 4 cm. Patients were discharged after a median hospital stay of 4 days (range 2–8 days). Patients in the EFTR group had a shorter procedure time (P = 0.016, Table 2) but a longer median postoperative hospital stay than the STER group (P = 0.004, Table 2). In addition, further Student t test indicated that the pure operation time (excluding the suturing process) of STER was significantly longer than that of EFTR (P = 0.022).

STER-based NOTES for EGISTs

Among the 60 patients, 11 patients with EGISTs who received STER-based NOTES were included. Their median tumor size was 2.5 cm (range 1.5–3.5 cm) (Table 4). Although 1 patient had an irregular tumor shape, all patients with EGISTs achieved en bloc resection by STER (100%). The median procedure time was 85 minutes (range 45–130 minutes). Among the 11 patients, 2 patients (18.2%) experienced adverse events. Both patients had tumor sizes ≥ 3 cm and relatively long procedure time. The median hospital stay of patients was 4 days (range 2–6 days). No recurrence or distant metastasis was noted during the follow-up period.

Table 4.

Clinicopathological characteristics, procedure details, outcomes and follow-up of 11 patients with extra-gastrointestinal stromal tumors

NO	Sex	Age	Tumor size	Shape	Location	Procedure time	En bloc	Adverse events	Hospital staya	Risk stratification	Follow-up
1	M	56 yrs	3.5 cm	R	Upper body	120 mins	Yes	Without	4 d	Low	33mons
2	F	67 yrs	3.5 cm	R	Middle body	129 mins	Yes	With	6 d	Low	31mons
3	M	61 yrs	2.0 cm	I	Upper body	85 mins	Yes	Without	6 d	Very low	55mons
4	F	32 yrs	3.5 cm	R	Upper body	95 mins	Yes	Without	2 d	Low	45mons
5	F	69 yrs	3.0 cm	R	Upper body	130 mins	Yes	With	4 d	Low	43mons
6	F	53 yrs	3.5 cm	R	Upper body	90 mins	Yes	Without	3 d	Low	45mons
7	M	66 yrs	2.5 cm	R	Upper body	50 mins	Yes	Without	4 d	Intermediate	55mons
8	F	82 yrs	1.8 cm	R	Upper body	45 mins	Yes	Without	4 d	Very low	34mons
9	F	56 yrs	1.5 cm	R	Upper body	75 mins	Yes	Without	4 d	Very low	60mons
10	M	56 yrs	1.5 cm	R	Upper body	55 mins	Yes	Without	3 d	Very low	47mons
11	F	26 yrs	2.0 cm	R	Upper body	50 mins	Yes	Without	4 d	Very low	34mons

EGIST, extra-gastrointestinal stromal tumor; F, female; I, irregular; M, male; R, regular.

^aHospital stay indicate days after procedure.

Follow-up results

The follow-up period of the 60 patients treated by EFTR or STER ranged from 6 to 60 months, and there was no loss to follow-up. During the study period, all tumor-related symptoms were relieved, and all patients were free from local recurrence and distant metastasis.

DISCUSSION

As the most common type of sarcoma in the gastrointestinal tract, GISTs are potentially malignant neoplasms primarily occurring in the stomach (60%) (11,12). Notably, gastric GISTs have a primary extraluminal growth type and can arise outside the gastrointestinal tract as EGISTs (2). EFTR is now a well-established methodology for the resection of tumors that originate from the deeper muscularis propria layer or with an extraluminal growth pattern. Recently, an increasing amount of clinical research has reported the efficacy and safety of EFTR for treating SMTs (13,14). Meanwhile, STER is newly developed by us to prevent full-thickness perforation when gastric GISTs originating from the muscularis propria layer are removed by a submucosal tunnel, which theoretically reduces the potential risk of leakage and secondary infection. However, few studies have compared the outcomes of EFTR and STER for treating gastric SMTs, which we believe is essential for EFTR and STER to achieve maximum clinical efficacy.

