Endoscopic Mucosal Resection: This Is Our Turf

Endoscopic mucosal resection (EMR) is emerging as one of the most important techniques to be incorporated into gastrointestinal endoscopy. Any doubts should be dispelled by four papers relating to EMR in this issue of Endoscopy [1] [2] [3] [4]. EMR greatly expands the already potent combination of the therapeutic power of endoscopic surgery with the diagnostic power of surgical pathology.

EMR is a technique designed to remove flat, slightly protruding, or slightly depressed mucosal lesions by cutting through the submucosa. The method can be practiced using a variety of endoscopic techniques that have been primarily developed in Japan, where gastric cancer remains common [5]. Early gastric cancer almost always occurs in flat lesions, unlike colon cancer which largely presents in its earliest stages as polyps that can be more readily snared at endoscopy [6]. Japanese endoscopists have extended EMR techniques to the esophagus and colon. In the West, EMR has been used primarily to facilitate the removal of sessile colon polyps and, recently, in the staging and treatment of high-grade intraepithelial neoplasia (HGIN; previously called high-grade dysplasia) and intramucosal cancer in Barrett’s esophagus.

The incidences of Barrett’s esophagus related to chronic gastroesophageal reflux disease, and of esophageal adenocarcinoma complicating Barrett’s esophagus, have increased rapidly in the United States and Western Europe over the past three decades [7]. Most patients with esophageal adenocarcinoma present with symptoms of dysphagia and weight loss, and have advanced disease. Endoscopic screening and surveillance for Barrett’s esophagus have detected an increasing number of patients with HGIN and superficial adenocarcinoma [8] [9].

The management of HGIN in Barrett’s esophagus has been a particularly difficult area in clinical decision making. Even for expert pathologists, the differentiation of HGIN from invasive adenocarcinoma on endoscopic biopsy can be problematic [10]. In several surgical series, esophagectomy specimens in patients with HGIN on endoscopic biopsy have been found to have invasive cancer at rates of about 30 % and higher. In this context, esophagectomy has been a standard recommendation for treatment of HGIN in Barrett’s esophagus [9].

Esophagectomy, which is usually curative of HGIN and intramucosal cancer, is associated with significant morbidity and mortality. While operative mortality has decreased in centers of excellence in recent years, it remains a significant risk at 1 - 2 %, and markedly increases in hospitals where esophagectomy is infrequently performed [11]. The incidence of HGIN and cancer increases with age. The probability of co-morbid illness in older patients increases the morbidity and mortality associated with esophagectomy, and often makes surgery an unsatisfactory or unacceptable option.

Endoscopic therapy of high-grade dysplasia and adenocarcinoma limited to the mucosa has been proposed as a minimally invasive method of treatment [12]. The basis for endoscopic therapy is that the rate of lymph node metastasis in HGIN is nil, and in invasive cancer confined to the mucosa it is very low, as confirmed in recent reports analyzing surgical pathology results following esophagectomy in patients with Barrett’s esophagus [13] [14] [15].

Endoscopic ablative therapies, including photodynamic therapy and argon plasma coagulation, have been used successfully for HGIN in Barrett’s esophagus [16] [17]. The problem with ablative therapies is the absence of confirmation that disease is confined to the mucosa. Endoscopic ultrasound has improved the staging of tumor depth of invasion (T), with accuracy in the range of 75 - 85 % when high-frequency probes are used in distinguishing between mucosal (T1m) and submucosal (T1sm) disease [18]. EUS remains extremely useful in identifying some patients with submucosal invasion where the risk of lymph node metastasis is high; T1sm detection on EUS should generally preclude endoscopic therapy with curative intent.

The great advantage of EMR over ablative methods is the provision of a large pathology specimen extending into the submucosa [19]. The depth of cancer invasion can be assessed with great accuracy [18]. Disease confined to the mucosa with clear margins can be considered to be completely resected. Patients with submucosal invasion are usually referred for surgery and oncology therapies [19]. The staging power of EMR specimens is amply demonstrated by Vieth et al. in a multicenter study reported in this issue [1]. In an analysis of 711 EMR specimens in 295 patients, mucosal carcinomas were most common, accounting for 80.3 % of the patients. Submucosal invasion was less common, accounting for 16 % of the patients overall. Deeper invasion of the submucosa could be differentiated from a focal minimal invasion, which might represent an acceptable risk for endoscopic therapy in selected patients.

Careful follow-up of larger numbers of patients treated with endoscopic therapy based on EMR pathology staging will be needed, but accumulating data continues to look highly promising. The pathological techniques applied by Vieth et al. were meticulous; specimens were cut into 1.5 mm wide strips, and at least eight stepped 3 - 4 µm thick slices were cut from each block. As EMR becomes more widely used, pathologists will need to be trained in processing and interpretation of EMR specimens. Good communication and dialogue between endoscopist and pathologist is essential. Artifacts associated with the EMR technique, for example those related to submucosal fluid injection, must be understood by the pathologist and should not become pitfalls.

The fact that most of the cancers in EMR specimens examined by Vieth et al. were well to moderately differentiated, and did not invade lymphatic or blood vessels, gave further reassurance that patients treated with EMR for intramucosal cancer were unlikely to develop lymph node metastases. Basal resection margins were clear in all but 8.8 % of patients on first resection specimens. On the other hand, 70.5 % of the patients showed tumor extending to the lateral resection lines of EMR specimens, and in an additional 2 % of patients, cautery artifacts precluded definitive lateral margin staging. Thus, repeat EMR procedures were required in many patients, with clear margins eventually achieved in 74.5 % of the study patients.

How to manage the remaining Barrett’s metaplasia after removal of all evident neoplasia by EMR remains an open question. Residual metaplasia leaves the possibility of synchronous undetected neoplasia and the development of subsequent metachronous neoplasia. One possible response is ablation of the residual Barrett’s tissue with thermal treatments or with photochemical ablation using photodynamic therapy [20]. These methods preclude pathological confirmation of removed tissue. The attraction of circumferential EMR is that all evident Barrett’s tissue can be removed and submitted for pathological analysis.

A major consideration when contemplating circumferential EMR is the length of the Barrett’s segment. Short segments are rather easily and effectively removed [21]. The removal of long segments with circumferential EMR has resulted in healing being accompanied by contractures, fibrosis, or strictures in some cases [22]. Giovannini et al. in this issue, report on the performance of circumferential EMR, mostly with an inject, lift, and cut technique, where half the circumferential area of Barrett’s mucosa, with the index neoplasm, was removed at an initial session, followed by removal of the second half of the circumferential area 1 month later [2]. The mean length of the Barrett’s segment was about 3.5 cm (range 2 - 5 cm), obviously not very long. Still it was impressive that no strictures developed when this two-step method was employed. We have used a similar two-step piecemeal approach, with no development of strictures [19]. In the series of Giovannini et al., two patients required additional EMR. Complete replacement of Barrett’s tissue with squamous mucosa was achieved in 75 %.

It seems likely that EMR techniques will continue to evolve and improve [23]. Japanese investigators have developed methods for en bloc resections of large neoplastic lesions using a technique called endoscopic submucosal dissection (ESD), employing variations on the needle-knife. En bloc resections would largely eliminate the problem of positive tumor margins associated with piecemeal EMR. ESD has been primarily used in the stomach and colon, and the procedures tend to be lengthy and associated with a significant perforation rate. Although the successful use of ESD at the esophagogastric junction and gastric cardia has recently been reported [24], a satisfactory method for en bloc resection of larger esophageal cancers in Barrett’s esophagus has yet to be perfected.

EMR techniques have also been employed successfully to achieve histological diagnosis and remove submucosal tumors throughout the gastrointestinal tract. EUS remains essential in determining tumor size, location, and vascularity, but usually is not sufficient to fully predict histology (with the possible exception of lipomas). Specific tumor markers are now required for diagnosis of gastrointestinal stromal tumors (GIST) [25].

