Endoscopic management of large nonpedunculated colorectal polyps: selective treatment algorithms are needed

In this issue of Endoscopy, separate studies from two well-credentialed tertiary referral centers provide important information on colorectal endoscopic submucosal dissection (ESD). Together they tell us that in experienced tertiary centers, ESD for large (> 2 cm) nonpedunculated colorectal polyps (LNPCPs) is effective and safe, with recurrence being uncommon in long-term follow-up. Moreover, cure for those with submucosal invasive cancer (SMIC) is possible in a subgroup with low risk histological features (submucosal invasion < 1000 µm, no lympovascular invasion, well- or moderately well-differentiated tumor) if R0 excision has been achieved. However, despite the more than 10-year global experience with this technique in the colorectum, many questions about its true role in the treatment of LNPCPs remain unanswered.

“Clearly, based on current evidence related to recurrence risk, a universal ESD strategy is not justified.”

In a prospective single-center, two-operator study, Probst et al. from Germany, report on 330 rectal (0 – 15 cm from the anal verge) LNPCPs, of median size 40 mm, 16.7 % having nongranular morphology, referred for endoscopic resection over a 12-year period up till March 2016 [1]. Of the 330 referred patients, 302 had lesions suitable for ESD and of those 250 were benign. En bloc resection was possible in 208/250 (83 %) with an R0 resection achieved in 175/250 (70 %). The recurrence rate for benign lesions was 4.8 %. All of the recurrent lesions, except one, were resected by piecemeal endoscopic mucosal resection (EMR). Delayed bleeding and perforation rates were 5.2 % and 0.8 % respectively.

Arbitrarily, they divided their experience into two approximately numerically equal cohorts where 149 (10/2005 to 07/2013) and 145 (07/2013 to 03/2016) lesions were resected. From this the authors were able to include learning curve data, showing improvements in en bloc and R0 resection and reduced bleeding complications and reduced use of intraprocedural clipping during the study. Both en bloc and R0 resection rates improved significantly, from 75.2 % to 91.2 % and from 55.2 % to 84.8 %, respectively, between the two time periods, indicating the importance of procedural experience in optimizing treatment outcomes.

Overall ESD was attempted in 52 lesions later found to have SMIC in this study. In 9 cases ESD was stopped because features suspicious for SMIC were found during the ESD. In those treated endoscopically, en bloc resection was possible in 35/43 (81 %) R0 resection in 28 (65 %), and curative resection in 13 (30 %). The increase in curative resection rate was significant between the two study periods (13.6 % vs. 47.6 %, P = 0.038). The 13 patients in whom curative resection was achieved were able to avoid the morbidity of rectal surgery by virtue of low risk histology. Interestingly, all of these lesions had no endoscopic features predicting SMIC identified prior to resection, whilst those with prior biopsy-proven cancer all had high risk SMIC and were not suitable for endoscopic treatment alone. There was no recurrence during median follow-up of 35 months for this group. 

Yamada et al., from the National Cancer Center (NCC) Tokyo, describe a retrospective case series of long-term outcomes of ESD for 423 LNPCPs treated over 10 years until 2008; median size was 37 mm, 40 % were nongranular, and 47 % were in the proximal colon [2]. Although the lesions are consecutive, in contrast to Probst et al., the authors have included only successful ESD procedures and thus we do not know the complete cohort from which these patients were drawn; this is a limitation of the study that inhibits the ability to place the results in context. En bloc resection was possible in 88 % and R0 excision in 81 %. The endoscopic and cancerous recurrence rates for the 168 eligible lesions at 5 years of follow-up were 4.7 % and 1.2 %, respectively. If an R0 excision was not achieved then recurrence was approximately 15 % at 5 years of follow-up. Risk factors for recurrence were lesion size of ≥ 40 mm and piecemeal resection, in accordance with previously reported studies [3]. Delayed bleeding and perforation rates were 1 % and 3 %, respectively. Overall, 94 lesions were found to have SMIC after ESD. Low risk histology and thus curative resection was possible in only 32 (34 %) lesions. The remaining lesions were found to have submucosal invasion > 1000 µm.

The two studies are comparable for lesion size, adverse events, and cure of low-risk SMIC; however the German study includes only rectal lesions which are at low risk for complications and technically substantially easier to treat, with comparatively easy endoscopic access and scope positioning when compared with more proximal colonic lesions. This is highly relevant as the clinical benefits from endoscopic cure of low risk SMIC are greatest and most relevant in the rectal location where the prospects of short- and long-term morbidity from surgery are very substantial.

The main endoscopic alternative to ESD for treating LNPCP is endoscopic mucosal resection (EMR). Multiple lines of evidence indicate that EMR for LNPCP is efficient (taking 20 % – 30 % of the time of ESD for an equivalent-sized lesion), safe and effective, and that it is mostly performed in an outpatient setting [3] [4] [5], making it substantially more convenient and cost-effective than either ESD or surgery [6]. The technique is generalizable beyond tertiary centers, particularly for granular lesions and those < 40 mm in size. Despite these advantages two significant limitations are often cited: recurrence in follow-up and failure of cure of low risk SMIC.

The first of these is not a major issue provided scheduled surveillance is adhered to as is expected for all oncological procedures. Despite the higher recurrence rates compared to ESD, recurrence after EMR is usually unifocal and diminutive and can be managed endoscopically in most cases [3] [4]. In a large prospective cohort of more than 1000 lesions treated by EMR (mean size 37 mm, 20 % nongranular, 66 % located in the proximal colon), recurrence was16 % at 4 – 6 months and 4 % at 16 months of follow-up. Recurrence was successfully treated in 93 % of cases [3].

Recent innovations in technique mean that recurrence is becoming both much less common and even more easily treated. Moreover it can be predicted and surveillance can be stratified based on data from the index EMR, thus substantially reducing unnecessary follow-up procedures [7]. Irrespective of the primary technique to treat LNPCPs, both ESD and EMR require scheduled surveillance at 6 – 18 months, so ultimately both groups likely require a similar surveillance schedule.

If lesion size is > 15 – 20 mm EMR cannot reliably cure low risk SMIC, and this is where ESD has clear advantages. Selective use of ESD in these cases would offer major clinical advantages, particularly in the rectum or distal left colon where it is technically easier and clinically superior to surgery, which has risks of major morbidity and stoma. The challenge is to accurately identify the LNPCPs most likely to contain low risk SMIC and use ESD in these cases. We need better means of doing this. Current imaging techniques when applied correctly are reasonably accurate for flat lesions [8] [9] [10]; however in our experience they lose their sensitivity in the presence of bulky disease, such as sessile or nodular components where the SMIC may be hidden within and the surface pattern is uniform. More precise and easily applied morphological classification systems are necessary to stratify this risk when surface patterns (pit and vascular) are normal or unhelpful. In LNPCPs with normal surface patterns but stratified to an elevated covert SMIC risk, ESD can be justifiably selected as the most clinically effective and cost-effective treatment, especially in the rectum and distal left colon.

These two studies provide important insights, which must be viewed in context. With the widespread adoption of national bowel cancer screening programs, colonoscopy demands are justifiably burgeoning whilst health expenditure is constrained and treatment decisions are increasingly under pressure. At what point do we strike a logical balance between endoscopic elegance and clinical pragmatism (which delivers its own form of elegance)? If most patients are cured by either EMR or ESD, can we argue for a substantially more burdensome and expensive treatment that delivers no clear clinical gains in the long run for the majority? Clearly, based on current evidence related to recurrence risk, a universal ESD strategy is not justified. The higher cost, longer procedure time, opportunity cost (how many EMRs or screening colonoscopies could be completed in the allocated ESD time), mandatory multiday hospital admission, and higher complication rate of ESD [11] [12] [13] impose major obstacles for ESD as a first-line treatment for LNPCPs. Training is also a significant issue, particularly in the West where the low prevalence of early gastric lesions greatly limits the possibility of training in the comparatively low risk environment of the stomach.

In summary, ESD is generally effective for low risk SMIC but offers no clear benefit, and compared with EMR it is a suboptimal treatment choice for the majority of LNPCPs, which are mostly noninvasive lesions. Better methods of treatment selection are clearly required. Selective ESD can be offered to those 5 % – 10 % of the group at substantial risk of low risk SMIC, and the safer, more efficient and cost-effective alternative of EMR to the remaining majority. Future studies of the treatment of LNPCPs should start from the position of the pathology, not of the treatment offered. We now know enough about the treatment options in isolation, but need to unravel how best to treat the pathology in clinical practice with all the available tools.

• References

• 1 Probst A, Ebigbo A, Maerkl B. et al. Endoscopic submucosal dissection for early rectal neoplasia – experience from a European center. Endoscopy 2017; 49: 222-232
• 2 Yamada M, Saito Y, Takamaru H. et al. Long-term clinical outcomes of endoscopic submucosal dissection for colorectal neoplasms in 423 case series: a retrospective study. Endoscopy 2017; 49: 233-242
• 3 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; 64: 57-65
• 4 Conio M, Repici A, Demarquay JF. et al. EMR of large sessile colorectal polyps. Gastrointest Endosc 2004; 60: 234-241
• 5 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: 1908-1918
• 6 Jayanna M, Burgess NG, Singh R. et al. Cost analysis of endoscopic mucosal resection vs surgery for large laterally spreading colorectal lesions. Clin Gastroenterol Hepatol 2016; 14: 271-278
• 7 Tate DJ, Desomer L, Klein A. et al. Adenoma recurrence after piecemeal colonic EMR is predictable; the Sydney EMR recurrence tool. Gastrointest Endosc 2016; DOI: http://dx.doi.org/10.1016/j.gie.2016.11.027.
• 8 Ikematsu H, Matsuda T, Emura F. et al. Efficacy of capillary pattern type IIIA/IIIB by magnifying narrow band imaging for estimating depth of invasion of early colorectal neoplasms. BMC Gastroenterol 2010; 10: 33
• 9 Hayashi N, Tanaka S, Hewett DG. et al. Endoscopic prediction of deep submucosal invasive carcinoma: Validation of the Narrow-Band Imaging international colorectal endoscopic (NICE) classification. Gastrointest Endosc 2013; 78: 625-632
• 10 Matsuda T, Fujii T, Saito Y. et al. Efficacy of the invasive/non-invasive pattern by magnifying chromoendoscopy to estimate the depth of invasion of early colorectal neoplasms. Am J Gastroenterol 2008; 103: 2700-2706
• 11 Tanaka S, Terasaki M, Kanao H. et al. Current status and future perspectives of endoscopic submucosal dissection for colorectal tumors. Dig Endosc 2012; 24 (Suppl. 01) 73-79
• 12 Saito Y, Uraoka T, Yamaguchi Y. et al. A prospective, multicenter study of 1111 colorectal endoscopic submucosal dissections (with video). Gastrointest Endosc 2010; 72: 1217-1225
• 13 Saito Y, Fukuzawa M, Matsuda T. et al. Clinical outcome of endoscopic submucosal dissection versus endoscopic mucosal resection of large colorectal tumors as determined by curative resection. Surg Endosc 2010; 24: 343-352