Endoscopic submucosal dissection and hybrid endoscopic submucosal dissection for stage 1 rectal neuroendocrine tumors

Abstract

BACKGROUND

The technical complexity and potential for complications associated with endoscopic submucosal dissection (ESD) pose limitations on the widespread use of this procedure for stage 1 rectal neuroendocrine tumors (NETs), despite its high success rate in achieving complete resection (R0).

AIM

To examine the results of ESD and hybrid ESD, a simpler adaptation of the ESD technique, for stage 1 rectal NETs.

METHODS

Seventy-nine patients with 84 lesions of clinical stage 1 rectal NETs who received treatment at Sun Yat-sen University Cancer Center from January 2010 to June 2021 were reviewed retrospectively.

RESULTS

Sixty-one lesions in 58 patients were treated with ESD, while 23 in 21 patients were treated with hybrid ESD. The 84 rectal NETs had a median diameter of 8 (5) mm (range, 3-20 mm), with the median lesion size 8 (5) mm for ESD and 8 (4) mm for hybrid ESD (P = 0.359). For ESD, the median duration of procedure was 46.00 (14.00) minutes, while for hybrid ESD, it was 32.00 (15.00) minutes (P < 0.001). Both the ESD and hybrid ESD groups had identical rates of en bloc resection (100.00% vs 100.00%, P = 1.000), R0 resection (86.89% vs 86.96%, P = 1.000), perforation (1.64% vs 0.00%, P = 1.000), and delayed bleeding (1.64% vs 4.35%, P = 0.475). After a median of 27.50 (30.00) months of observation, neither group had recurrence.

CONCLUSION

For endoscopic excision of stage 1 rectal NETs, both ESD and hybrid ESD were well tolerated and produced positive results, with similar efficacy and safety.

Key Words: Rectal neuroendocrine tumors; Endoscopic submucosal dissection; Hybrid endoscopic submucosal dissection; R0 resection; Safety

Core Tip: Despite a high complete resection (R0) rate, its technical difficulty and risk of complications limit the widespread use of endoscopic submucosal dissection (ESD) for stage 1 rectal neuroendocrine tumors (NETs). One simplified modification to ESD is hybrid ESD (ESD with snaring). This research analyzed 84 lesions who were diagnosed with rectal NETs and treated with ESD or hybrid ESD. We found that both ESD and hybrid ESD were safe and effective for endoscopic resection of stage 1 rectal NETs, and hybrid ESD can be a good alternative to ESD for R0 resection of stage 1 rectal NETs.

INTRODUCTION

Rectal neuroendocrine tumors (NETs) are uncommon but potentially malignant tumors. The detection rate with colonoscopy is around 0.05%-0.17%, although the worldwide incidence of rectal NETs has increased several to a dozen times as a result of the significant rise in colonoscopy popularity over the last several decades[1-3]. Clinicians are becoming more and more concerned about the variable prognosis of rectal NETs due to factors such as tumor size, grade, depth of invasion and lymphatic vascular. It has been noticed that between 80% and 90% of rectal NETs are quite small when first discovered during a routine colonoscopy[4,5].

Low-risk lymph node metastases and local endoscopic excision are now accepted for rectal NETs less than 20 mm in diameter that are limited to the mucosa and submucosa[4,6,7]. To achieve a more histologically complete excision of the lesion and a lower incidence of remaining tissue, endoscopic submucosal dissection (ESD) is also recommended for the endoscopic treatment of rectal NETs[6,8-10]. However, ESD has not been commonly utilized for the removal of rectal NETs, presumably because it is technically more difficult, takes longer, and has a greater risk of major consequences including perforation when compared to other techniques. Hybrid ESD (ESD with snaring) is a straightforward variant on the original method[11]. While the first part of the hybrid ESD process is performed similarly to conventional ESD, the last step includes the use of a snare to capture and remove the undissected, constricted submucosal tissue.

This research analyzed patients with stage 1 rectal NETs who were treated with ESD or hybrid ESD. En bloc and R0 resection rates, surgery time, complications, histopathological findings and follow-up data were all evaluated.

MATERIALS AND METHODS

Patients

A single-center retrospective cohort study was conducted on a cohort of 79 patients with 84 lesions of clinical stage 1 (cT1N0M0)[7] rectal NETs who received treatment at Sun Yat-sen University Cancer Center from January 2010 to June 2021 (Figure 1). Prior to the surgery, invasion depth, as well as to assess the presence of local lymph node involvement or distant metastases were ruled out by magnetic resonance imaging, abdominopelvic computed tomography or endoscopic ultrasonography in all patients. The excision of the lesions was performed by ESD or hybrid ESD, followed by histological examination. The criteria for inclusion in the study were: (1) The lesion was situated in the rectum, specifically at a distance of less than 15 cm from the anus; (2) The lesion exhibited a lengthy diameter of less than 20 mm; (3) Prior to the endoscopic resection, examinations using endoscopic ultrasonography/magnetic resonance imaging/computed tomography did not reveal any evidence of invasion into the muscularis propria, involvement of local lymph nodes, or presence of distant metastases; and (4) the diagnosis was confirmed by histopathology. The exclusion criteria included: (1) Patients combined with other types of malignant tumors; and (2) participants with a follow-up period of less than 6 months.

Figure 1 Flowchart of the inclusion of patients in this study. NETs: Neuroendocrine tumors; ESD: Endoscopic submucosal dissection; EMR: Endoscopic mucosal resection.

Prior to undergoing endoscopic resection, all patients were duly informed about the possible risks and advantages associated with the procedure. In addition, they provided written informed permission as a prerequisite for their participation. The techniques conducted in this study adhere to the guidelines outlined in the 1964 Helsinki Declaration and subsequent ethics that are pertinent to the research. The retrospective research was approval by the Ethics Committee of the Institutional Review Board of the Sun Yat-sen University Cancer Center, approval No. SL-B2023-722-01 and informed consent was exempt.

ESD and hybrid ESD procedures

The study used a single-channel endoscope (manufactured by Olympus Co, Tokyo, Japan) for ESD. A transparent hood was affixed to the distal end of the endoscope. Following a thorough examination of the lesion, a submucosal injection was administered in the vicinity of the lesion in order to elevate it from the muscular layer. The solution used in this study consisted of sodium hyaluronate that was diluted with a normal saline solution at a ratio of 1:4. Additionally, a tiny quantity of indigo carmine was introduced into the solution. Following the administration of the injection, an incision was made in the mucosa around the elevated lesion using a stationary and adaptable snare knife. A surgical procedure was performed in which the mucosa was carefully cut, ensuring a distance of about 2 mm to 3 mm from the outside edge of the lesion. This was done to ensure that the lateral resection margin was free from any tumor tissue. Following a mucosal incision that encompassed around 50% of the lesion's diameter, the submucosal tissue underlying the lesion was separated from the muscular layer using the same surgical instrument. Following the procedure of submucosal dissection, the remaining portion of the lesion that was not initially raised was subsequently elevated with an additional submucosal injection. Subsequently, the mucosal incision was fully executed. The lesion was fully removed from the muscle layer with further submucosal dissection. Following the resection procedure, the visible blood vessels inside the ulcer that occurred during ESD were effectively eliminated with the use of either hemostatic forceps or argon plasma coagulation. The electrosurgical machines utilized in this study were manufactured by Elektromedizin GmbH, Tuebingen, Germany.

The hybrid ESD procedure entails the first execution of ESD, followed by the use of snaring as the final resection technique. The ESD procedure was conducted in a manner consistent with previous descriptions, with the exception that submucosal dissection was deferred until the lesion had been fully detached from the muscular layer. Instead, the submucosal dissection was advanced in a circumferential manner only up to the point where the remaining undissected submucosal tissue accounted for less than 1/4 of the lesion’s width. Then, the enclosed, restricted submucosal tissue was then captured and excised by the use of an electrical current with snaring.

The selection between ESD or hybrid ESD was determined by the judgment of the endoscopist. The primary indications for hybrid ESD were twofold. Firstly, it was used for faster resection in cases which snaring the remaining undissected submucosal portion appeared to be a straightforward task. Secondly, hybrid ESD was employed to complete the resection in cases where the location of the lesion or the presence of submucosal fibrosis made complete submucosal dissection challenging. In such instances, performing ESD alone could have potentially increased the risk of perforation.

An ESD specialist refers to an endoscopist who has autonomously conducted over 100 instances of colorectal ESD. An ESD beginner refers to an endoscopist who has autonomously conducted fewer than 100 instances of colorectal ESD.

