Defining endoscopic candidacy for intermediate size rectal neuroendocrine tumors

Abstract

Endoscopic resection is standard for small, well-differentiated rectal neuroendocrine tumors, but management of intermediate lesions remains unsettled. In a large single-center cohort with propensity score matching, endoscopic treatment of grade 1 tumors measuring 1.0-1.5 cm achieved high negative-margin rates and no observed recurrences over a median 54-month follow-up, mirroring outcomes for lesions < 1 cm. Lymphovascular invasion was absent in the intermediate group, and endoscopic submucosal dissection was used more often than mucosal resection. These data support endoscopic resection as a feasible organ preserving option in carefully staged, well differentiated intermediate size tumors. Priorities now include prospective, multicenter validation; standardized pre-resection staging with high-resolution endoscopy and endoscopic ultrasound; refined risk stratification incorporating depth and lymphovascular invasion; pragmatic surveillance schedules; and assessment of patient-reported outcomes, function, and costs. Developing decision tools and targeted training to optimize endoscopic technique may further expand safe, individualized care, and health economic analyses.

Key Words: Rectal neuroendocrine tumor; Endoscopic submucosal dissection; Endoscopic mucosal resection; Propensity score matching; Lymphovascular invasion; Surveillance

Core Tip: A matched analysis from a high-volume center suggests that well differentiated rectal neuroendocrine tumors measuring 1.0-1.5 cm can be managed safely with endoscopic resection when selection and technique are optimized. The absence of lymphovascular invasion and the lack of observed recurrences in the intermediate group argue for an organ preserving pathway with structured staging and follow up.

TO THE EDITOR

A matched analysis from a high-volume center suggests that well differentiated rectal neuroendocrine tumors (NETs) measuring 1.0-1.5 cm can be managed safely with endoscopic resection when selection and technique are optimized. The absence of lymphovascular invasion (LVI) and the lack of observed recurrences in the intermediate group argue for an organ preserving pathway with structured staging and follow up.

Kim et al[1] conducted a single center retrospective cohort with 1:10 propensity score matching (age, sex, and resection method) comparing 27 grade one rectal NETs (rNETs) measuring 1.0-1.5 cm with 225 lesions smaller than one centimeter. After matching, the R0 resection rates were 92.6% and 97.8%, respectively; no recurrences were observed in the intermediate size group, and recurrence free survival did not differ (P = 0.48). Endoscopic submucosal dissection (ESD) was used more often for intermediate lesions before matching, but its use was similar across groups after matching (45%-48%). In addition to summarizing Kim et al[1], we propose an actionable pathway for 1.0-1.5 cm rNETs in which standardized staging, morphology in relation to technique [endoscopic full-thickness resection (EFTR) for cap-capturable non-lifting lesions and endoscopic intermuscular dissection (EID) for deeper margins] are defined, margin terminology and referral thresholds are established enabling reproducible, organ-preserving care across centers.

ENDOSCOPIC FEATURES AND RESECTION METHODS OF RNETS

The rNETs typically present as small, smooth, yellowish subepithelial nodules in the distal rectum, sometimes with a central depression or umbilication; high-resolution and image-enhanced endoscopy help delineate margins and pit or vascular patterns and reduce inadvertent snare polypectomy[2]. Selective endoscopic ultrasound is useful when morphology suggests submucosal invasion or when size approaches 10-20 mm, and cross-sectional imaging is reserved for higher grade features or suspicious nodes in line with European Neuroendocrine Tumor Society guidance[3]. Resection options include modified endoscopic mucosal resection (EMR) techniques such as cap-assisted and ligation-assisted EMR, and underwater EMR, which achieve high en bloc and R0 rates for lesions ≤ 10 mm with short procedure time[2,4]. ESD provides wider and deeper margins and is preferred for 10-20 mm or broad-based lesions where en bloc histology is required. EFTR can be considered for non-lifting or previously manipulated lesions in selected cases.

DEFINE CANDIDACY WITH A UNIFORM PRE-RESECTION STAGING BUNDLE

We suggest a simple bundle for all lesions between one and one and a half centimeters: (1) Meticulous high-resolution inspection with cap assistance and dye or virtual chromoendoscopy to confirm a submucosal origin and to map borders; (2) Targeted endoscopic ultrasound when the pit or vascular pattern suggests deeper invasion; and (3) Cross sectional imaging when grade two features or suspicious nodes are present. This bundle would operationalize what Kim et al[1] already achieved implicitly in a high-volume setting and would help replicate their low risk profile beyond a single center. Risk factors for nodal spread in this size range (size, grade, depth, and LVI) are well established and can guide selection for endoscopic therapy and surgery[5]. Where grade 2 and 3 features are present, obtain cross-sectional imaging; prefer pelvic magnetic resonance imaging (MRI) for nodal staging (computed tomography use reserved to when MRI unavailable/contraindicated).

MATCH TECHNIQUE TO GOALS WHILE RECOGNIZING THAT OUTCOMES WERE SIMILAR

In the matched analysis, negative margins were obtained in more than ninety percent of intermediate lesions, and recurrence free survival did not differ by technique, even though ESD was used more often for intermediate tumors[1]. A practical approach is to favor en bloc resection with an adequate deep margin for sessile lesions with a broad base or when suspicion for submucosal extension is present, while reserving modified mucosal resection methods for small, well circumscribed domes. Contemporary reviews also support tailoring the method to lesion morphology, operator expertise, and the need for histologic assessment of margins[2]. Consider clip-assisted EFTR with an over-the-scope full-thickness resection device within our en bloc strategy for resection of rNETs. Use EFTR for lesions that can be completely drawn into the cap (< 1.5 cm; 20-25 mm, with adequate distance from the dentate line) are amenable to en bloc resection and are suitable for non-lifting scars or when a vertical margin is essential[6]. In a multicenter registry, EFTR achieved 100% complete resection margins without major adverse events and a median tumor size of 8 mm[7], supporting EFTR as an organ-preserving approach for well-differentiated tumors. Comparative data suggest that for small rNETs, EFTR can achieve R0 rates approaching those of transanal endoscopic microsurgery while being a more practical choice in terms of procedure time where expertise and device availability is a barrier, in a pragmatic alternative[8]. Newer modalities to facilitate en bloc/R0 resection when clipping/EFTR are not suitable include underwater EMR for ≤ 10 mm rNETs (similar resection quality, shorter procedure time and lower resource use), traction-assisted ESD to improve efficiency/margins, and refinements including anchored snare-tip EMR, and pocket-creation ESD in cases of submucosal fibrosis[4,9-11], enabling individualized technique selection based on morphology and depth to preserve function and minimize costs when oncologic risk is low.

USE A STRUCTURED MARGIN AND SURVEILLANCE PATHWAY

Kim et al[1] observed very few positive margins after endoscopic resection, and no recurrence among intermediate lesions despite occasional margin positivity; one metachronous recurrence occurred in the smaller size group during follow up. When the deep plane appears clear endoscopically and margin positivity likely reflects cautery artifact, careful endoscopic surveillance may be reasonable, whereas clearly transected submucosa should prompt immediate re resection. Prior work links lymphatic invasion and incomplete excision to higher risk and can inform these decisions[12,13]. Immunostaining such as D2-40 may improve detection of lymphatic invasion and refine risk estimates[14].

SURVEILLANCE THAT IS PRAGMATIC AND MEASURABLE

Margin definitions and pathway. We define R0 resection as histologically negative lateral and vertical (deep) margins. A positive vertical margin means tumor at the deep cut edge; an indeterminate vertical margin means margins cannot be reliably assessed owing to cautery artifact or specimen fragmentation. We avoid ambiguous terms and instead report histology (positive or indeterminate vertical margin) and whether a clear deep plane was achieved endoscopically. When R0 is achieved in well-differentiated rNETs ≤ 15 mm without LVI or nodal suspicion, endoscopic surveillance is appropriate. If the vertical margin is indeterminate but the deep plane appeared clear, early scar inspection with biopsy and possible salvage resection is reasonable. If the vertical margin is positive, or LVI or grade 2 and 3 is present, escalate management[3,5,12,13].

Achieving R0 in deep submucosal invasion. For lesions with optical or endoscopic ultrasonography (EUS) signs of deep submucosal invasion where a wider vertical margin is required, EID, which is dissection between the inner circular and outer longitudinal muscle layers, can yield a deeper en bloc specimen and reduce positive deep margins in selected distal rectal lesions. Small series show feasibility in rNETs; restrict EID to expert operators with pre-procedure staging and surgical backup[15].

When to consider additional radical resection with regional lymph-node dissection. After endoscopic resection, consider radical surgery when any of the following are present: (1) Grade 2 and 3; (2) LVI; (3) Positive vertical margin; (4) Muscularis propria invasion; (5) Suspicious nodes on MRI or EUS; and (6) Size ≥ 1-2 cm with adverse histology. Grade independently predicts nodal metastasis in 1-2 cm rNETs[5], while prior series identified size, depth, and LVI as key surgical triggers[12,13]. Evidence after non-curative endoscopic resection shows that positive or indeterminate margins and LVI often prompt radical surgery, with favorable long-term outcomes[16]. Pelvic MRI is preferred for nodal staging when higher-grade or adverse features are present[3].

Centers adopting this pathway should track predefined key performance indicators to ensure safe scale-up: En bloc and R0 rates overall and by technique (modified EMR, ESD, EFTR, EID), vertical-margin positivity, nodal upstaging after salvage or radical surgery, adverse events patient-reported outcomes (bowel function and quality of life) and cost per R0 resection provide complementary patient and system level signals. A quarterly review of these metrics, with feedback into case selection and training, can harmonize outcomes across operators and institutions.

A RESEARCH AGENDA THAT MATTERS TO PATIENTS

Future work should validate these findings across centers with standardized reporting of grade, depth, and LVI; test decision tools that combine endoscopic morphology with histology; incorporate patient reported outcomes and anorectal function; and perform cost analyses that compare endoscopic treatment and surgery in this intermediate size window[1,5,17].

CONCLUSION

The analysis by Kim et al[1] highlights the opportunity for endoscopic therapy in well-differentiated rNETs measuring 1.0-1.5 cm when staging confirms favorable biology. The challenge is to produce higher-level evidence from prospective multicenter cohorts or a pragmatic registry using a uniform pre-resection staging bundle (high-resolution inspection with chromoendoscopy, selective EUS, and cross-sectional imaging for grade 2 features or suspicious nodes), centralized pathology for depth and LVI (with adjunctive immunostains), and prespecified endpoints: En bloc/R0 resection, nodal metastasis, recurrence-free survival, adverse events, anorectal function, quality of life, and costs to compare comparative effectiveness of ESD, modified EMR, and EFTR with stratification by grade, depth, and LVI, as well as credentialed training and decision tools matching morphology to technique. A shared margin management and surveillance algorithm that has been tested for efficiency and patient experience can then serve as a standard of care as evidence accrues, facilitating faster, more widespread, safe, and organ-preserving treatment.