In our study, patients treated with STER had a shorter median postoperative hospital stay than those treated with EFTR (P = 0.004). However, the STER procedure time was significantly longer, possibly because submucosal tunnel creation requires considerable time in the presence of complex anatomical configurations, restricted operative spaces, and friable mucosal layers. Another technical issue for STER is that the tunnel space used for excision operation is limited and narrow, which may cause breaching of the tumor capsule when resecting large tumors and increasing the risk of local recurrence. Further Student t test also indicated that the pure operation time (excluding the suturing process) of STER was also significantly longer than that of EFTR (P = 0.022). In a previous report (6), a large size was a contributor to STER-related adverse events (P = 0.029). In our study, STER and EFTR-related adverse events in this present study occurred in patients with 3 tumor sizes, i.e., 3, 3.5, and 4 cm. Although the results of data analysis showed that tumor size > 2 cm was not a risk factor of adverse events (P = 0.085), this may be because our sample size was relatively small, and a larger sample size may be needed to validate the result of the previous study. Moreover, we found it was difficult and time-consuming to resect tumors > 2 cm by χ² analysis and Mann-Whitney U Test (Table 3 and Supplementary Table S1, http://links.lww.com/CTG/B322). The largest tumor size was 4 cm in the EFTR group, whereas it was only 3.5 cm in the STER group. Our experience informs us that endoscopists may face challenges in successfully resecting large tumors using the STER technique. For tumors > 3.5 cm, EFTR is a safer and more suitable option because it avoids dissection near the capsule and is not limited by tunnel space. In addition, the Mann-Whitney U test revealed that GIST had a longer median procedure time compared with leiomyomas, which may be attributed to the fact that GISTs typically exhibit an infiltrative or extraluminal growth pattern. For small tumors resected with EFTR, the postoperative perforation is relatively small; so, the use of several metallic clips is sufficient for suturing. However, for larger gastric perforations, the most practical way to close the defect is to use the endoloop-clips method: metallic clips with nylon loop suturing. Currently, endoscopic suturing have been developed to effectively close defects, however, the application of endoscopic suturing in our country was limited by market and equipment availability, and it has not yet reached full maturity during the study period. Therefore, in our clinical practice, the most commonly used methods in STER and EFTR remain metallic clips and endoloop-clips. The endoloop-clips method is performed through a dual-channel endoscope to ensure total closure of the gastric wall defect. According to our statistical analysis, use of the endoloop-clips suture method is also a contributor to long procedure time overall (P = 0.004, Table 3). Further χ² analysis indicated that the suture method significantly influenced the procedure time, regardless of whether it was STER or EFTR (Supplementary Table S2 and S3, http://links.lww.com/CTG/B322). It is important to note that only skilled endoscopic endoscopists who can execute safe oncologic procedures and close the defect smoothly should perform endoloop-clips method to prevent negative outcomes and ensure safety. Another limitation of STER is that effective tunneling in the stomach can only be performed on the greater curvature of the gastric antrum or the lesser curvature of the gastric corpus. It is very difficult to successfully establish the tunnel in the gastric fundus, lesser curvature of the gastric antrum, or the greater curvature of the gastric corpus, successfully (6). In addition, the entry of the submucosal tunnel should be located 3 cm orally to the tunnel exit on the serous side of the stomach, mainly because of gastric flexibility. However, EFTR can be conducted across the stomach because there is no need for submucosal tunneling. Thus, when deciding between STER and EFTR for the treatment of gastric tumors, the location of the tumor should be considered.