Evaluation of endoscopic resection and complications

Histological examination was conducted on the resected tumors to assess parameters such as lymphovascular invasion, the depth of invasion, and the status of the resected margin of the specimen, determining whether it was positive or negative. The assessment of mitotic rate and Ki-67 index was conducted in accordance with the guidelines provided by the World Health Organization[6,7].

En bloc resection was originally described as a surgical procedure involving the removal of a tumor or lesion in a single, intact piece. A R0 resection is operationally defined as a surgical procedure in which both the lateral margin and vertical margin are found to be negative upon microscopic inspection. Curative resection refers to a surgical procedure known as R0 resection, which involves the complete removal of a tumor without any evidence of vascular invasion or perineurium invasion. Non-curative resection was operationally defined as a surgical procedure with evidence of positive vertical margins, positive lateral margins, vascular invasion, or perineurium invasion upon microscopic inspection of the resected tumors.

The procedure time was operationally defined as the duration from the initiation of submucosal injection and the full completion of lesion excision. The size of the lesion was assessed endoscopically, with open biopsy forceps serving as a reference. Delayed bleeding is characterized by the presence of hematochezia and/or melena after the conclusion of endoscopic resection, necessitating another endoscopy for the purpose of achieving hemostasis. The identification of perforation was determined by the examination of endoscopic and/or radiographic evidence. The American Society of Gastrointestinal Endoscopy (ASGE) lexicon classification was used for assessment of the severity of complications.

Follow-up after endoscopic resection

For patients with non-curative resection of stage 1 rectal NETs with G2/G3, a salvage radical surgical procedure was recommended, while for patients with non-curative resected stage 1 rectal NETs with G1, salvage local resection (endoscopic resection or transanal endoscopic microsurgery) or close follow-up was recommended[10].

The clinical results were determined by retrospective examination of the medical data. The first clinical follow-up was conducted within 1 weeks to 2 weeks subsequent to the endoscopic resection procedure. The assessment was conducted to determine the incidence of problems, such as delayed bleeding. It was suggested that individuals who have had curative resection get a colonoscopy examination every six months during the first year after the surgical procedure. In the absence of any indications of recurrence, it was advised to do an annual colonoscopy check. Colonoscopy was advised at intervals of 3 months, 6 months, and 12 months after the procedure for patients who had undergone R0 resection but exhibited vascular invasion, or for patients who had undergone non-R0 resection and declined further surgical intervention. Subsequently, an annual colonoscopy was indicated in the absence of any local recurrence. In the event that a potential recurrence was seen during a follow-up colonoscopy, it was advised to proceed with a biopsy.

Statistical analysis

Typically, continuous variables that follow a normal distribution were represented using means and standard deviations. On the other hand, for continuous variables that did not follow a normal distribution, medians and interquartile ranges were employed. Categorical variables were often represented as proportions. The variables were compared via statistical tests such as t-tests, Kruskal-Wallis tests, χ² tests, or Fisher's exact tests, depending on the nature of the data. The missing data was removed.

A binary logistic regression analysis was conducted to determine the risk variables associated with non-R0 resection. The odds ratio (OR) for each item was computed along with a 95% confidence interval (CI). The significance level was set at P < 0.05 for a two-tailed test. Statistical testing was conducted using SAS v9.4 (SAS Institute Inc., United States).

RESULTS

Baseline characteristics

The summary of the baseline characteristics analyses may be seen in Table 1. The investigation included a collective of 84 lesions among 79 patients. A total of 61 lesions in 58 patients had ESD, whereas 23 lesions in 21 patients had hybrid ESD. There were 33 (56.89%) men in the ESD group and 12 (57.14%) men in the hybrid ESD group (P = 0.984). The mean ages were 45.38 ± 13.15 years and 42.61 ± 14.38 years (P = 0.424) and the median lesion size was 8 (5) mm and 8 (4) mm in the respective groups (P = 0.359).

Table 1 The characteristics of 84 lesions with rectal neuroendocrine tumors, n (%).

