Pocket Creation Method of Endoscopic Submucosal Dissection to Ensure Curative Resection of Rectal Neuroendocrine Tumors

Since the first report by Oberndorfer in 1907 [1], the term “carcinoid” has been used for neuroendocrine tumor (NET) of the gastrointestinal tract. According to the 2010 revision of the World Health Organization classification [2], a pancreatic/gastrointestinal tract tumor with characteristics and phenotype of the endocrine system is called a neuroendocrine neoplasm (NEN), with further division into NET and neuroendocrine carcinoma (NEC) according to the grade of malignancy and clinical prognosis. NET is a well-differentiated tumor, heterogeneity and malignancy are relatively low, and endoscopic treatment is often used, particularly for rectal NET. Previously, endoscopic mucosal resection (EMR) was mainly performed for rectal NET [3], but stumps showing positive results for residual tumor cells, local recurrence, and tumor remnants were the most important problems [4].

EMR with a ligation device (EMR-L) [5, 6] and endoscopic submucosal dissection (ESD) [6] have been used to treat rectal NET and have been reported to be more useful than EMR [7, 8]. On the other hand, the pocket creation method (PCM) was reported by Hayashi et al. [9, 10] as an ESD method for colon tumor, and has been described as a safe technique that offers a high en bloc resection rate compared with conventional colonic ESD [11]. In addition, the characteristics of the procedure make this method useful to ensure pathological evaluation in the vertical direction. PCM thus seems to have strong potential as a method for resecting NET, which exists within the very thin submucosal layer between the NET and muscularis propria, as the deep part of the stump is easily exposed during the procedure. The present study performed ESD using PCM for rectal NET and evaluated the complete resection rate.

From November 2015 to February 2017, a total of 4 patients with rectal NET underwent ESD using PCM at Dokkyo Medical University Saitama Medical Center. The pathological complete resection rate was examined for cases in which ESD was performed using PCM for rectal NET. In addition, rates of complications such as perforation and delayed postoperative bleeding were examined. Lesions considered for endoscopic treatment were tumors less than 10 mm [12] that were restricted to the submucosal layer as evaluated by endoscopic ultrasonography and showed no distant metastasis/lymph node metastasis on computed tomography. For PCM, we used a LUCERA-ELITE video processor and GIF-260J videoendoscope (Olympus, Tokyo, Japan). We used an upper gastroscope for the PCM procedure, because this method offers better operability than colonoscope in the rectum. For endoscopic products, we used an MD-47951 SB soft hood (Sumitomo Bakelite, Tokyo, Japan) and a DualKnife electrosurgical knife (KD-650Q; Olympus). PCM was performed as previously described [9, 10]. Briefly, the initial mucosal incision was made after injection of 0.4% sodium hyaluronate solution. Submucosal dissection was performed to create the pocket. Finally, the remaining surrounding incision was added to finish the procedure (Fig. 1). This study was approved by our institutional review board.

ESD using PCM proved technically feasible in all patients. Patient characteristics and technical results are shown in Table 1. The mean tumor size was 6.5 ± 2.1 cm, and the grade was G1 in all 4 cases. Endoscopically, all cases were resected en bloc, and histopathological examination indicated that complete resection was achieved in all cases. In particular, a sufficient amount of the submucosal layer remained for accurate pathological examination in the deep part of the tumor (Fig. 2). One case showed vascular invasion but has not shown any recurrence as of the time of writing. No complications such as perforation or delayed postoperative bleeding were encountered.

In the endoscopic treatment of rectal NET, EMR treatment was initially mainstream. However, EMR often proved problematic in that pathologically confirmed complete resection was not achieved. EMR-L and conventional ESD techniques for rectal NET have represented advances, but incomplete resection is still often reported [13].

We likewise experienced inadequate results with EMR-L and ESD, and several cases have required follow-up surgery. Moon et al. [14] reported that 7 of the 148 patients showed positive margins and therefore needed additional treatment. In patients with long-term follow-up with positive margins following endoscopic treatment without additional treatment, the local recurrence rate was 1.4% (2/141).

On the other hand, the PCM method reported in 2014 by Hayashi et al. [10] appears to be effective for laterally spreading tumors of the large intestine. Traction is important in ESD but difficult to apply in every situation. However, insertion of the transparent hood into the pocket in the PCM allows easy application of traction and countertraction [11]. PCM allows easy recognition of the muscular layer and thus placement of the incision just above this layer, facilitating dissection of the submucosal layer and reducing the rate of positive deep margins compared with EMR-L or conventional ESD. While PCM is more expensive and time-consuming than EMR-L, follow-up of a positive stump or incomplete resection is difficult and increases costs. Sufficient efficacy of the initial intervention is thus important and must be weighed against the potential economic and time costs of procedures for additional treatment and surveillance.

Although NET are not considered relatively low-grade tumors, NET with lymph node metastasis have the same prognosis as adenocarcinomas with metastasis [15]. Tumor diameter ≥10 mm and vascular invasion have been reported as risk factors for both lymph node metastasis and distant metastasis. According to the current consensus, adaptation of NET endoscopic treatment is considered to be warranted for lesions ≤10 mm in diameter [15]. A sufficient amount of the submucosal layer is necessary for accurate pathological examination, but resection of the tumor with a sufficient sample of submucosa is difficult in ESD or EMR-L. ESD by PCM can be performed leaving a sufficient amount of the submucosal layer on the deep margin side of the resected specimen, and the distance to the vertical stump can be secured.

In cases of NET with a small diameter, treatment can be performed simply and quickly using EMR-L. However, which treatment should be chosen to treat rectal NET according to tumor size remains unclear. If the distance to the vertical stump is considered difficult to secure because the tumor diameter is too large, ESD by PCM seems preferable. Proper use of PCM and EMR-L depending on the size of the tumor will therefore be a subject for future study. Interestingly, we observed vascular invasion in a 5-mm NET in in 1 of our patients. According to a report by Kitagawa et al. [16], addition of immunohistochemistry for D2-40 and Elastica van Gieson staining revealed lymphovascular invasion in 48.5% (32/66) of NET cases ≤5 mm in diameter. These included 1 case of lymph node metastasis, so careful follow-up is necessary for cases showing lymphovascular invasion.

In conclusion, PCM should be considered when treating NET of appropriate size.

Informed consent was obtained from the patients for publication of this report. The ethics committee at Dokkyo Medical University Saitama Medical Center approved this study in accordance with the ethical guidelines of the 2013 Declaration of Helsinki.

The authors have no conflicts of interest to disclose.

I.K. performed endoscopic therapy, Y. Katayama wrote and edited the manuscript, T.K. collected data, K.T., Y. Kusano, Y.F., and R.O. provided patient care, S.B. performed pathological studies, and M.T. edited the manuscript.