Endoscopy 2021; 53(07): 710-712
DOI: 10.1055/a-1309-1721
Editorial

Rectal neuroendocrine tumors: is it all on first sight?

Referring to Chen L et al. p. 702–709
Louis de Mestier
1  Department of Gastroenterology and Pancreatology, ENETS Centre of Excellence, Beaujon Hospital (APHP), Clichy, France
2  Université de Paris, Centre of Research on Inflammation, INSERM U1149, Paris, France
,
Guillaume Cadiot
3  Department of Hepato-Gastroenterology and Digestive Oncology, Robert Debré Hospital and Reims-Champagne-Ardenne University, Reims, France
› Author Affiliations

The incidence of rectal neuroendocrine tumors (NETs) has been steadily rising during recent decades, reaching approximately 1.2 /100 000 per year in 2012 [1]. This is notably due to the increasing number of colonoscopies performed, with a prevalence of rectal NETs of about 0.17 % in adults undergoing screening colonoscopy [2]. As a consequence, all endoscopists face situations of the per-procedure discovery of a rectal polyp evocative of a rectal NET, or will face it during their activity.

“…advanced imaging methods stand as an interesting area of research and can surely help in the diagnosis and characterization of rectal NETs, in order to move toward better case-by-case management.”

In the present issue of Endoscopy, Chen and colleagues [3] report the use of the endoscopic appearances of rectal NETs to predict their malignant behavior and prognosis in a retrospective study including 309 consecutive patients with a rectal NET. Three endoscopic features, i. e. tumor size (< 1 cm, 1 – 2 cm, and > 2 cm), shape (sessile, semi-pedunculated, and flat or fungating), and mucosal surface appearance (smooth, depression, or erosion/ulceration), were independently associated with advanced disease (invasion through the muscularis propria, or regional or distant metastases). These three characteristics were used to compute a nomogram-based score. A total score ≥ 110 accurately predicted the presence of advanced disease with a sensitivity, specificity, and accuracy above 91 % (area under the curve [AUC] of 0.953 and 0.960 in the training and validation sets, respectively), and was associated with worse prognosis. This score, which combines three easily assessed endoscopic features of a rectal NET, could help identify those that are not suitable for an endoscopic resection, but must instead be considered for radical surgery because of the high risk of concurrent metastases ([Fig. 1]).

Zoom Image
Fig. 1 Proposed algorithm for the management of a rectal neuroendocrine tumor suspected during initial colonoscopy. This personal view from the authors relies on the current recommendations [5-6] and integrates the score proposed by Chen et al. [3] in the present issue of Endoscopy. CT, computed tomography; EMR, endoscopic mucosal resection; ESD, endoscopic submucosal dissection; EUS, endoscopic ultrasonography; LNM, lymph node metastases; MRI, magnetic resonance imaging; PET, positron emission tomography; TEM, transanal endoscopic microsurgery. * EMR with cap aspiration or ligation assistance.

Importantly, rectal NETs predicted to have a low risk of advanced disease (most rectal NETs < 1 cm in size, and some of those < 2 cm) could be good candidates for local resection, although it requires an appropriate resection technique [1] [3] [4]. As local excision of these small rectal NETs is often also the definitive treatment, endoscopic resection should best be performed radically. It is now well recognized that polypectomy or endoscopic mucosal resection (EMR) do not provide sufficient resection margins [1] [5]. Instead, advanced EMR (with cap aspiration or ligation assistance) or endoscopic submucosal dissection (ESD) appear more appropriate for rectal NETs < 10 mm, and ESD or transanal endoscopic microsurgery (TEM) for those ≥ 10 mm ([Fig. 1]) [1] [5] [6]. Should these advanced resection techniques not be available on site, it is preferable to refer the patient to an expert endoscopy center rather than attempt standard polypectomy/EMR.

The prognostic evaluation performed by initial endoscopic evaluation must then be supplemented by pathological analysis of the locally resected rectal NET, in order to decide whether to perform additional explorations or not. No additional explorations are required for rectal NETs that are grade 1, < 10 mm in size, T1, and R0 after the initial resection. Conversely, if factors predictive of lymph node metastases exist, it is appropriate to perform additional explorations to search for metastatic involvement ([Fig. 1]) [1] [6].

Hence, the endoscopic recognition of rectal NETs is of paramount importance for appropriate prognostic evaluation and therapeutic decision-making, in order to increase the chances of curative resection. However, recent series reported that only 18 % – 33 % of rectal NETs were suspected and managed as NETs during the initial endoscopy [4] [7]. Rectal NETs that were initially suspected underwent upfront resection using appropriate techniques in a higher proportion of cases than unsuspected rectal NETs [4], which resulted in a significantly higher R0 rate (69 % vs. 23 % if a rectal NET was not suspected; P < 0.001) [7]. However, while training in the recognition of other adenomatous polyps (Paris and Kudo classifications) is well advanced, training in the recognition of rectal NETs currently remains in its early stages. In response to this area of unmet need, the European Neuroendocrine Tumor Society have proposed that endoscopic training programs should include a rectal NET recognition module and such a development has started [5].

Finally, the basis for endoscopic diagnosis of rectal lesions is shifting from histopathology, with subsequently made diagnosis, to advanced imaging methods, including the use of narrow-band imaging [8]. Such advanced imaging methods stand as an interesting area of research and can surely help in the diagnosis and characterization of rectal NETs, in order to move toward better case-by-case management.



Publication History

Publication Date:
24 June 2021 (online)

© 2021. Thieme. All rights reserved.

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany