- Email: [email protected]
Large Sessile Serrated Polyps Can Be Safely and Effectively Removed by Endoscopic Mucosal Resection
Accepted Manuscript Large Sessile Serrated Polyps Can Be Safely and Effectively Removed by Endoscopic Mucosal Resection Aarti K. Rao, MD, Roy Soetikno, MD, MS, Gottumukkala S. Raju, MD, Phillip Lum, Robert V. Rouse, MD, Tohru Sato, MD, Diane Titzer-Schwarzl, James Aisenberg, MD, Tonya Kaltenbach, MD, MS PII: DOI: Reference:
S1542-3565(15)01412-3 10.1016/j.cgh.2015.10.013 YJCGH 54512
To appear in: Clinical Gastroenterology and Hepatology Accepted Date: 4 October 2015 Please cite this article as: Rao AK, Soetikno R, Raju GS, Lum P, Rouse RV, Sato T, Titzer-Schwarzl D, Aisenberg J, Kaltenbach T, Large Sessile Serrated Polyps Can Be Safely and Effectively Removed by Endoscopic Mucosal Resection, Clinical Gastroenterology and Hepatology (2015), doi: 10.1016/ j.cgh.2015.10.013. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Large Sessile Serrated Polyps Can Be Safely and Effectively Removed by Endoscopic Mucosal Resection
RI PT
Aarti K. Rao MD1, Roy Soetikno MD, MS1, Gottumukkala S. Raju MD2, Phillip Lum2, Robert V. Rouse MD1, Tohru Sato MD1, Diane Titzer-Schwarzl1, James Aisenberg MD3, and Tonya Kaltenbach MD, MS1
SC
1. Veterans Affairs Palo Alto Health Care System and Stanford University School of Medicine, Palo Alto, CA; 2. The University of Texas MD Anderson Cancer Center, Houston, TX; 3. Icahn School of Medicine at Mount Sinai, New York, NY
Grant support: none
M AN U
Abbreviations: SSP (sessile serrated adenoma/polyp), EMR (endoscopic mucosal resection), CRC (colorectal cancer), APC (argon plasma coagulation), HP (hyperplastic polyp), TSA (traditional serrated adenoma)
TE D
Corresponding Author: Tonya Kaltenbach MD MS Clinical Assistant Professor of Medicine (Affiliate) Stanford University School of Medicine Veterans Affairs Palo Alto 3801 Miranda Avenue, GI 111, Palo Alto, CA 94304 650-814-0629 [email protected]
AC C
EP
Conflict of Interest: Aarti K. Rao: no conflict of interest to declare Roy Soetikno has received non-salaried research support and is a consultant for Olympus America, Inc. Gottumukkala S. Raju: no conflict of interest to declare Phillip Lum: no conflict of interest to declare Robert V. Rouse: no conflict of interest to declare Tohru Sato: no conflict of interest to declare Diane Titzer-Schwarzl: no conflict of interest to declare James Aisenberg: no conflict of interest to declare Tonya Kaltenbach has received non-salaried research support and is a consultant for Olympus America, Inc. Author Contributions: Aarti Rao • Acquisition of data • Analysis and interpretation of data • Drafting of the manuscript
ACCEPTED MANUSCRIPT
•
Critical revision of the manuscript for intellectual content
RI PT
Roy Soetikno • Study concept and design • Acquisition of data • Analysis and interpretation of data • Drafting of the manuscript • Critical revision of the manuscript for intellectual content • Study supervision
Phillip Lum • Acquisition of data
M AN U
SC
Gottumukkala S. Raju • Study concept and design • Acquisition of data • Critical revision of the manuscript for intellectual content • Study supervision
Robert V. Rouse • Acquisition of data • Critical revision of the manuscript for intellectual content
TE D
Tohru Sato • Acquisition of data • Critical revision of the manuscript for intellectual content
EP
Diane Titzer-Schwarzl • Acquisition of data
AC C
James Aisenberg • Study concept and design • Critical revision of the manuscript for intellectual content Tonya Kaltenbach • Study concept and design • Acquisition of data • Analysis and interpretation of data • Statistical analysis • Drafting of the manuscript • Critical revision of the manuscript for intellectual content • Study supervision
ACCEPTED MANUSCRIPT
Abstract
Background & Aims
RI PT
As many as 50% of large sessile serrated adenomas/polyps (SSPs) are removed
incompletely, which is significant because SSPs have been implicated in development of interval cancers. It is unclear if endoscopic mucosal resection (EMR) is an optimal
SC
method for removal of SSPs. We assessed the efficacy and safety of removal of SSPs
M AN U
≥10 mm using a standardized inject-and-cut EMR technique.
Methods
We performed a retrospective analysis of colonoscopy data, collected over 7 years (2007–2013) at 2 centers, from 199 patients with proximal colon SSPs ≥10 mm (251
TE D
polyps) removed by EMR by 4 endoscopists. The primary outcome measure was local recurrence. The secondary outcome measure was safety.
EP
Results
At the index colonoscopy, patients had a median of 1 serrated lesion (range, 1–12) and
AC C
1 non-serrated neoplastic lesion (range, 0–15). The mean SSP size was 15.9±5.3 mm; most were superficially elevated (84.5%) and located in the ascending colon (51%), and 3 SSPs (1.2%) had dysplasia. Surveillance colonoscopies were performed on 138 patients (69.3%) over a mean follow up time of 25.5±17.4 months. Of these patients, 5 had local recurrences (3.6%; 95% confidence interval, 0.5–6.7%), detected after 17.8±15.4 months, with a median size of 4 mm. No patients developed post-procedural
ACCEPTED MANUSCRIPT
bleeding, perforation, or advanced colon cancer or had a death related to the index colorectal lesion during the study period.
RI PT
Conclusions
Inject-and-cut EMR is a safe and effective technique for the resection of SSPs. Less than 5% of patients have a local recurrence, which is usually small and can be treated
SC
endoscopically.
AC C
EP
TE D
M AN U
KEY WORDS: endoscopy; tumor; early detection; malignancy
ACCEPTED MANUSCRIPT
Background The endoscopic features of sessile serrated adenomas/polyps (SSPs) predominantly flat in shape and with indistinct borders1-3 - make them difficult to detect
RI PT
and remove. SSPs are more likely to be incompletely resected compared to
conventional adenomas (31% vs. 7.2%) and larger lesions have even higher rates (47.6%) of incomplete removal.1 The development of interval cancers has been
SC
attributed to incompletely resected lesions.4 A recent study of large proximal serrated polyps showed that 3 of 18 patients with such lesions biopsied but not completely
M AN U
removed developed interval colorectal cancer (CRC) at the same location.5 At present, the optimal method for removal of SSPs is unknown. The inject-andcut EMR technique, which utilizes saline injection into the submucosa to lift the lesion of interest and subsequent stiff electrocautery snare to resect the polyp, may improve the
TE D
likelihood of complete resection. The outcomes of EMR for serrated lesions, however, are largely unavailable.6, 7 Prior studies of EMR have primarily focused on its safety and efficacy for the resection of conventional adenomas.8-13 We examined the safety and
AC C
centers.
EP
efficacy of EMR for proximal SSPs ≥10mm performed by multiple endoscopists at 2
Methods
Patients and study design We systematically reviewed outcomes data of a serrated lesion cohort who underwent colonoscopy by 1 of 4 endoscopists with significant experience in EMR at 2
ACCEPTED MANUSCRIPT
centers (Veterans Affairs [VA] Palo Alto Health Care System and University of Texas MD Anderson Cancer Center) over a 7-year period (2007-2013). We identified the serrated lesion cohort using a Filemaker Pro pathology database, and ascertained data
RI PT
using an endoscopic database and computerized medical record. We included serrated lesions with specific criteria: (1) pathologic diagnosis of SSP as defined by consensusmodified World Health Organization criteria14, (2) size ≥10mm, (3) location proximal to
SC
and including the splenic flexure, (4) endoscopic resection by the inject-and-cut EMR technique, and (5) due for surveillance colonoscopy based on multi-society
M AN U
recommendations (6 months for piecemeal resection and 3 years for en bloc resection). Serrated lesions with pathologic diagnosis of hyperplastic polyp (HP) and traditional serrated adenoma (TSA) were excluded. We analyzed the recurrence rate of SSPs in patients who had at least one surveillance colonoscopy. The institutional review boards
Procedure
TE D
at both sites approved the study.
EP
Patients completed standard bowel preparation prior to the procedure and received moderate sedation with midazolam and fentanyl or meperidine. All patients
AC C
were prescribed split bowel preparation the evening prior to and the morning of the colonoscopy. VA Palo Alto patients received polyethylene glycol-based electrolyte solution and/or magnesium citrate and MD Anderson patients received polyethylene glycol-based solution and bisacodyl. The endoscopists performed colonoscopy using an adult or pediatric high definition colonoscope equipped with an auxiliary water jet (Olympus PCFH180AL, CFH180AL, CFH190AL) and carbon dioxide insufflation. Both
ACCEPTED MANUSCRIPT
sites used a standardized endoscopic resection approach that includes lesion assessment, inject-and-cut EMR techniques, immediate re-assessment and treatment of residual lesion (Figure 1).
RI PT
Specifically, we visualized the index SSP lesion using white light and then
enhanced visualization with narrow band imaging, followed by selective spray of 0.2% diluted indigo carmine onto the area of interest for suspected nonpolypoid serrated
SC
lesions, by a syringe through the accessory channel. Injection of a mixture of dilute indigo carmine and saline into the submucosa with a 25-gauge needle was used to
M AN U
create a submucosal bleb. We used a variety of 25-gauge needles (LDVI-25-240, Cook Medical Inc., Bloomington, Indiana; 00711811, US Endoscopy, Mentor, Ohio; M00518161, Boston Scientific, Marlborough, Massachusetts; Injector Force Max Injection Needle, Olympus America, Center Valley, Pennsylvania). Indigo carmine in the
TE D
injection solution also aided the rapid visual contrast of resection depth. The distended colon was slightly collapsed with suctioning of the lumen and the stiff electrosurgical snare (10-mm oval snare [SD-210–10] and/or 20-mm spiral snare [SD-230–20],
EP
Olympus America, Center Valley, Pennsylvania) was placed around the lifted area of interest to complete snare resection. We applied cautery for snare resection using the
AC C
Endostat II Microvasive electrosurgical unit (Boston Scientific, Marlborough, Massachusetts) at setting of 35 watts or the ERBE electrosurgical generator (VIO 330D; ERBE USA, Marietta, Georgia) using the fractionated cutting mode EndoCutQ (effect 3, cut duration 1, and cut interval 4 at VA Palo Alto; effect 3, cut duration 1, and cut interval 3 at MD Anderson Cancer Center) for snare resection. In cases where en bloc resection was not possible, we resected the lesion piecemeal. After resection, the site
ACCEPTED MANUSCRIPT
was closely examined and multiple photographs were taken to confirm eradication. Argon plasma coagulation (APC) at 60W and 1L/min flow on a continuous setting using the ERBE electrosurgical generator (VIO 330D; ERBE USA, Marietta, Georgia) was
RI PT
applied to the periphery and bridges in piecemeal resection, and to exposed superficial vessels. Prophylactic endoscopic clips (Resolution Clip, Boston Scientific, Marlborough, Massachusetts; Instinct Clip, Cook Medical Inc., Bloomington, Indiana; QuikClip 2,
SC
Olympus America, Center Valley, Pennsylvania) were selectively placed to close
Histopathology Analysis
M AN U
mucosal defects.
A blinded pathologist reviewed all identified serrated lesions (SSP, HP and TSA) that met inclusion criteria of size, location and resection technique. A diagnosis of SSP
TE D
required at least one abnormal crypt, defined as crypt dilation with a flattened/horizontal base and serration variably present throughout the crypt length.14 Synchronous lesions classified as non-serrated neoplasia included tubular adenomas, villous adenomas or
EP
cancer. These were sub-categorized as non-serrated advanced neoplasia if the tubular
AC C
adenoma was ≥10 mm or had villous, high-grade dysplastic or cancerous features.
Study Outcomes
We evaluated the efficacy of the inject-and-cut EMR technique for complete
resection of large SSPs by measuring local recurrence rate at surveillance colonoscopy as the primary endpoint. We defined local recurrence as lesion residual or regrowth at the index EMR location. Local recurrence was assessed with visualization of the post-
ACCEPTED MANUSCRIPT
EMR scar site with elective use of image enhancement (narrow band imaging or sprayed or injected indigo carmine) or biopsy. In cases where the prior EMR scar was not explicitly visible, we reviewed the index colonoscopy report and images to determine
RI PT
whether recurrence was grossly seen in the vicinity of the prior EMR site. We also
evaluated the safety of the EMR technique by secondarily measuring post-procedure bleeding, perforation and need for surgery. We defined index colonoscopy as the initial
SC
colonoscopy at which the serrated lesion was identified and resected. Subsequent
colonoscopies were categorized as surveillance. Patients were recalled for surveillance
M AN U
by letter and/or phone call. Patients at VA Palo Alto received a follow up call after the procedure and any complications or hospitalizations within 30 days after the procedure were recorded on a standardized form. At MD Anderson, patients were followed up with
Statistical Analysis
TE D
daily emails for 5 days and by telephone at 14 days.
We performed descriptive statistical analysis (STATA/SE 13.1 for Macintosh,
EP
Stata Corporation) for patients and polyps. We primarily analyzed local recurrence rate of SSPs. We report percentages for categorical variables and mean, standard deviation,
AC C
range and median for continuous variables.
Results
Patients
We identified 199 patients with 251 proximal SSPs that were ≥10mm and removed by inject-and-cut EMR. The characteristics of the patients are shown in Table
ACCEPTED MANUSCRIPT
1. Colonoscopies were performed in 129 patients by 3 endoscopists at VA Palo Alto and 70 patients by 1 endoscopist at MD Anderson. The majority were men (73.4%) and had a mean age of 61.0 ± 8.4 years. Ninety patients in the cohort (45%) underwent the
RI PT
index colonoscopy for average risk colorectal cancer screening. At the index examination, the patients had a median of 1 (range 1-7) proximal SSP ≥10mm. In addition, they had a median of 1 (range 0-15) non-serrated synchronous lesion.
SC
The characteristics of SSPs are shown in Table 2. The mean size SSP was 15.9 ± 5.3 mm (range 10-50mm, median 15mm). The majority was nonpolypoid superficially
M AN U
elevated (84.5%) and located in the ascending colon (51.0%). Three SSPs (1.2%, 95%CI:0.25-3.5%) had dysplastic histologic foci. These SSPs with dysplasia were 10, 12 and 20 mm in size.
TE D
Outcomes
At an average follow-up of over 2 years, SSPs removed with EMR had a low local recurrence rate, 3.6%, 95% CI = 0.5-6.7%. Overall, the median size of recurrence
EP
was 4mm (mean size 10.2 ± 9.6, range 3-25). There was no recurrence with high-grade dysplasia or cancer. Of the 5 SSPs that developed recurrence, the median size of the
AC C
index lesion was 20mm, only one had high-grade dysplasia and all were flat. Table 3 details the characteristics of each local recurrence. The majority of the cohort (138 of 199, 69.3% patients and 176 of 251, 70.1%
lesions) had surveillance colonoscopy at a mean of 25.5 months (range 2-74). The reasons for not undergoing surveillance were varied, and included death from unrelated causes, co-morbidities, advanced age and non-adherence to recommendations (Figure
ACCEPTED MANUSCRIPT
2). At surveillance, we found a mean of 1.1 ± 2.0 (range 0-13, median 0) metachronous non-serrated neoplastic lesions and 0.4 ± 1.0 (range 0-5, median 0) metachronous serrated lesions. The five patients with local recurrence had it detected at a mean
RI PT
surveillance interval of 17.8 months (range 4-37months). We successfully treated all of the local recurrences endoscopically – the majority (4/5) at the time of the first surveillance colonoscopy. One SSP showed diminutive recurrence at multiple
SC
surveillance colonoscopies until ultimately no recurrence was detected at the sixth colonoscopy exam. All lesions were completely eradicated based on macroscopic
M AN U
inspection of the polyp site perimeter post-EMR.
We resected the majority of lesions (86.5%, 217/251) en bloc. The mean size of lesions resected en bloc was 14.9 ± 4.0mm (range 10-30 mm). The mean size of lesions resected piecemeal was 22.2 ± 8.6mm (range 10-50 mm). Of the 176 polyps
TE D
that received surveillance, we resected 142 lesions en bloc (size 15.7 ± 4.2 mm [range 10-30 mm]) and 34 lesions piecemeal (size 21.1 ± 7.7 mm [range 10-50 mm]). Three of the 5 recurrences (60%) originated from index lesions that had been resected en bloc.
EP
Recurrence rate for lesions resected en bloc was 2.1% (3/142) and 5.9% (2/34) for piecemeal. While lesions resected piecemeal were more likely to recur than those
AC C
resected en bloc, the difference was not statistically significant (OR 2.9, p=0.23).
Adverse Events
No patients developed post-procedural bleeding, perforation, advanced colon
cancer or died related to the index colorectal lesion during the study period. We applied
ACCEPTED MANUSCRIPT
clips in 84.5% (n=212) to close the lesion defect. Of the 56 SSPs ≥20mm in the cecum or ascending colon, we clipped 48 (85.7%) lesions.
RI PT
Discussion
The findings of our two-site multi-endoscopist study show high rates of safe and complete resection of SSPs ≥10mm using the inject-and-cut EMR technique with a low
SC
(3.6%) recurrence rate and no complications. All five local recurrences were small (median 4mm) and four lesions were endoscopically cured after one surveillance
M AN U
colonoscopy. Our outcomes expand upon the previously published single-operator findings that EMR of large SSPs achieves recurrence rates similar to EMR of conventional adenomas.6 Our study further emphasizes the importance of EMR in the removal of SSPs and in addition illustrates the low prevalence of dysplasia in SSPs.
TE D
Our study provides important data on the outcomes of EMR in the resection of SSPs. In this study, we found 85% of index SSPs to be non-polypoid, and, thus, the need to use EMR to resect SSPs. A recent single endoscopist study using the inject-
EP
and-cut EMR technique showed low recurrence rates (8.7%) in 46 SSPs, similar to the low recurrence rate in our study.6 This Rex et al study examined SSPs ≥20 mm. SSPs
AC C
between 10 and 20 mm are much more frequent than SSPs ≥20mm, and a lesion size cut-off of 10mm is used in SSP surveillance guidelines. We, thus, focused on SSPs ≥10mm.
We found EMR to be safe and effective to remove SSPs in this first and largest
long-term US multi-site study. Our findings suggest that EMR for complete removal of SSPs is highly efficacious when performed by endoscopists who are facile with EMR.
ACCEPTED MANUSCRIPT
Our study is also in agreement with our prior paper that found the EMR technique to be effective in removing non-polypoid lesions8 and others, which reported recurrence to be small in size and occur infrequently.2, 8, 9, 13, 15, 16 In contrast, prior data showed
RI PT
polypectomy to be ineffective in achieving complete resection of SSPs. Pohl and
colleagues reported the inadequacies of hot snare resection of SSPs, with high rates of incomplete resections in SSPs compared to other neoplastic polyps (31.0% vs. 7.2%;
SC
p=0.001). Nearly half of all large (10-20mm) SSPs were incompletely removed.1 We suggest that EMR may mitigate the variability in incomplete resection of SSPs using
M AN U
standard polypectomy.
Our data support the current multi-society recommendations for follow up of SSA/P. Current surveillance guidelines suggest follow up in 3 years for SSPs ≥10mm, 1-3 years if there are 2 or more SSPs ≥10mm or dysplasia and 3-6 months for index
TE D
lesions resected piecemeal.14 The recurrence rate of SSPs after EMR is low (3.6%) and the majority (4 of 5) of recurrences were detected in larger SSPs ≥20mm. The mean time to detection of recurrence was approximately 1.5 years (range 0.25-3) and we
EP
detected 4 of 5 recurrences during the first surveillance colonoscopy at 4, 7, 9 and 37 months. One patient whose initial lesion was resected en bloc at the ileo-cecal valve did
AC C
not have recurrence identified until the second follow up examination. While lesions resected piecemeal were more likely to recur than those resected
en bloc, the difference was not statistically significant (OR 2.9, p=0.23). We found 60% (3 of 5) of our recurrences originated from en bloc-resected SSPs. This is in contrast to prior studies that have shown piecemeal EMR results in higher recurrence rates than en bloc EMR.2, 8 Although these numbers are too small to draw any conclusions,
ACCEPTED MANUSCRIPT
inconspicuous serrated pathology may be easily overlooked in this group compared to traditional adenomas. A subset of lesions resected en bloc may have other features that characterize them as high risk for earlier surveillance, such as size, but this requires
RI PT
further study.
We found a low rate of dysplasia in our population (1.2%). In comparison, prior endoscopic and pathologic studies described rates between 8.7% and 32.4%.6, 17, 18
SC
Burgess and colleagues found a 32.4% dysplasia rate in large SSPs ≥20mm. Their high dysplasia rates may reflect referral bias in a tertiary care center or an association
M AN U
between size ≥20mm and the presence of dysplasia. It might also be due to the smaller size of lesions in our cohort (median 15 mm) that had not yet grown enough to develop dysplasia.
Our findings show promise for the implementation of the EMR technique for
TE D
SSPs ≥10mm, which are associated with the development of CRC. Sessile serrated adenomas/polyps are important precursors of CRC via distinct pathways involving microsatellite instability and hypermethylation events. Prior studies have shown that
EP
compared to conventional adenomas, serrated lesions have a higher association with CRC and higher rates of incomplete resection, thought to be due to their flat shape with
AC C
indistinct borders.1, 20, 21 In fact, large serrated polyps ≥10mm have been shown to be the greatest risk factor in the development of CRC compared to age, presence of adenoma ≥10mm or ≥4 adenomas. This is especially true for proximal CRC.20 Guidelines suggest lesions ≥10mm are the highest risk for significant pathology.22 The findings from our phase I/II study are favorable for the safety and efficacy of EMR for large serrated lesions—with no complications and high endoscopic removal rates.
ACCEPTED MANUSCRIPT
Although we report the outcomes of multiple endoscopists who practice at a tertiary care medical center and an academic veterans hospital to improve generalizability of the technique to physicians and patients, our study has limitations.
RI PT
While detection of SSPs has improved in recent years due to increasing endoscopic and pathologic recognition, our recent study period (2007-2013) resulted in the
exclusion of a over one quarter of patients (71/270) who were not yet due for follow up
SC
at the time of data analysis. However, since data on resection of SSPs is sparse, we believe it is important to share our current findings. Importantly, the majority (nearly
M AN U
70%) of patients due for follow up did have surveillance performed. As a result of the retrospective nature of this study, assessment for local recurrence was not standardized and did not require scar visualization or biopsy. In the absence of scar visualization or biopsy, we used the colonoscopy report and images to determine whether recurrence
TE D
was grossly seen near the prior EMR site. A prospective study could standardize the method for recurrence detection, for example by placing a tattoo at every EMR site. Furthermore, we only examined lesions amenable to resection by inject-and-cut EMR
EP
(though no SSP underwent surgical resection during the study time period), and we did not compare recurrence rates of SSPs to nonpolypoid adenomas removed by EMR.
AC C
In conclusion, inject-and-cut EMR is a safe and efficacious technique for the
resection of SSPs. Our results are based on a high surveillance colonoscopy rate— nearly 70% of patients due for follow up in the serrated lesion cohort after EMR. Local recurrence occurs the minority of the time (3.6%), is small in size and can be treated endoscopically.
ACCEPTED MANUSCRIPT
Figure Legend Figure 1: Endoscopic Mucosal Resection
RI PT
A: Identify the lesion initially with white light, then enhance visualization with narrow band imaging followed by 0.2% diluted indigo carmine sprayed onto the area of interest by a syringe through the accessory channel.
SC
B: Inject dilute indigo carmine into the submucosal space with a 25-gauge sclerotherapy needle to create a submucosal bleb.
M AN U
C: Suction air to collapse the distended colon and place the stiff electrosurgical snare around the lifted area of interest to complete resection. D: Re-assess the post-resection margin immediately for residual and use APC
AC C
EP
TE D
for residual or endoscopic clips to close mucosal defects as appropriate.