EUS-guided fine-needle aspiration or cutting-needle biopsy of submucosal tumors can often be achieved, but sometimes this is with difficulty, and the tumor remains in place. Wehrmann et al., in this issue, describe their results using EMR for submucosal tumors of the esophagus [3]. EUS was used to avoid attempts at EMR in tumors involving the muscularis propria. The endoscopists used a variety of EMR methods (band and cut, lift and cut, cap-assisted) to remove tumors of mean diameter 17 mm (8 - 34 mm). Complete resection was achieved in 18/20 patients without significant complication.

Rösch, et al., in this issue, report their initial experience (certainly one of the largest outside Japan) of using a needle-knife with an insulated tip to achieve or assist in a one-piece en bloc resection [4]. In 37 patients with upper gastrointestinal lesions, en bloc resection was possible in only 25 % of mucosal lesions and 37 % of submucosal tumors. Extending the definition of success to include piecemeal resections, a satisfactory outcome was achieved in 65 % of mucosal lesions and 78 % of submucosal tumors. There were significant complications in 16 %, including a perforation and three hemorrhages requiring re-intervention, and many of the procedures were quite lengthy, in the range of 1 - 2 hours, in the hands of the expert and experienced endoscopists.

The ideal EMR technique would provide an en bloc specimen to allow definitive pathological assessment of resection margins even for larger tumors greater than 20 mm in diameter. This would almost certainly result in a lower recurrence rate than piecemeal resection, as Japanese endoscopists have demonstrated for early gastric cancer. Instruments and methods will no doubt continue to improve, and make the ideal more approachable.

As Rösch et al. point out, the insulated-tip knife is useful and a step in the right direction [2], and I suppose the same can be said of other needle-knife variations, such as the hook knife, triangle-tip knife, and sheathed knife. Hyaluronic acid seems to be better than saline for prolonged lifting during ESD, and cheaper fluids such as methylcellulose may also be helpful. Problems remain to be solved: how to effectively retract the mucosa during submucosal dissection; how to achieve a good angle of approach in different locations; and how to securely close perforations. Endoscopes and accessories specifically designed for EMR will likely be required.

The evolving nature of EMR technique should in no way detract from its current capabilities and its implications for gastrointestinal endoscopy. As radiological (virtual) diagnosis will certainly improve, and new endoscopic optical methods will increasingly detect early neoplastic change in gastrointestinal mucosa, the need for EMR will increase. The ability to both diagnose and definitively treat early neoplasia should maintain the primary role of endoscopy for this purpose. No surgical method for resection approaching from the serosal side of the gastrointestinal tract should be competitive. The gastrointestinal mucosa, of course, has become the principal turf of the endoscopist. EMR should guarantee that it remains that way.