Although the basic concept of NOTES emerged in 1994, it was not taken up in the following decades and did not become widely applied until recently (15). With the development of the flexible endoscope, endoscopic surgery provided an opening door for NOTES application and initiated the superscope era (16,17). Because flexible endoscopes provide greater freedom for tissue access, manipulation, traction, resection, and suturing, NOTES has the potential to be applied in more areas of surgery. The advantage of NOTES technique based on STER is preservation of the mucosal layer to reduce the risk of leaking, secondary infection, and stricture formation at the resection site (18,19). During STER-based NOTES, a mucosal flap valve serving as a protective barrier is used between the gastrointestinal lumen and peritoneal cavity, allowing safe access to extra-gastrointestinal lesions by a submucosal tunnel (20,21). Based on our experience, to increase the STER-based NOTES success rate and avoid adverse events when treating large SMTs, preoperative EUS/CT evaluation of the tumor, adequate preparation, and an experienced endoscopist are important. GISTs should be resected due to their malignant potential, particularly for those > 2.0 cm (22). Most EGISTs are aggressive, and their prognosis is poor (2). Thus, timely removal of EGISTs and tumors with a predominant extraluminal growth type is necessary. Traditionally, laparoscopic or surgical resection remains the main method (23). The STER technique has revolutionized NOTES practices, as it allows endoscopists to be able to treat certain extra-gastrointestinal tumors in a minimally invasive manner. In this study, 11 EGISTs were successfully removed by using the STER-based NOTES technique, which all achieved en bloc resection. Two adverse events occurred, both associated with large tumor size and long procedure time. Intraoperative bleeding was effectively treated during the procedure, and retroperitoneal emphysema in a patient who had received CO2 insufflation resolved spontaneously within 1–2 hours. We believe that STER-based NOTES may potentially offer a safer alternative for EGISTs compared with EFTR. However, further studies are needed to validate it. During NOTES, the entrance of the submucosal tunnel can be enlarged, and the endoloop-clips technique can be used to close the entrance. An experienced endoscopist is also important for completing this procedure smoothly and successfully.

We are fully aware that this study has several limitations. First, our inclusion of patients was not by random allocation, and the number of patients was small, and the different tumors are undoubtedly associated with the preference of endoscopists. Secondly, we not only compared STER with EFTR but also further explored different technical variants of STER (with or without NOTES), which inevitably introduced heterogeneity and may have reduced the overall statistical power of our study. Finally, the retrospective study design in this study is used, where selection bias was inevitable. A large, prospective, randomized controlled study should be performed in the future to validate our findings.

CONFLICTS OF INTEREST

Guarantor of the article: Tao Chen, MD.

Specific author contributions: T.C. and M.X.: the project was conceived and designed. S.L.: manuscript was written. S.L. and Z.C.: proof-reading, verification of references were performed. S.L., Z.C., L.H., A.X., Z.Z., X.D., S.Y., H.Z., L.Z., J.L.: data collection was performed. S.L. and Z.C.: data analysis was performed. All authors read and approved the final manuscript.

Financial support: This work was supported by grants from the Shanghai Municipal Health Commission (2024ZDXK0001 [M.X.]), the Medical Discipline Construction Project of Pudong Health Committee of Shanghai (PWZxq2022-6 [M.X.] and 2024-PWXZ-07 [M.X.]), Outstanding Young Medical Talents and Outstanding Projects of Shanghai Municipal Health Commission (20224Z0016 [T.C.]), and the grants from Shanghai East Hospital [2024-DFZD-005 (M.X.)]. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. All other authors disclosed no financial relationships relevant to this publication.

Potential competing interests: None.

IRB approval statement: Our study did not require further ethics committee approval as it did not involve animal or human clinical trials and was not unethical. In accordance with the ethical principles outlined in the Declaration of Helsinki, all participants provided informed consent before participating in the study. The anonymity and confidentiality of the participants were guaranteed, and participation was completely voluntary.

Study Highlights.

WHAT IS KNOWN

• ✓ Endoscopic resection of gastric submucosal tumors is minimally invasive.

• ✓ Submucosal tunneling endoscopic resection can preserve the integrity of the gastrointestinal tract.

WHAT IS NEW HERE

• ✓ Tumor size > 2 cm and endoloop-clips suture were significantly associated with long-time procedure.

• ✓ Patients who undergo endoscopic full-thickness resection have a shorter procedure time but longer postoperative hospital stay than those who undergo submucosal tunneling endoscopic resection (STER).

• ✓ Natural Orifice Transluminal Endoscopic Surgery based on STER is successful for extra-gastrointestinal stromal tumors.

Supplementary Material

ct9-16-e00869-s003.docx ^{(22KB, docx)}

ct9-16-e00869-s004.docx ^{(19.2KB, docx)}

ct9-16-e00869-s005.pdf ^{(593.7KB, pdf)}