ACCEPTED MANUSCRIPT
References
6. 7.
8.
9.
10.
11.
12.
13.
14.
15.
RI PT
SC
5.
M AN U
4.
TE D
3.
EP
2.
Pohl H, Srivastava A, Bensen SP, et al. Incomplete polyp resection during colonoscopy-results of the complete adenoma resection (CARE) study. Gastroenterology 2013;144:74-80 e1. Mannath J, Subramanian V, Singh R, et al. Polyp recurrence after endoscopic mucosal resection of sessile and flat colonic adenomas. Dig Dis Sci 2011;56:2389-95. Tadepalli US, Feihel D, Miller KM, et al. A morphologic analysis of sessile serrated polyps observed during routine colonoscopy (with video). Gastrointest Endosc 2011;74:1360-8. Anderloni A, Jovani M, Hassan C, et al. Advances, problems, and complications of polypectomy. Clin Exp Gastroenterol 2014;7:285-96. Bouwens MWE, Riedl RG, Bosman FT, et al. Large proximal serrated polyps: natural history and colorectal cancer risk in a retrospective series. Journal of clinical gastroenterology 2013;47:734-735. Rex KD, Vemulapalli KC, Rex DK. Recurrence rates after EMR of large sessile serrated polyps. Gastrointestinal endoscopy 2015. Moss A, Williams SJ, Hourigan LF, et al. Long-term adenoma recurrence following wide-field endoscopic mucosal resection (WF-EMR) for advanced colonic mucosal neoplasia is infrequent: results and risk factors in 1000 cases from the Australian Colonic EMR (ACE) study. Gut 2014:gutjnl-2013-305516. Kaltenbach T, Friedland S, Maheshwari A, et al. Short- and long-term outcomes of standardized EMR of nonpolypoid (flat and depressed) colorectal lesions > or = 1 cm (with video). Gastrointest Endosc 2007;65:857-65. Buchner AM, Guarner-Argente C, Ginsberg GG. Outcomes of EMR of defiant colorectal lesions directed to an endoscopy referral center. Gastrointest Endosc 2012;76:255-63. Arebi N, Swain D, Suzuki N, et al. Endoscopic mucosal resection of 161 cases of large sessile or flat colorectal polyps. Scandinavian journal of gastroenterology 2007;42:859-866. Ferrara F, Luigiano C, Ghersi S, et al. Efficacy, safety and outcomes of'inject and cut'endoscopic mucosal resection for large sessile and flat colorectal polyps. Digestion 2009;82:213-220. Moss A, Bourke MJ, Williams SJ, et al. Endoscopic mucosal resection outcomes and prediction of submucosal cancer from advanced colonic mucosal neoplasia. Gastroenterology 2011;140:1909-1918. Mahadeva S, Rembacken BJ. Standard "inject and cut" endoscopic mucosal resection technique is practical and effective in the management of superficial colorectal neoplasms. Surg Endosc 2009;23:417-22. Rex DK, Ahnen DJ, Baron JA, et al. Serrated lesions of the colorectum: review and recommendations from an expert panel. Am J Gastroenterol 2012;107:1315-29; quiz 1314, 1330. Serrano M, de Ferro SM, Fidalgo P, et al. Endoscopic mucosal resection of superficial colorectal neoplasms: review of 140 procedures. Acta Médica Portuguesa 2012;25:288-296.
AC C
1.
ACCEPTED MANUSCRIPT
22.
RI PT
21.
SC
20.
M AN U
19.
TE D
18.
EP
17.
Santos CEOd, Malaman D, Pereira-Lima JC. Endoscopic mucosal resection in colorectal lesion: a safe and effective procedure even in lesions larger than 2 cm and in carcinomas. Arquivos de gastroenterologia 2011;48:242-247. Burgess NG, Pellise M, Nanda KS, et al. Clinical and endoscopic predictors of cytological dysplasia or cancer in a prospective multicentre study of large sessile serrated adenomas/polyps. Gut 2015. Lash RH, Genta RM, Schuler CM. Sessile serrated adenomas: prevalence of dysplasia and carcinoma in 2139 patients. J Clin Pathol 2010;63:681-6. Snover DC. Update on the serrated pathway to colorectal carcinoma. Human pathology 2011;42:1-10. Hiraoka S, Kato J, Fujiki S, et al. The presence of large serrated polyps increases risk for colorectal cancer. Gastroenterology 2010;139:1503-10, 1510 e1-3. Lu FI, van Niekerk de W, Owen D, et al. Longitudinal outcome study of sessile serrated adenomas of the colorectum: an increased risk for subsequent right-sided colorectal carcinoma. Am J Surg Pathol 2010;34:927-34. Lieberman DA, Rex DK, Winawer SJ, et al. Guidelines for colonoscopy surveillance after screening and polypectomy: a consensus update by the US Multi-Society Task Force on Colorectal Cancer. Gastroenterology 2012;143:844-57.
AC C
16.
ACCEPTED MANUSCRIPT
61.0 ± 8.4 (35-83), 61
Male (n, %)
146 (73.4)
Ethnicity/Race (n, %) Non-Hispanic White
174 (87.4)
Non-Hispanic Black
4 (2.0)
Hispanic
11 (5.5)
Other
10 (5.0) 26 (13.1)
M AN U
Family hx of CRC (n, %)
SC
Age, years (Mean ± SD, range, median)
RI PT
Table 1. Baseline Patient Characteristics, N=199
Hx of CRC (n,%)
3 (1.5)
Indication (n, %)
90 (45.2)
Surveillance
88 (44.2)
Diagnostic
21 (10.6)
TE D
Screening
Total number of polyps at index colonoscopy (Mean ± SD, range, median)
EP
Number of Serrated Lesions*
AC C
Number of Synchronous Neoplasia** Number of Synchronous Advanced Neoplasia**
1.9 ± 1.6 (1-12), 1 2.0 ± 2.5 (0-15), 1 0.5 ± 0.9 (0-4), 0
Other includes 5 Asian, 2 Pacific Islander and 3 Unknown; *Serrated lesions include sessile serrated adenomas/polyps, hyperplastic polyps and traditional serrated adenomas **Neoplasia includes tubular adenomas, villous adenomas and cancer. These were subcategorized as advanced neoplasia if the tubular adenoma was ≥10 mm or had villous, highgrade dysplastic or cancerous features. SSP with or without dysplasia were not included in the category of neoplasia.
ACCEPTED MANUSCRIPT
Table 2. Baseline Characteristics of the Sessile Serrated Polyp (SSP), N=251 Size, mm
Shape, n(%)
38 (15.1)
Superficially Elevated (0-IIa)(flat)
212 (84.5)
Depressed (0-IIc)
1 (0.4)
M AN U
SC
Sessile (0-Is)
Location, n(%)
Ileo-cecal valve
10 (4.0)
Cecum
40 (15.9)
Ascending
128 (51.0)
Splenic flexure
55 (21.9) 2 (0.8)
EP
Pathology, n(%)
16 (6.4)
TE D
Hepatic flexure Transverse
AC C
Sessile Serrated Polyp with Dysplasia Sessile Serrated Polyp without Dysplasia
RI PT
15.9 ± 5.3 (10-50), 15
(Mean ± SD, range, median)
3 (1.2)
248 (98.8)
ACCEPTED MANUSCRIPT
Table 3. Recurrence Characteristics Index polyp pathology
Index polyp location
Index polyp shape
En bloc vs. piecemeal
APC
1
20
SSP
Ascending
Flat
Piecemeal
No
Time to detect recurrence (months) 7
Method of recurrence detection
Recurrence size (mm)
Method of recurrence removal
Recurrence pathology
4
EMR
Lost specimen
37 9
Indigo carmine, NBI and biopsy Biopsy Biopsy
2 3
10 20
SSP SSP
Ascending Ascending
Flat Flat
En bloc En bloc
No Yes
15 3
4
20
SSP
Flat
En bloc
5
20
SSP with high grade dysplasia
Ileo-cecal valve Ascending
Yes
32**
Biopsy
Flat
Piecemeal
Yes
4
Indigo carmine, NBI and biopsy
RI PT
Index polyp size
SSP SSP
25
EMR Cold forceps EMR
4
EMR
Tubular adenoma
SC
Patient
nd
AC C
EP
TE D
M AN U
**Recurrence was detected on the 2 surveillance colonoscopy. This patient was initially resected en bloc at the ileo-cecal valve, returned in 3 months and no recurrence was found. The patient came back in almost 3 years at which point recurrence was detected.
SSP
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S1542-3565(15)01412-3 10.1016/j.cgh.2015.10.013 YJCGH 54512
To appear in: Clinical Gastroenterology and Hepatology Accepted Date: 4 October 2015 Please cite this article as: Rao AK, Soetikno R, Raju GS, Lum P, Rouse RV, Sato T, Titzer-Schwarzl D, Aisenberg J, Kaltenbach T, Large Sessile Serrated Polyps Can Be Safely and Effectively Removed by Endoscopic Mucosal Resection, Clinical Gastroenterology and Hepatology (2015), doi: 10.1016/ j.cgh.2015.10.013. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Large Sessile Serrated Polyps Can Be Safely and Effectively Removed by Endoscopic Mucosal Resection
RI PT
Aarti K. Rao MD1, Roy Soetikno MD, MS1, Gottumukkala S. Raju MD2, Phillip Lum2, Robert V. Rouse MD1, Tohru Sato MD1, Diane Titzer-Schwarzl1, James Aisenberg MD3, and Tonya Kaltenbach MD, MS1
SC
1. Veterans Affairs Palo Alto Health Care System and Stanford University School of Medicine, Palo Alto, CA; 2. The University of Texas MD Anderson Cancer Center, Houston, TX; 3. Icahn School of Medicine at Mount Sinai, New York, NY
Grant support: none
M AN U
Abbreviations: SSP (sessile serrated adenoma/polyp), EMR (endoscopic mucosal resection), CRC (colorectal cancer), APC (argon plasma coagulation), HP (hyperplastic polyp), TSA (traditional serrated adenoma)
TE D
Corresponding Author: Tonya Kaltenbach MD MS Clinical Assistant Professor of Medicine (Affiliate) Stanford University School of Medicine Veterans Affairs Palo Alto 3801 Miranda Avenue, GI 111, Palo Alto, CA 94304 650-814-0629 [email protected]
AC C
EP
Conflict of Interest: Aarti K. Rao: no conflict of interest to declare Roy Soetikno has received non-salaried research support and is a consultant for Olympus America, Inc. Gottumukkala S. Raju: no conflict of interest to declare Phillip Lum: no conflict of interest to declare Robert V. Rouse: no conflict of interest to declare Tohru Sato: no conflict of interest to declare Diane Titzer-Schwarzl: no conflict of interest to declare James Aisenberg: no conflict of interest to declare Tonya Kaltenbach has received non-salaried research support and is a consultant for Olympus America, Inc. Author Contributions: Aarti Rao • Acquisition of data • Analysis and interpretation of data • Drafting of the manuscript
ACCEPTED MANUSCRIPT
•
Critical revision of the manuscript for intellectual content
RI PT
Roy Soetikno • Study concept and design • Acquisition of data • Analysis and interpretation of data • Drafting of the manuscript • Critical revision of the manuscript for intellectual content • Study supervision
Phillip Lum • Acquisition of data
M AN U
SC
Gottumukkala S. Raju • Study concept and design • Acquisition of data • Critical revision of the manuscript for intellectual content • Study supervision
Robert V. Rouse • Acquisition of data • Critical revision of the manuscript for intellectual content
TE D
Tohru Sato • Acquisition of data • Critical revision of the manuscript for intellectual content
EP
Diane Titzer-Schwarzl • Acquisition of data
AC C
James Aisenberg • Study concept and design • Critical revision of the manuscript for intellectual content Tonya Kaltenbach • Study concept and design • Acquisition of data • Analysis and interpretation of data • Statistical analysis • Drafting of the manuscript • Critical revision of the manuscript for intellectual content • Study supervision
ACCEPTED MANUSCRIPT
Abstract
Background & Aims
RI PT
As many as 50% of large sessile serrated adenomas/polyps (SSPs) are removed
incompletely, which is significant because SSPs have been implicated in development of interval cancers. It is unclear if endoscopic mucosal resection (EMR) is an optimal
SC
method for removal of SSPs. We assessed the efficacy and safety of removal of SSPs
M AN U
≥10 mm using a standardized inject-and-cut EMR technique.
Methods
We performed a retrospective analysis of colonoscopy data, collected over 7 years (2007–2013) at 2 centers, from 199 patients with proximal colon SSPs ≥10 mm (251
TE D
polyps) removed by EMR by 4 endoscopists. The primary outcome measure was local recurrence. The secondary outcome measure was safety.
EP
Results
At the index colonoscopy, patients had a median of 1 serrated lesion (range, 1–12) and
AC C
1 non-serrated neoplastic lesion (range, 0–15). The mean SSP size was 15.9±5.3 mm; most were superficially elevated (84.5%) and located in the ascending colon (51%), and 3 SSPs (1.2%) had dysplasia. Surveillance colonoscopies were performed on 138 patients (69.3%) over a mean follow up time of 25.5±17.4 months. Of these patients, 5 had local recurrences (3.6%; 95% confidence interval, 0.5–6.7%), detected after 17.8±15.4 months, with a median size of 4 mm. No patients developed post-procedural
ACCEPTED MANUSCRIPT
bleeding, perforation, or advanced colon cancer or had a death related to the index colorectal lesion during the study period.
RI PT
Conclusions
Inject-and-cut EMR is a safe and effective technique for the resection of SSPs. Less than 5% of patients have a local recurrence, which is usually small and can be treated
SC
endoscopically.
AC C
EP
TE D
M AN U
KEY WORDS: endoscopy; tumor; early detection; malignancy
ACCEPTED MANUSCRIPT
Background The endoscopic features of sessile serrated adenomas/polyps (SSPs) predominantly flat in shape and with indistinct borders1-3 - make them difficult to detect
RI PT
and remove. SSPs are more likely to be incompletely resected compared to
conventional adenomas (31% vs. 7.2%) and larger lesions have even higher rates (47.6%) of incomplete removal.1 The development of interval cancers has been
SC
attributed to incompletely resected lesions.4 A recent study of large proximal serrated polyps showed that 3 of 18 patients with such lesions biopsied but not completely
M AN U
removed developed interval colorectal cancer (CRC) at the same location.5 At present, the optimal method for removal of SSPs is unknown. The inject-andcut EMR technique, which utilizes saline injection into the submucosa to lift the lesion of interest and subsequent stiff electrocautery snare to resect the polyp, may improve the
TE D
likelihood of complete resection. The outcomes of EMR for serrated lesions, however, are largely unavailable.6, 7 Prior studies of EMR have primarily focused on its safety and efficacy for the resection of conventional adenomas.8-13 We examined the safety and
AC C
centers.
EP
efficacy of EMR for proximal SSPs ≥10mm performed by multiple endoscopists at 2
Methods
Patients and study design We systematically reviewed outcomes data of a serrated lesion cohort who underwent colonoscopy by 1 of 4 endoscopists with significant experience in EMR at 2
ACCEPTED MANUSCRIPT
centers (Veterans Affairs [VA] Palo Alto Health Care System and University of Texas MD Anderson Cancer Center) over a 7-year period (2007-2013). We identified the serrated lesion cohort using a Filemaker Pro pathology database, and ascertained data
RI PT
using an endoscopic database and computerized medical record. We included serrated lesions with specific criteria: (1) pathologic diagnosis of SSP as defined by consensusmodified World Health Organization criteria14, (2) size ≥10mm, (3) location proximal to
SC
and including the splenic flexure, (4) endoscopic resection by the inject-and-cut EMR technique, and (5) due for surveillance colonoscopy based on multi-society
M AN U
recommendations (6 months for piecemeal resection and 3 years for en bloc resection). Serrated lesions with pathologic diagnosis of hyperplastic polyp (HP) and traditional serrated adenoma (TSA) were excluded. We analyzed the recurrence rate of SSPs in patients who had at least one surveillance colonoscopy. The institutional review boards
Procedure
TE D
at both sites approved the study.
EP
Patients completed standard bowel preparation prior to the procedure and received moderate sedation with midazolam and fentanyl or meperidine. All patients
AC C
were prescribed split bowel preparation the evening prior to and the morning of the colonoscopy. VA Palo Alto patients received polyethylene glycol-based electrolyte solution and/or magnesium citrate and MD Anderson patients received polyethylene glycol-based solution and bisacodyl. The endoscopists performed colonoscopy using an adult or pediatric high definition colonoscope equipped with an auxiliary water jet (Olympus PCFH180AL, CFH180AL, CFH190AL) and carbon dioxide insufflation. Both
ACCEPTED MANUSCRIPT
sites used a standardized endoscopic resection approach that includes lesion assessment, inject-and-cut EMR techniques, immediate re-assessment and treatment of residual lesion (Figure 1).
RI PT
Specifically, we visualized the index SSP lesion using white light and then
enhanced visualization with narrow band imaging, followed by selective spray of 0.2% diluted indigo carmine onto the area of interest for suspected nonpolypoid serrated
SC
lesions, by a syringe through the accessory channel. Injection of a mixture of dilute indigo carmine and saline into the submucosa with a 25-gauge needle was used to
M AN U
create a submucosal bleb. We used a variety of 25-gauge needles (LDVI-25-240, Cook Medical Inc., Bloomington, Indiana; 00711811, US Endoscopy, Mentor, Ohio; M00518161, Boston Scientific, Marlborough, Massachusetts; Injector Force Max Injection Needle, Olympus America, Center Valley, Pennsylvania). Indigo carmine in the
TE D
injection solution also aided the rapid visual contrast of resection depth. The distended colon was slightly collapsed with suctioning of the lumen and the stiff electrosurgical snare (10-mm oval snare [SD-210–10] and/or 20-mm spiral snare [SD-230–20],
EP
Olympus America, Center Valley, Pennsylvania) was placed around the lifted area of interest to complete snare resection. We applied cautery for snare resection using the
AC C
Endostat II Microvasive electrosurgical unit (Boston Scientific, Marlborough, Massachusetts) at setting of 35 watts or the ERBE electrosurgical generator (VIO 330D; ERBE USA, Marietta, Georgia) using the fractionated cutting mode EndoCutQ (effect 3, cut duration 1, and cut interval 4 at VA Palo Alto; effect 3, cut duration 1, and cut interval 3 at MD Anderson Cancer Center) for snare resection. In cases where en bloc resection was not possible, we resected the lesion piecemeal. After resection, the site
ACCEPTED MANUSCRIPT
was closely examined and multiple photographs were taken to confirm eradication. Argon plasma coagulation (APC) at 60W and 1L/min flow on a continuous setting using the ERBE electrosurgical generator (VIO 330D; ERBE USA, Marietta, Georgia) was
RI PT
applied to the periphery and bridges in piecemeal resection, and to exposed superficial vessels. Prophylactic endoscopic clips (Resolution Clip, Boston Scientific, Marlborough, Massachusetts; Instinct Clip, Cook Medical Inc., Bloomington, Indiana; QuikClip 2,
SC
Olympus America, Center Valley, Pennsylvania) were selectively placed to close
Histopathology Analysis
M AN U
mucosal defects.
A blinded pathologist reviewed all identified serrated lesions (SSP, HP and TSA) that met inclusion criteria of size, location and resection technique. A diagnosis of SSP
TE D
required at least one abnormal crypt, defined as crypt dilation with a flattened/horizontal base and serration variably present throughout the crypt length.14 Synchronous lesions classified as non-serrated neoplasia included tubular adenomas, villous adenomas or
EP
cancer. These were sub-categorized as non-serrated advanced neoplasia if the tubular
AC C
adenoma was ≥10 mm or had villous, high-grade dysplastic or cancerous features.
Study Outcomes
We evaluated the efficacy of the inject-and-cut EMR technique for complete
resection of large SSPs by measuring local recurrence rate at surveillance colonoscopy as the primary endpoint. We defined local recurrence as lesion residual or regrowth at the index EMR location. Local recurrence was assessed with visualization of the post-
ACCEPTED MANUSCRIPT
EMR scar site with elective use of image enhancement (narrow band imaging or sprayed or injected indigo carmine) or biopsy. In cases where the prior EMR scar was not explicitly visible, we reviewed the index colonoscopy report and images to determine
RI PT
whether recurrence was grossly seen in the vicinity of the prior EMR site. We also
evaluated the safety of the EMR technique by secondarily measuring post-procedure bleeding, perforation and need for surgery. We defined index colonoscopy as the initial
SC
colonoscopy at which the serrated lesion was identified and resected. Subsequent
colonoscopies were categorized as surveillance. Patients were recalled for surveillance
M AN U
by letter and/or phone call. Patients at VA Palo Alto received a follow up call after the procedure and any complications or hospitalizations within 30 days after the procedure were recorded on a standardized form. At MD Anderson, patients were followed up with
Statistical Analysis
TE D
daily emails for 5 days and by telephone at 14 days.
We performed descriptive statistical analysis (STATA/SE 13.1 for Macintosh,
EP
Stata Corporation) for patients and polyps. We primarily analyzed local recurrence rate of SSPs. We report percentages for categorical variables and mean, standard deviation,
AC C
range and median for continuous variables.
Results
Patients
We identified 199 patients with 251 proximal SSPs that were ≥10mm and removed by inject-and-cut EMR. The characteristics of the patients are shown in Table
ACCEPTED MANUSCRIPT
1. Colonoscopies were performed in 129 patients by 3 endoscopists at VA Palo Alto and 70 patients by 1 endoscopist at MD Anderson. The majority were men (73.4%) and had a mean age of 61.0 ± 8.4 years. Ninety patients in the cohort (45%) underwent the
RI PT
index colonoscopy for average risk colorectal cancer screening. At the index examination, the patients had a median of 1 (range 1-7) proximal SSP ≥10mm. In addition, they had a median of 1 (range 0-15) non-serrated synchronous lesion.
SC
The characteristics of SSPs are shown in Table 2. The mean size SSP was 15.9 ± 5.3 mm (range 10-50mm, median 15mm). The majority was nonpolypoid superficially
M AN U
elevated (84.5%) and located in the ascending colon (51.0%). Three SSPs (1.2%, 95%CI:0.25-3.5%) had dysplastic histologic foci. These SSPs with dysplasia were 10, 12 and 20 mm in size.
TE D
Outcomes
At an average follow-up of over 2 years, SSPs removed with EMR had a low local recurrence rate, 3.6%, 95% CI = 0.5-6.7%. Overall, the median size of recurrence
EP
was 4mm (mean size 10.2 ± 9.6, range 3-25). There was no recurrence with high-grade dysplasia or cancer. Of the 5 SSPs that developed recurrence, the median size of the
AC C
index lesion was 20mm, only one had high-grade dysplasia and all were flat. Table 3 details the characteristics of each local recurrence. The majority of the cohort (138 of 199, 69.3% patients and 176 of 251, 70.1%
lesions) had surveillance colonoscopy at a mean of 25.5 months (range 2-74). The reasons for not undergoing surveillance were varied, and included death from unrelated causes, co-morbidities, advanced age and non-adherence to recommendations (Figure
ACCEPTED MANUSCRIPT
2). At surveillance, we found a mean of 1.1 ± 2.0 (range 0-13, median 0) metachronous non-serrated neoplastic lesions and 0.4 ± 1.0 (range 0-5, median 0) metachronous serrated lesions. The five patients with local recurrence had it detected at a mean
RI PT
surveillance interval of 17.8 months (range 4-37months). We successfully treated all of the local recurrences endoscopically – the majority (4/5) at the time of the first surveillance colonoscopy. One SSP showed diminutive recurrence at multiple
SC
surveillance colonoscopies until ultimately no recurrence was detected at the sixth colonoscopy exam. All lesions were completely eradicated based on macroscopic
M AN U
inspection of the polyp site perimeter post-EMR.
We resected the majority of lesions (86.5%, 217/251) en bloc. The mean size of lesions resected en bloc was 14.9 ± 4.0mm (range 10-30 mm). The mean size of lesions resected piecemeal was 22.2 ± 8.6mm (range 10-50 mm). Of the 176 polyps
TE D
that received surveillance, we resected 142 lesions en bloc (size 15.7 ± 4.2 mm [range 10-30 mm]) and 34 lesions piecemeal (size 21.1 ± 7.7 mm [range 10-50 mm]). Three of the 5 recurrences (60%) originated from index lesions that had been resected en bloc.