References

• 1 Vieth M, Ell C, Gossner L. et al . Histological analysis of endoscopic resection specimens from 326 patients with Barrett’s esophagus and early neoplasia.  Endoscopy. 2004;  36 776-781
• 2 Giovannini M, Bories E, Presenti C. et al . Circumferential endoscopic mucosal resection in Barrett’s esophagus with high-grade intraepithelial neoplasia or mucosal cancer. Preliminary results in 21 patients.  Endoscopy. 2004;  36 782-787
• 3 Wehrmann T, Martchenko K, Nakamura M. et al . Endoscopic resection of submucosal esophageal tumors. A prospective case series.  Endoscopy. 2004;  36 802-807
• 4 Rösch T, Sarbia M, Schumacher B. et al . Attempted endoscopic en bloc resection of mucosal and submucosal tumors using insulated-tip knives: a pilot series.  Endoscopy. 2004;  36 788-801
• 5 Soetikno R M, Gotoda T, Nakanishi Y, Soehendra . Endoscopic mucosal resection.  Gastrointest Endosc. 2003;  57 567-579
• 6 . The Paris endoscopic classification of superficial neoplastic lesions: esophagus, stomach, and colon.  Gastrointest Endosc. 2003;  58 (Suppl) S3-S43
• 7 Devesa S S, Blot W J, Fraumeni J F. Changing patterns in the incidence of esophageal and gastric cancer in the United States.  Cancer. 1998;  83 2049-2053
• 8 Sampliner R E, and the Practice Parameters Committee of the American College of Gastroenterology. Updated guidelines for the diagnosis, surveillance, and therapy of Barrett’s esophagus.  Am J Gastroenterol. 2002;  97 1888-1895
• 9 Spechler S J. Barrett’s esophagus.  N Engl J Med. 2002;  346 836-842
• 10 Ormsby A, Petras R E, Henricks W H. et al . Observer variation in the diagnosis of superficial oesophageal adenocarcinoma.  Gut. 2002;  51 671-676
• 11 Birkmyer J D, Siewers A E, Finlayson E V. et al . Hospital volume and surgical mortality in the United States.  N Engl J Med. 2002;  346 1128-1137
• 12 Lightdale C J. Ablation therapy for Barrett’s esophagus: is it time to choose our weapons?.  Gastointest Endosc. 1999;  49 122-125
• 13 Nigro J J, Hagen J A, DeMeester T R. et al . Prevalence and location of nodal metastases in distal esophageal adenocarcinoma confined to the wall: implications for therapy.  J Thorac Cardiovasc Surg. 1999;  117 16-25
• 14 Rice T W, Zuccaro G, Adelstein D J. et al . Esophageal carcinoma: depth of tumor invasion is predictive of regional lymph node status.  Ann Thorac Surg. 1998;  65 787-792
• 15 Stein H J, Feith M, Mueller J. et al . Limited resection for early adenocarcinoma in Barrett’s esophagus.  Ann Surg. 2000;  232 733-742
• 16 Overholt B F, Panjehpour M, Halberg D L. Photodynamic therapy for Barrett’s esophagus with dysplasia and/or early stage carcinoma: long-term results.  Gastrointest Endosc. 2003;  58 183-188
• 17 Attwood S EA, Lewis C J, Caplin S. et al . Argon beam plasma coagulation as therapy for high-grade dysplasia in Barrett’s esophagus.  Clin Gastroenterol Hepatol. 2003;  1 258-263
• 18 Ell C, May A, Gossner L. et al . Endoscopic mucosal resection of early cancer and high-grade dysplasia in Barrett’s esophagus.  Gastroenterology. 2000;  118 670-677
• 19 Lightdale C J, Larghi A, Rotterdam H, Okpara N. Endoscopic ultrasonography (EUS) and endoscopic mucosal resection (EMR) for staging and treatment of high-grade dysplasia (HGD) and early adenocarcinoma (EAC) in Barrett’s esophagus (BE).  Gastrointest Endosc. 2004;  59 AB90
• 20 Pacifico R J, Wang K K, Wongkeesong L-M. et al . Combined endoscopic mucosal resection and photodynamic therapy versus esophagectomy for management of early adenocarcinoma in Barrett’s esophagus.  Clin Gastroenterol Hepatol. 2003;  1 252-257
• 21 May A, Gossner L, Pech O. et al . Intraepithelial high-grade neoplasia and early adenocarcinoma in short-segment Barrett’s esophagus (SSBE): curative treatment using local endoscopic treatment techniques.  Endoscopy. 2002;  34 604-610
• 22 Seewald S, Akaraviputh T, Seitz U. et al . Circumferential EMR and complete removal of Barrett’s epithelium: a new approach to management of Barrett’s esophagus containing high-grade intraepithelial neoplasia and intramucosal carcinoma.  Gastrointest Endosc. 2003;  57 854-859
• 23 Miyamoto S, Muto M, Hamamoto Y. et al . A new technique for endoscopic mucosal resection with an insulated-tip electrosurgical knife improves the completeness of resection of intramucosal gastric neoplasms.  Gastrointest Endosc. 2002;  55 576-581
• 24 Yahagi N, Fujishiro M, Kakushima N. et al . Endoscopic submucosal dissection for lesions of the esophago-gastric junction and gastric cardia.  Gastrointest Endosc. 2004;  59 AB171
• 25 Chandu d e, Reid R. Gastrointestinal stromal tumors (GIST): C-kit mutations, CD 117 expression, differential diagnosis and targeted cancer therapy with imatinib.  Pathol Oncol Res. 203;  9 13-19