OUTCOMES OF ENDOSCOPIC SUBMUCOSAL DISSECTION FOR PREVIOUSLY ATTEMPTED COLORECTAL LESIONS: AN INTERNATIONAL MULTICENTER EXPERIENCE

Background: The most common complication after endoscopic mucosal resection (EMR) is delayed bleeding (DB), especially in the proximal colon. Randomized controlled trials in high volume centers suggest that prophylactic clipping (PC) of the resection defect reduces DB in patients with a high DB risk. Guidelines already recommend PC for proximal polyps, despite being technical difficult and expensive. We aimed to evaluate the value of PC in patients receiving EMR for proximal flat polyps in reducing DB in daily clinical practice.
Methods: We performed a randomized controlled trial in 19 Dutch hospitals with patients referred for EMR of lateral spreading and sessile polyps ≥ 20mm in the proximal colon. Patients were randomly assigned (1:1) to groups treated with PC (intervention group) or no PC (control group). PC was standardized in tutorial meetings focusing on approximating the resection margins with aligning clips 5-10mm apart. The primary endpoint was clinically significant DB defined as hematochezia necessitating emergency department presentation, hospitalization, or re-intervention within 30 days post-EMR, which was analyzed according to the intention-to-treat principle. The trial is registered at ClinicalTrials.gov, NCT03309683.
Results: Between May 15, 2018 and December 14, 2021, 356 patients with a median polyp size of 30mm (IQR 25,40) were included of whom 179 were randomly assigned to the control group and 177 to the intervention group. DB occurred in 11 (6.1%) patients of the control group and in 16 (9.0%) patients of the intervention group (p=0.30). Endoscopists reported complete defect closure in 70.6% of cases. There were no differences between the control and intervention group in serious adverse events including perforation (two versus one, p=0.57), post polypectomy syndrome (zero versus three, p=0.08) and intensive care unit admission (one versus one). No deaths were reported.
Conclusion: PC did not reduce DB in patients undergoing EMR for large lateral spreading and sessile polyps in the proximal colon. Therefore, this study demonstrates that the burden of laborious and expensive PC is not justified in daily clinical practice.
Funding: The CLIPPER trial is investigator initiated and is financially supported by the Dutch Digestive Foundation (MLDS). Olympus (Japan) contributed Quick Clip Pro endoclips for this trial.

Introduction
The US Multi-Society Task Force on Colorectal Cancer recommends referral to an endoscopist experienced in advanced polypectomy for management of large colorectal polyps. It is known that prior manipulation of polyps such as biopsy, attempted resection, or tattoo placement increases the risk of fibrosis and adverse outcomes including incomplete endoscopic mucosal resection (EMR). Delays in referral and scheduling for definitive endoscopic resection are often encountered in practice. It is unclear if timing to between the index procedure and definitive resection affects outcomes of previously manipulated polyps.

Methods
All patients who underwent EMR between 2016 and 2021 were identified from a prospectively-maintained procedure database at a tertiary referral center. Index colonoscopy reports were reviewed and only previously manipulated polyps >10 mm referred for EMR were included. Patient, polyp, procedure, and outcome-related variables were retrospectively collected. Timing between index colonoscopy and definitive resection was calculated. Polyps referred and resected within 6 weeks were compared to those referred and resected after.

Results
Of 320 procedures screened, 247 lesions were included; 131 were females (54%). At the time of referral for EMR, 201 (69%) had been biopsied, 29 (12%) had previous attempted resection, and 22 (9%) had a tattoo extending to the base of the lesion. At EMR, fibrosis was noted in 62 (25.3%). Mean number of weeks between index colonoscopy and definitive resection was 9.1 ± 10.5. There were no significant differences in age, gender, polyp size, polyp location between polyps resected within 6 weeks and those resected after [Table 1]. Previously manipulated polyps referred and resected within 6 weeks of index colonoscopy had a significantly lower incidence of fibrosis (19.2%) versus those resected after (31.2%), p=0.030. A multivariate analysis controlling for polyp size showed that resection within 6 weeks continued to show 49% lower odds of fibrosis (p=0.028). Polyps with fibrosis had significantly lower en-bloc resection rates (9.7% vs. 22.4%, p=0.028), longer procedure time (72 vs. 52 minutes, p<0.0001), and lower Sydney resection quotient (3.9 vs. 8.2, p=0.001). There was no significant difference in adverse events or recurrence [Table 2].

Conclusion
In this study of previously manipulated polyps >10mm undergoing EMR, we found that delay between referral and definitive resection beyond 6 weeks was associated with a higher risk of fibrosis. Presence of fibrosis was associated with inferior outcomes including longer procedure time, lower en-bloc resection rate, and lower Sydney resection quotient. To our knowledge, this is the first series examining timing of definitive resection and EMR outcomes. If externally validated, our findings support expedited definitive resection for previously manipulated lesions.

Background: Endoscopic resection is highly effective for large colorectal neoplasia removal. However, recurrent or residual lesions are common and pose a clinical challenge. Salvage endoscopic submucosal dissection (ESD) has been recommended as a treatment option for previously attempted colorectal lesions (PACL). ESD has the advantage of achieving en bloc resection, yet the presence of extensive submucosal fibrosis in PACL impacts procedural difficulty and resection outcomes. A few small single center studies, mostly from Asia, have evaluated the role of salvage ESD for PACL, but data from Western centers is limited.

Aims: To evaluate the efficacy and outcomes of salvage ESD for the treatment of PACL in the Western population.

Methods: Consecutive patients who underwent ESD for the management of PACL between 05/2014-10/2022 at 19 tertiary care centers (15 North America, 2 South America, 2 Europe) were included. Primary outcomes were rate of en bloc and R0 resection. Secondary outcomes were rates of adverse events (AEs) [severity graded per Classification for Adverse events Gastrointestinal Endoscopy (AGREE)], procedure-related mortality, need for surgical management, and local recurrence.

Results: A total of 360 PACL treated by ESD were included (Table 1). En bloc rate was 89.7%, R0 81.1%, and curative resection of 79.4%. Overall, AEs occurred in 41 (11.4%) patients, most common being intraprocedural perforation (n=17; 4.7%). One patient required surgical management and 16 were managed endoscopically. Severe intraprocedural bleeding occurred in 3 (0.8%), all managed endoscopically. The mean ± SD procedure time was 94.7 ± 58 minutes. In 291 (80.8%) patients, ESD was performed in an outpatient setting; the remaining patients were admitted for observation (median (IQR) length of hospital stay (LOS): 1 (0-2) days). 21 patients (5.8%) had post-ESD AEs at the median of 7 (1-36) days, most commonly being delayed bleeding (1.9%). A total of 4 (1.1%) patients required surgery (1 intraprocedural perforation, and 3 post-ESD AEs [colonic stricture, fistula formation, and abscess]. Of 281 patients with follow-up data, local recurrence occurred in 14 (5%) patients after a median follow-up of 49 (25, 72) weeks. There was no mortality from procedure-related or colorectal cancer (Table 2).

Conclusion: Salvage ESD is highly effective in the treatment of PACL. Our data showed a high rate of en bloc and R0 resection for this challenging lesion, with very low overall rate (1.1%) of surgery. When performed by experts, AEs were uncommon and mostly managed endoscopically. These results support role of ESD for PACL in the Western setting.