EP
Recurrence rate for lesions resected en bloc was 2.1% (3/142) and 5.9% (2/34) for piecemeal. While lesions resected piecemeal were more likely to recur than those
AC C
resected en bloc, the difference was not statistically significant (OR 2.9, p=0.23).
Adverse Events
No patients developed post-procedural bleeding, perforation, advanced colon
cancer or died related to the index colorectal lesion during the study period. We applied
ACCEPTED MANUSCRIPT
clips in 84.5% (n=212) to close the lesion defect. Of the 56 SSPs ≥20mm in the cecum or ascending colon, we clipped 48 (85.7%) lesions.
RI PT
Discussion
The findings of our two-site multi-endoscopist study show high rates of safe and complete resection of SSPs ≥10mm using the inject-and-cut EMR technique with a low
SC
(3.6%) recurrence rate and no complications. All five local recurrences were small (median 4mm) and four lesions were endoscopically cured after one surveillance
M AN U
colonoscopy. Our outcomes expand upon the previously published single-operator findings that EMR of large SSPs achieves recurrence rates similar to EMR of conventional adenomas.6 Our study further emphasizes the importance of EMR in the removal of SSPs and in addition illustrates the low prevalence of dysplasia in SSPs.
TE D
Our study provides important data on the outcomes of EMR in the resection of SSPs. In this study, we found 85% of index SSPs to be non-polypoid, and, thus, the need to use EMR to resect SSPs. A recent single endoscopist study using the inject-
EP
and-cut EMR technique showed low recurrence rates (8.7%) in 46 SSPs, similar to the low recurrence rate in our study.6 This Rex et al study examined SSPs ≥20 mm. SSPs
AC C
between 10 and 20 mm are much more frequent than SSPs ≥20mm, and a lesion size cut-off of 10mm is used in SSP surveillance guidelines. We, thus, focused on SSPs ≥10mm.
We found EMR to be safe and effective to remove SSPs in this first and largest
long-term US multi-site study. Our findings suggest that EMR for complete removal of SSPs is highly efficacious when performed by endoscopists who are facile with EMR.
ACCEPTED MANUSCRIPT
Our study is also in agreement with our prior paper that found the EMR technique to be effective in removing non-polypoid lesions8 and others, which reported recurrence to be small in size and occur infrequently.2, 8, 9, 13, 15, 16 In contrast, prior data showed
RI PT
polypectomy to be ineffective in achieving complete resection of SSPs. Pohl and
colleagues reported the inadequacies of hot snare resection of SSPs, with high rates of incomplete resections in SSPs compared to other neoplastic polyps (31.0% vs. 7.2%;
SC
p=0.001). Nearly half of all large (10-20mm) SSPs were incompletely removed.1 We suggest that EMR may mitigate the variability in incomplete resection of SSPs using
M AN U
standard polypectomy.
Our data support the current multi-society recommendations for follow up of SSA/P. Current surveillance guidelines suggest follow up in 3 years for SSPs ≥10mm, 1-3 years if there are 2 or more SSPs ≥10mm or dysplasia and 3-6 months for index
TE D
lesions resected piecemeal.14 The recurrence rate of SSPs after EMR is low (3.6%) and the majority (4 of 5) of recurrences were detected in larger SSPs ≥20mm. The mean time to detection of recurrence was approximately 1.5 years (range 0.25-3) and we
EP
detected 4 of 5 recurrences during the first surveillance colonoscopy at 4, 7, 9 and 37 months. One patient whose initial lesion was resected en bloc at the ileo-cecal valve did
AC C
not have recurrence identified until the second follow up examination. While lesions resected piecemeal were more likely to recur than those resected
en bloc, the difference was not statistically significant (OR 2.9, p=0.23). We found 60% (3 of 5) of our recurrences originated from en bloc-resected SSPs. This is in contrast to prior studies that have shown piecemeal EMR results in higher recurrence rates than en bloc EMR.2, 8 Although these numbers are too small to draw any conclusions,
ACCEPTED MANUSCRIPT
inconspicuous serrated pathology may be easily overlooked in this group compared to traditional adenomas. A subset of lesions resected en bloc may have other features that characterize them as high risk for earlier surveillance, such as size, but this requires
RI PT
further study.
We found a low rate of dysplasia in our population (1.2%). In comparison, prior endoscopic and pathologic studies described rates between 8.7% and 32.4%.6, 17, 18
SC
Burgess and colleagues found a 32.4% dysplasia rate in large SSPs ≥20mm. Their high dysplasia rates may reflect referral bias in a tertiary care center or an association
M AN U
between size ≥20mm and the presence of dysplasia. It might also be due to the smaller size of lesions in our cohort (median 15 mm) that had not yet grown enough to develop dysplasia.
Our findings show promise for the implementation of the EMR technique for
TE D
SSPs ≥10mm, which are associated with the development of CRC. Sessile serrated adenomas/polyps are important precursors of CRC via distinct pathways involving microsatellite instability and hypermethylation events. Prior studies have shown that
EP
compared to conventional adenomas, serrated lesions have a higher association with CRC and higher rates of incomplete resection, thought to be due to their flat shape with
AC C
indistinct borders.1, 20, 21 In fact, large serrated polyps ≥10mm have been shown to be the greatest risk factor in the development of CRC compared to age, presence of adenoma ≥10mm or ≥4 adenomas. This is especially true for proximal CRC.20 Guidelines suggest lesions ≥10mm are the highest risk for significant pathology.22 The findings from our phase I/II study are favorable for the safety and efficacy of EMR for large serrated lesions—with no complications and high endoscopic removal rates.
ACCEPTED MANUSCRIPT
Although we report the outcomes of multiple endoscopists who practice at a tertiary care medical center and an academic veterans hospital to improve generalizability of the technique to physicians and patients, our study has limitations.
RI PT
While detection of SSPs has improved in recent years due to increasing endoscopic and pathologic recognition, our recent study period (2007-2013) resulted in the
exclusion of a over one quarter of patients (71/270) who were not yet due for follow up
SC
at the time of data analysis. However, since data on resection of SSPs is sparse, we believe it is important to share our current findings. Importantly, the majority (nearly
M AN U
70%) of patients due for follow up did have surveillance performed. As a result of the retrospective nature of this study, assessment for local recurrence was not standardized and did not require scar visualization or biopsy. In the absence of scar visualization or biopsy, we used the colonoscopy report and images to determine whether recurrence
TE D
was grossly seen near the prior EMR site. A prospective study could standardize the method for recurrence detection, for example by placing a tattoo at every EMR site. Furthermore, we only examined lesions amenable to resection by inject-and-cut EMR
EP
(though no SSP underwent surgical resection during the study time period), and we did not compare recurrence rates of SSPs to nonpolypoid adenomas removed by EMR.
AC C
In conclusion, inject-and-cut EMR is a safe and efficacious technique for the
resection of SSPs. Our results are based on a high surveillance colonoscopy rate— nearly 70% of patients due for follow up in the serrated lesion cohort after EMR. Local recurrence occurs the minority of the time (3.6%), is small in size and can be treated endoscopically.
ACCEPTED MANUSCRIPT
Figure Legend Figure 1: Endoscopic Mucosal Resection
RI PT
A: Identify the lesion initially with white light, then enhance visualization with narrow band imaging followed by 0.2% diluted indigo carmine sprayed onto the area of interest by a syringe through the accessory channel.
SC
B: Inject dilute indigo carmine into the submucosal space with a 25-gauge sclerotherapy needle to create a submucosal bleb.
M AN U
C: Suction air to collapse the distended colon and place the stiff electrosurgical snare around the lifted area of interest to complete resection. D: Re-assess the post-resection margin immediately for residual and use APC
AC C
EP
TE D
for residual or endoscopic clips to close mucosal defects as appropriate.
ACCEPTED MANUSCRIPT
References
6. 7.
8.
9.
10.
11.
12.
13.
14.
15.
RI PT
SC
5.
M AN U
4.
TE D
3.
EP
2.