Characteristic	ESD (n = 61)	Hybrid ESD (n = 23)	t/χ2 value	P value
Lesion size, mm	8 (5)	8 (4)	0.842	0.359
Age (year), mean ± SD	45.38 ± 13.15	42.61 ± 14.38	0.800	0.424
Procedure time, minute	46.00 (14.00)	32.00 (15.00)	25.093	< 0.001
Follow-up time, month	24.00 (23.00)	34.00 (32.00)	5.010	0.024
Male	33 (56.89)	12 (57.14)	0.000	0.984
Location
Anus ≤ 5 cm	24 (39.34)	8 (34.78)	0.147	0.701
Anus > 5 cm	37 (60.66)	15 (65.21)
Operator
Specialist	53 (86.89)	17 (73.91)	2.024	0.192
Beginner	8 (13.11)	6 (26.09)
Perforation	1 (1.64)	0 (0.00)	0.382	1.000
Delayed hemorrhage	1 (1.64)	1 (4.35)	0.527	0.475
Histologic grade
G1	55 (90.16)	20 (86.96)	0.672	0.700
G2	6 (9.84)	3 (13.04)
Invasion depth
Mucosa	5 (8.20)	1 (4.35)	0.541	1.000
Submucosa	56 (91.80)	22 (95.65)
En-bloc resection	61 (100.00)	23 (100.00)
R0 resection	53 (86.89)	20 (86.96)	0.000	1.000
Vascular invasion	0 (0.00)	1 (4.35)	2.684	0.274
Recurrence	0 (0.00)	0 (0.00)

ESD: Endoscopic submucosal dissection.

Procedure-related results

The average duration of the procedure was substantially higher for ESD compared to hybrid ESD, with median times of 46.00 (14.00) minutes and 32.00 (15.00) minutes, respectively (P < 0.001) (Table 1). The en bloc resection rate achieved a 100.00% success rate for both procedures.

There was one case of perforation, which was classified as moderate adverse event according to the ASGE lexicon, resulting in prolongation of the hospital stay by 4 days. The perforated case was effectively treated using the OTSC technique, obviating the need for further surgical intervention. Two cases of delayed bleeding occurred, both of which were classified as moderate adverse events according to the ASGE lexicon and were treated with repeat endoscopy (endoscopic hemostasis was used for the medical management of the two cases of delayed bleeding). The incidence of perforation (1.64% vs 0.00%, P = 1.000) and delayed bleeding (1.64% vs 4.35%, P = 0.475) exhibited no significant differences between the two groups.

Histopathologic results

A total of 11 subjects had positive vertical margins, leading to a R0 resection rate of 86.90% (73/84) (Table 1). Only a single instance was completely resected (R0), although vascular involvement was evident. Hence, the rate of curative resection was 85.71% (72/84). In the ESD group, the rate of R0 resection was 86.89% (53/61 cases), whereas in the hybrid ESD group, the rate of R0 resection was 86.96% (20/23 cases) (P = 1.000). Based on the categorization of mitoses and the Ki-67 proliferation index, it was determined that 89.29% (75/84) of cases fell into the G1 category, whereas 10.71% (9/84) of cases were categorized as G2. In 7.14% (6/84) of the cases, the lesions were limited to the mucosal layer, whereas in 92.96% (78/84) of the cases, the lesions had penetrated into the submucosal layer.

Subgroup analysis according to lesion size

The results of outcome analyses based on lesion size are shown in Table 2. The research included a collective of 56 lesions measuring less than 10 mm and 28 lesions measuring between 10-20 mm. In both lesions measuring less than 10 mm and lesions measuring 10-20 mm, hybrid ESD demonstrated comparable outcomes to ESD in terms of having no significant difference for en bloc resection rate, R0 resection rate and complication rate. Nevertheless, hybrid ESD exhibited was more efficient compared to ESD, in terms of having shorter procedure time.

Table 2 Outcomes between endoscopic submucosal dissection and hybrid endoscopic submucosal dissection groups according to lesion size, n (%).

Characteristic	1A (< 10 mm, n = 56)				1B (10-20 mm, n = 28)
	ESD (n = 41)	Hybrid ESD (n = 15)	t/χ2 value	P value	ESD (n = 20)	Hybrid ESD (n = 8)	t/χ2 value	P value
Procedure time, minute	44.00 (16.00)	30.00 (13.00)	14.905	< 0.001	50.50 (10.00)	33.00 (20.00)	11.471	< 0.001
Follow-up time, month	22.00 (24.00)	30.00 (27.00)	1.426	0.232	26.50 (31.50)	48.00 (22.00)	4.905	0.027
Perforation	0 (0.00)	0 (0.00)	/	/	1 (5.00)	0 (0.00)	0.415	1.000
Delayed hemorrhage	1 (2.43)	0 (0.00)	0.373	1.000	0 (0.00)	1 (12.50)	2.593	0.286
En bloc resection	41 (100.00)	15 (100.00)	/	/	20 (100.00)	8 (100.00)	/	/
R0 resection	37 (90.24)	13 (86.67)	0.147	0.654	16 (80.00)	7 (87.50)	0.219	1.000
Recurrence	0 (0.00)	0 (0.00)	/	/	0 (0.00)	0 (0.00)	/	/

ESD: Endoscopic submucosal dissection.

Follow-up data

The subsequent data is succinctly described in Table 3. Out of the 12 cases of non-curative resection, three cases with positive vertical margins underwent salvage surgery, resulting in the absence of residual tumor cells upon microscopic examination. Additionally, five cases with positive vertical margins underwent mucosal biopsy in the postoperative scar, revealing no residual tumor cells upon microscopic examination. Furthermore, three cases with positive vertical margins and one case with vascular invasion underwent colonoscopy follow-up without mucosal biopsy. The median duration of follow-up for the 12 patients was 25.00 (18.00) months, during which no instances of tumor recurrence were observed.

Table 3 Follow-up of the 12 non-curable resection cases.

Number	Gender	Age	Size (mm)	Method	Grade	Invasion depth	Salvage therapy	Microscopic examination	Follow-up (month)	Recurrence
1	Male	41	10	ESD	G1	Submucosa	Colonoscopy	/	76	No
2	Female1	20	5	hybrid ESD	G1	Submucosa	Colonoscopy	/	30	No
3	Male	34	10	ESD	G1	Submucosa	Salvage local resection	Negative	24	No
4	Female	46	8	ESD	G1	Submucosa	Salvage radical surgery	Negative	16	No
5	Female	32	6	hybrid ESD	G1	Submucosa	Colonoscopy	/	12	No
6	Male	27	10	ESD	G1	Submucosa	Salvage local resection	Negative	34	No
7	Female	51	15	hybrid ESD	G1	Submucosa	Colonoscopy + biopsy	Negative	38	No
8	Female	31	8	hybrid ESD	G1	Submucosa	Colonoscopy + biopsy	Negative	25	No
9	Male	53	5	ESD	G1	Submucosa	Colonoscopy	/	20	No
10	Male	35	4	ESD	G1	Submucosa	Colonoscopy + biopsy	Negative	29	No
11	Male	36	10	ESD	G1	Submucosa	Colonoscopy + biopsy	Negative	15	No
12	Male	55	5	ESD	G1	Submucosa	Colonoscopy + biopsy	Negative	10	No

¹The case had vascular invasion, and the other 11 cases had positive vertical margins.

Colonoscopy: Colonoscopy follow-up without mucosal biopsy; Colonoscopy + biopsy: Colonoscopy follow-up with mucosal biopsy; ESD: Endoscopic submucosal dissection.

Factors influencing non-R0 resection

The factors associated with non-R0 resection for stage 1 rectal NETs are shown in Table 4. The proportion of beginners was significantly higher in the non-R0 resection group compared to in the R0 resection group (45.45% vs 12.33%, P = 0.016). There were no significant differences between the two groups in terms of age, lesion size, sex, tumor location, resection method, pathological grade and invasion depth.

Table 4 Factors associated with non-R0 resection for stage 1 rectal neuroendocrine tumors, n (%).

Characteristic	Non-R0 resection (n = 11)	R0 resection (n = 73)	t/χ2 value	P value
Age (year), mean ± SD	40.09 ± 9.71	45.38 ± 13.88	1.210	0.229
Lesion size, mm	8.00 (5.00)	8.00 (4.00)	0.186	0.666
Procedure time, minute	49.00 (18.00)	42.00 (16.00)	0.721	0.399
Male	7 (63.63)	38 (55.88)	0.232	0.749
Location
Anus ≤ 5 cm	4 (36.36)	28 (38.36)	0.016	1.000
Anus > 5 cm	7 (63.63)	45 (61.64)
Operator
Specialist	6 (54.55)	64 (87.67)	7.553	0.016
Beginner	5 (45.45)	9 (12.33)
Resection method
ESD	8 (72.73)	53 (72.60)	0.000	1.000
Hybrid ESD	3 (27.27)	20 (27.29)
Grade
G1	11 (100.00)	64 (87.67)	1.519	0.598
G2	0 (0.00)	9 (12.33)
Invasion depth
Mucosal	0 (0.00)	6 (8.22)	0.974	1.000
Submucosal	11 (100.00)	67 (91.78)

ESD: Endoscopic submucosal dissection.

Multivariate analysis was conducted to evaluate the variables that influence non-R0 resection (Table 5). In comparison to individuals with expertise in ESD, those classified as ESD beginners exhibited ORs of 7.31 (95%CI: 1.70-31.42, P = 0.008) for non-R0 resection. The non-R0 resection was not influenced by other variables such as age, sex, lesion size, location, and resection method, since they were not shown to be independent factors.

Table 5 Univariate and multivariate analysis for factors influencing non-R0 resection.

Characteristic	Univariate analysis		Multivariate analysis
	OR (95%CI)	P value	OR (95%CI)	P value
Age (year), mean ± SD	0.97 (0.92-1.02)	0.229	0.97 (0.92-1.03)	0.284
Lesion size, mm	1.03 (0.85-1.25)	0.730	1.01 (0.81-1.26)	0.913
Gender
Male	1.00		1.00
Female	0.72 (0.19-2.70)	0.631	0.74 (0.17-3.23)	0.688
Location
Anus ≤ 5 cm	1.00		1.00
Annus > 5 cm	1.09 (0.29-4.06)	0.899	1.18 (0.28-5.05)	0.824
Operator
Specialist	1.00		1.00
Beginner	5.93 (1.50-23.48)	0.011	7.31 (1.70-31.42)	0.008
Resection method
ESD	1.00		1.00
Hybrid ESD	0.99 (0.24-4.12)	0.993	0.81 (0.16-3.97)	0.792

CI: Confidence interval; OR: Odds ratio; ESD: Endoscopic submucosal dissection.

DISCUSSION

The current research proved that both ESD and hybrid ESD are viable and successful treatment methods for stage 1 (cT1N0M0) rectal NETs. Hybrid ESD was employed in cases where the execution of full submucosal dissection posed challenges or carried a heightened risk of perforation compared to ESD[11]. Consequently, hybrid ESD offers distinct advantages for lesions situated in anatomically challenging sites or exhibiting submucosal fibrosis[12,13]. Besides, it is evident that hybrid ESD is inherently less complex than ESD[14,15]. Consequently, we propose that hybrid ESD may be a viable alternative for the management of rectal NETs < 20 mm. It should be noted, however, that the present investigation only included a collective of 28 lesions ranging in size from 10 mm to 20 mm. The majority of the lesions seen ranged from 10 mm to 15 mm, with an average size of 11.5 ± 2.7 mm. Hence, the results of our study are more likely favor the safety and efficacy of hybrid ESD as a viable treatment option for rectal NETs measuring less than 15 mm in size.

In the present investigation, the rate of R0 resection was 86.89% in the ESD, a finding that aligns with past research in this field. Pan et al[16] conducted a comprehensive assessment of 14 papers with a total of 823 cases involving endoscopic resection of rectal NETs measuring less than 10 mm. The findings of their meta-analysis revealed an overall R0 resection rate of 90.2%. Specifically, the R0 resection rate for ESD was 84.1%. Kim et al[17] provided a comprehensive summary of 277 cases with rectal NETs measuring less than 10 mm. The authors reported that the percentage of R0 varied between 72% and 74% across various endoscopic resection techniques. Li et al[18] found that the R0 endoscopic resection rate of rectal NETs less than 10 mm among 101 cases was 88.7%, which was also similar with our results.

There was no recurrence in any of the 12 cases in our analysis that had a positive vertical resection margin or vascular invasion. Microscopic examination revealed no trace of tumor cells in the five patients who had mucosal biopsy in the postoperative scar or the three patients who underwent salvage surgery. Therefore, whether salvage treatment is necessary for non-R0 endoscopic resection of stage 1 rectal NETs? A total of 407 cases of rectal NETs were studied retrospectively by Moon et al[19], with 148 cases receiving non-R0 endoscopic resection. Among the 148 cases, 14 underwent salvage surgery, and 134 opted for follow-up monitoring. There was no significant difference in the local recurrence rate between the non-R0 resection group and the R0 resection group (2.1% vs 1.2%) at a median follow-up of 56.8 months. The 322 cases of rectal NETs that were studied retrospectively by Cha et al[20] included 76 cases of non-R0 resection, 38 cases of which had salvage surgery, and 38 cases of which were solely followed up. There was a median follow-up of 40.5 months, with no evidence of either local recurrence or distant metastases among the 322 patients. According to a retrospective study by Li et al[21], 54 of 428 cases included non-R0 resection with no salvage therapy for rectal NETs. Patients were followed for a median of 38 months, during which no evidence of local tumor recurrence or distant metastasis was found. Because cauterization may eliminate nearby tumor cells, some endoscopists may not see pathologically positive resection margins as indicative of recurring or residual malignancies[16,22,23]. Some endoscopists may have decided against salvage therapy and instead opted for routine follow-up colonoscopy and biopsy of the resection site. It is possible that the cheap cost and high availability of colonoscopy in China have contributed to a decrease in the number of salvage procedures.

Tumor size, invasion depth, pathological grade, and endoscopic methods are all potential risk factors that might influence R0 resection, as indicated in a small number of prior studies[17,21,24-27]. An independent risk factor for non-R0 endoscopic resection was endoscopic morphological manifestation of rectal NETs, according to Wang et al[25]. Flat or slightly raised lesions (type II) and depression or ulcer on the surface of lesions (type III) were associated with a considerably greater risk of non-R0 resection than protruding lesions (type I), with corresponding ORs of 6.65 (95%CI: 1.24-35.74, P = 0.03), and 6.81 (95%CI: 1.06-43.56, P = 0.04). Tumor size was a risk factor for non-R0 resection of stage 1 rectal NETs by Zheng et al[24], with a 64% increase in the probability of non-R0 resection for every 1mm increase in tumor size (OR = 1.64, 95%CI: 1.59-4.39, P < 0.001). Kim et al[17] demonstrated that the likelihood of endoscopic R0 resection of stage 1 rectal NETs dropped by 10% for every 1 mm rise in tumor size (OR 0.90, 95%CI 0.82-0.99, P = 0.02). Central depression on the surface of the tumor was associated with an increased probability of non-R0 endoscopic resection, as shown by Choi et al[27] (OR = 11.53, 95%CI: 2.38-55.92, P = 0.002). Uni- and multivariate analysis of our data revealed that, unlike tumor size and morphological presentation, ESD operating experience was linked with non-R0 resection. However, only 11 non-R0 endoscopic resection instances were included in our analysis. These negative findings should be approached with caution due to the limited sample size.

The present research exhibits some strengths and limitations that need careful assessment. One notable strength of this research is its comparison of the safety and effectiveness of ESD vs hybrid ESD in the treatment of stage 1 rectal NETs. Hybrid ESD is inherently less complex than ESD, and the present study found that hybrid ESD could be equally effective as ESD in achieving R0 resection, while also being less time-consuming, for stage 1 rectal NETs. Hybrid ESD could be advantageous in scenarios where the presence of submucosal fibrosis complicates the ESD process and poses a higher risk of perforation. Thus, the present results demonstrate a significant finding with clear clinical implications. Nevertheless, this research is constrained by its restricted sample size, increasing the risk of a type II error, indicating that there may be differences that the study failed to detect. Especially, the incidence of both a specific complication and overall complications is low, which make it difficult to find significant differences. As a result, it is imperative that the findings and conclusions drawn from our investigation are needed to be substantiated by the implementation of multicenter studies encompassing a larger sample size. Second, it is important to note that the present investigation was retrospective and single-center, potentially introducing possible selection bias when comparing the outcomes of ESD with hybrid ESD. Third, a subset of patients had a limited duration of follow-up. Fourth, the choice for ESD or hybrid ESD was not standardized and was made by the appointed endoscopist. Therefore, there is selection bias in this study.

CONCLUSION

It was determined that both ESD and hybrid ESD are deemed as safe and efficacious methods for the removal of stage 1 rectal NETs. Hybrid ESD may serve as a viable substitute for ESD in achieving R0 resection of rectal NETs measuring less than 20 mm. This is particularly relevant in cases where accomplishing a thorough submucosal dissection proved challenging or carried a greater likelihood of perforation when using ESD. It is necessary to conduct extensive, prospective, randomized studies to validate these recommendations.