Pohl H, Srivastava A, Bensen SP, et al. Incomplete polyp resection during colonoscopy-results of the complete adenoma resection (CARE) study. Gastroenterology 2013;144:74-80 e1. Mannath J, Subramanian V, Singh R, et al. Polyp recurrence after endoscopic mucosal resection of sessile and flat colonic adenomas. Dig Dis Sci 2011;56:2389-95. Tadepalli US, Feihel D, Miller KM, et al. A morphologic analysis of sessile serrated polyps observed during routine colonoscopy (with video). Gastrointest Endosc 2011;74:1360-8. Anderloni A, Jovani M, Hassan C, et al. Advances, problems, and complications of polypectomy. Clin Exp Gastroenterol 2014;7:285-96. Bouwens MWE, Riedl RG, Bosman FT, et al. Large proximal serrated polyps: natural history and colorectal cancer risk in a retrospective series. Journal of clinical gastroenterology 2013;47:734-735. Rex KD, Vemulapalli KC, Rex DK. Recurrence rates after EMR of large sessile serrated polyps. Gastrointestinal endoscopy 2015. Moss A, Williams SJ, Hourigan LF, et al. Long-term adenoma recurrence following wide-field endoscopic mucosal resection (WF-EMR) for advanced colonic mucosal neoplasia is infrequent: results and risk factors in 1000 cases from the Australian Colonic EMR (ACE) study. Gut 2014:gutjnl-2013-305516. Kaltenbach T, Friedland S, Maheshwari A, et al. Short- and long-term outcomes of standardized EMR of nonpolypoid (flat and depressed) colorectal lesions > or = 1 cm (with video). Gastrointest Endosc 2007;65:857-65. Buchner AM, Guarner-Argente C, Ginsberg GG. Outcomes of EMR of defiant colorectal lesions directed to an endoscopy referral center. Gastrointest Endosc 2012;76:255-63. Arebi N, Swain D, Suzuki N, et al. Endoscopic mucosal resection of 161 cases of large sessile or flat colorectal polyps. Scandinavian journal of gastroenterology 2007;42:859-866. Ferrara F, Luigiano C, Ghersi S, et al. Efficacy, safety and outcomes of'inject and cut'endoscopic mucosal resection for large sessile and flat colorectal polyps. Digestion 2009;82:213-220. Moss A, Bourke MJ, Williams SJ, et al. Endoscopic mucosal resection outcomes and prediction of submucosal cancer from advanced colonic mucosal neoplasia. Gastroenterology 2011;140:1909-1918. Mahadeva S, Rembacken BJ. Standard "inject and cut" endoscopic mucosal resection technique is practical and effective in the management of superficial colorectal neoplasms. Surg Endosc 2009;23:417-22. Rex DK, Ahnen DJ, Baron JA, et al. Serrated lesions of the colorectum: review and recommendations from an expert panel. Am J Gastroenterol 2012;107:1315-29; quiz 1314, 1330. Serrano M, de Ferro SM, Fidalgo P, et al. Endoscopic mucosal resection of superficial colorectal neoplasms: review of 140 procedures. Acta Médica Portuguesa 2012;25:288-296.
AC C
1.
ACCEPTED MANUSCRIPT
22.
RI PT
21.
SC
20.
M AN U
19.
TE D
18.
EP
17.
Santos CEOd, Malaman D, Pereira-Lima JC. Endoscopic mucosal resection in colorectal lesion: a safe and effective procedure even in lesions larger than 2 cm and in carcinomas. Arquivos de gastroenterologia 2011;48:242-247. Burgess NG, Pellise M, Nanda KS, et al. Clinical and endoscopic predictors of cytological dysplasia or cancer in a prospective multicentre study of large sessile serrated adenomas/polyps. Gut 2015. Lash RH, Genta RM, Schuler CM. Sessile serrated adenomas: prevalence of dysplasia and carcinoma in 2139 patients. J Clin Pathol 2010;63:681-6. Snover DC. Update on the serrated pathway to colorectal carcinoma. Human pathology 2011;42:1-10. Hiraoka S, Kato J, Fujiki S, et al. The presence of large serrated polyps increases risk for colorectal cancer. Gastroenterology 2010;139:1503-10, 1510 e1-3. Lu FI, van Niekerk de W, Owen D, et al. Longitudinal outcome study of sessile serrated adenomas of the colorectum: an increased risk for subsequent right-sided colorectal carcinoma. Am J Surg Pathol 2010;34:927-34. Lieberman DA, Rex DK, Winawer SJ, et al. Guidelines for colonoscopy surveillance after screening and polypectomy: a consensus update by the US Multi-Society Task Force on Colorectal Cancer. Gastroenterology 2012;143:844-57.
AC C
16.
ACCEPTED MANUSCRIPT
61.0 ± 8.4 (35-83), 61
Male (n, %)
146 (73.4)
Ethnicity/Race (n, %) Non-Hispanic White
174 (87.4)
Non-Hispanic Black
4 (2.0)
Hispanic
11 (5.5)
Other
10 (5.0) 26 (13.1)
M AN U
Family hx of CRC (n, %)
SC
Age, years (Mean ± SD, range, median)
RI PT
Table 1. Baseline Patient Characteristics, N=199
Hx of CRC (n,%)
3 (1.5)
Indication (n, %)
90 (45.2)
Surveillance
88 (44.2)
Diagnostic
21 (10.6)
TE D
Screening
Total number of polyps at index colonoscopy (Mean ± SD, range, median)
EP
Number of Serrated Lesions*
AC C
Number of Synchronous Neoplasia** Number of Synchronous Advanced Neoplasia**
1.9 ± 1.6 (1-12), 1 2.0 ± 2.5 (0-15), 1 0.5 ± 0.9 (0-4), 0
Other includes 5 Asian, 2 Pacific Islander and 3 Unknown; *Serrated lesions include sessile serrated adenomas/polyps, hyperplastic polyps and traditional serrated adenomas **Neoplasia includes tubular adenomas, villous adenomas and cancer. These were subcategorized as advanced neoplasia if the tubular adenoma was ≥10 mm or had villous, highgrade dysplastic or cancerous features. SSP with or without dysplasia were not included in the category of neoplasia.
ACCEPTED MANUSCRIPT
Table 2. Baseline Characteristics of the Sessile Serrated Polyp (SSP), N=251 Size, mm
Shape, n(%)
38 (15.1)
Superficially Elevated (0-IIa)(flat)
212 (84.5)
Depressed (0-IIc)
1 (0.4)
M AN U
SC
Sessile (0-Is)
Location, n(%)
Ileo-cecal valve
10 (4.0)
Cecum
40 (15.9)
Ascending
128 (51.0)
Splenic flexure
55 (21.9) 2 (0.8)
EP
Pathology, n(%)
16 (6.4)
TE D
Hepatic flexure Transverse
AC C
Sessile Serrated Polyp with Dysplasia Sessile Serrated Polyp without Dysplasia
RI PT
15.9 ± 5.3 (10-50), 15
(Mean ± SD, range, median)
3 (1.2)
248 (98.8)
ACCEPTED MANUSCRIPT
Table 3. Recurrence Characteristics Index polyp pathology
Index polyp location
Index polyp shape
En bloc vs. piecemeal
APC
1
20
SSP
Ascending
Flat
Piecemeal
No
Time to detect recurrence (months) 7
Method of recurrence detection
Recurrence size (mm)
Method of recurrence removal
Recurrence pathology
4
EMR
Lost specimen
37 9
Indigo carmine, NBI and biopsy Biopsy Biopsy
2 3
10 20
SSP SSP
Ascending Ascending
Flat Flat
En bloc En bloc
No Yes
15 3
4
20
SSP
Flat
En bloc
5
20
SSP with high grade dysplasia
Ileo-cecal valve Ascending
Yes
32**
Biopsy
Flat
Piecemeal
Yes
4
Indigo carmine, NBI and biopsy
RI PT
Index polyp size
SSP SSP
25
EMR Cold forceps EMR
4
EMR
Tubular adenoma
SC
Patient
nd
AC C
EP
TE D
M AN U
**Recurrence was detected on the 2 surveillance colonoscopy. This patient was initially resected en bloc at the ileo-cecal valve, returned in 3 months and no recurrence was found. The patient came back in almost 3 years at which point recurrence was detected.
SSP
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT