AGA Clinical Practice Update on the Utility of Endoscopic Submucosal Dissection in T1b Esophageal Cancer: Expert Review

Abstract

Among esophageal neoplasms, the 2 most common cancers by cell type are squamous cell carcinoma (ESCC) and adenocarcinoma (EAC). Whereas ESCC is the most common type of cancer worldwide,1Lin Y. Totsuka Y. He Y. et al.Epidemiology of esophageal cancer in Japan and China.J Epidemiol. 2013; 23: 233-242Crossref PubMed Scopus (374) Google Scholar EAC is more prevalent than ESCC in the Western world, especially in North America.2Hur C. Miller M. Kong C.Y. et al.Trends in esophageal adenocarcinoma incidence and mortality.Cancer. 2013; 119: 1149-1158Crossref PubMed Scopus (334) Google Scholar During the last 2 decades, EAC has shown an increasing trend in incidence of diagnosis. Yet overall survival for EAC remains disappointing with a poor 5-year survival rate. Early stage EACs (T1a and T1b) have a much better prognosis than late stage EACs. It is remarkable that these early staged cancers comprise approximately 20% of all cases of EAC diagnosed in the United States.3Das A. Singh V. Fleischer D.E. et al.A comparison of endoscopic treatment and surgery in early esophageal cancer: an analysis of surveillance epidemiology and end results data.Am J Gastroenterol. 2008; 103: 1340-1345Crossref PubMed Scopus (123) Google Scholar Esophagectomy has been the mainstay for esophageal cancer (EC) but has high morbidity and mortality. Even a high volume center such as Mayo Clinic reported a surgical mortality of 4% for T1a esophageal cancer.4Prasad G.A. Wu T.T. Wigle D.A. et al.Endoscopic and surgical treatment of mucosal (T1a) esophageal adenocarcinoma in Barrett's esophagus.Gastroenterology. 2009; 137: 815-823Abstract Full Text Full Text PDF PubMed Scopus (259) Google Scholar Moreover, 34% of patients developed postoperative complications such as anastomotic leaks, anastomotic strictures, cardiopulmonary complications, and feeding jejunostomy leaks. The majority of patients who successfully undergo esophagectomy will experience gastroesophageal reflux symptoms5Fuchs H.F. Schmidt H.M. Meissner M. et al.Endoscopic and histopathologic reflux-associated mucosal damage in the remnant esophagus following transthoracic esophagectomy for cancer-5-year long-term follow-up.Dis Esophagus. 2018; 31: 1-6Crossref PubMed Scopus (3) Google Scholar with the surgically absent lower esophageal sphincter, which brings a profound negative impact on their quality of life. Therefore, a less invasive alternative to esophagectomy would be extremely valuable in the management of early stage of EC if proven effective. Endoscopic submucosal dissection (ESD) has been gaining momentum as an alternative to surgery in treating early gastrointestinal neoplasms. If ESD provides treatment as effective as surgical resection for T1b EC, the patients with T1b EC will benefit from a shorter hospital stay without diminished quality of life from new or increased reflux symptoms. On the other hand, if ESD for T1b EC is an inadequate treatment with a decreased survival rate when compared with esophagectomy, it should be discouraged as a primary form of therapy for these curable patients. The combination of ESD with adjuvant chemotherapy and radiation therapy may also prove to be better tolerated without the need for long recovery needed for surgical therapy. Within early EC, tumors limited to mucosa are defined as T1a on the basis of TNM classification. T1b early EC is limited to the submucosa without any deeper invasion to the muscularis propria. T1b tumors can be subdivided into SM1, SM2, and SM3 tumors. SM1 tumors are limited to the superficial submucosa, with a limited depth of 200 μm or 500 μm for ESCC or EAC, respectively. SM2 and SM3 tumors are invading the submucosa, with a depth more than 200 μm for ESCC or more than 500 μm for EAC.6Japan Esophageal Society Japanese classification of esophageal cancer, 11th edition: part I.Esophagus. 2017; 14: 1-36Crossref PubMed Scopus (378) Google Scholar, 7The Paris endoscopic classification of superficial neoplastic lesions: esophagus, stomach, and colon—November 30 to December 1, 2002.Gastrointest Endosc. 2003; 58: S3-S43Abstract Full Text Full Text PDF PubMed Scopus (1568) Google Scholar The detection of early EC has improved because of the advances in optical imaging techniques such as narrow-band imaging (NBI), magnifying endoscopy, endomicroscopy, and chromoendoscopy. Lugol’s iodine is a chromoendoscopy agent that is commonly used for the detection of ESCC. Lugol’s iodine selectively stains normal squamous mucosa as dark brown, leaving precancerous and early cancerous lesions unstained, “void lesions”. (Figure 1 illustrates early ESCC lesion seen by chromoendoscopy as Lugol’s voiding lesion). Lugol’s voiding lesions can be false positive for areas of neosquamous tissue and in certain well-differentiated squamous cell cancers, so careful observation and biopsy are still required. For better identification of margins within Barrett’s esophagus, methylene blue or crystal violet is used. Both agents have affinity to intestinal epithelium within Barrett’s esophagus, and they highlight dysplastic pit patterns within Barrett’s epithelium. Recognizing features of deep invasion is important in assessing early EC and in differentiating T1b tumor from other stages of EC. Assessment of the depth of invasion by using endoscopy has been more rigorously investigated in ESCC as compared with EAC. In a retrospective analysis of 203 patients with ESCC, surface nodularity, granularity, deep depression, and thick notch (a very deep groove within a mass) were all associated with higher risk of deep invasion beyond the upper third of the submucosa (SM1).8Bae J. Shin I.S. Min Y.W. et al.Endoscopic prediction model for differentiating upper submucosal invasion (<200 mum) and beyond in superficial esophageal squamous cell carcinoma.Oncotarget. 2018; 9: 9156-9165Crossref PubMed Scopus (2) Google Scholar Paris classification of lesion’s morphology subdivided superficial tumors (Paris type 0) into 3 categories, type 0-I for protruding lesions, type 0-II for flat lesions, and type 0-III for excavated lesions. Paris type 0-II lesions can be subdivided into subtypes 0-II A, B, or C according to the degree of elevation/depression7The Paris endoscopic classification of superficial neoplastic lesions: esophagus, stomach, and colon—November 30 to December 1, 2002.Gastrointest Endosc. 2003; 58: S3-S43Abstract Full Text Full Text PDF PubMed Scopus (1568) Google Scholar (Figure 2). In ESCC, type 0-I protruding lesions and type 0-III excavated lesions carry the highest risk of deep submucosal invasion (79% and 84%, respectively).9Kodama M. Kakegawa T. Treatment of superficial cancer of the esophagus: a summary of responses to a questionnaire on superficial cancer of the esophagus in Japan.Surgery. 1998; 123: 432-439Abstract Full Text Full Text PDF PubMed Scopus (375) Google Scholar This contrasts with the morphology of the lesions in the colon in which type 0-I protruding lesions are generally superficial. Japan Esophageal Society developed a NBI classification with magnified endoscopy to assist in prediction of depth of invasion for ESCC. The classification depends on the degree of microvascular irregularity in intrapapillary capillary loops (IPCLs). Generally speaking, IPCLs with irregularity, dilation, and loop-like formation are labeled as type B vessels (type A is normal IPCL pattern). When significant dilation, increase in caliber, or tortuosity is noted in type B vessels, the suspicion for tumor invasion into deep submucosa (SM2) increases. Type B vessels had 90.5% accuracy in estimating the depth of invasion in early ESCC.10Oyama T. Inoue H. Arima M. et al.Prediction of the invasion depth of superficial squamous cell carcinoma based on microvessel morphology: magnifying endoscopic classification of the Japan Esophageal Society.Esophagus. 2017; 14: 105-112Crossref PubMed Scopus (3) Google Scholar It is worth mentioning that the unavailability of magnified endoscopy poses limitations for the application of NBI with magnified endoscopy classifications in many centers in the United States. Newer endoscopes are available with similar magnification systems as in Japan, which should improve the situation in the United States. Endoscopic ultrasound (EUS) can overstage neoplasia in Barrett’s esophagus, especially T1b tumors. In a meta-analysis of 11 studies that evaluated the diagnostic accuracy of EUS in staging patients with Barrett’s esophagus, the pooled false-positive rate was 35.85% (95% confidence interval, 25.3%–47.8%). In other words, 36% of patients diagnosed as having T1b lesions on EUS were found to have low-grade dysplasia, high-grade dysplasia, or T1a tumor on histology.11Qumseya B.J. Bartel M.J. Gendy S. et al.High rate of over-staging of Barrett's neoplasia with endoscopic ultrasound: systemic review and meta-analysis.Dig Liver Dis. 2018; 50: 438-445Abstract Full Text Full Text PDF PubMed Scopus (9) Google Scholar Similar findings were also noted in ESCC. In a retrospective study of 174 patients with ESCC, NBI was shown to have higher accuracy than EUS in differentiating tumors limited to superficial mucosa compared with tumors with muscularis mucosa or SM1 invasion.12Mizumoto T. Hiyama T. Oka S. et al.Diagnosis of superficial esophageal squamous cell carcinoma invasion depth before endoscopic submucosal dissection.Dis Esophagus. 2018; 31: 1-7Crossref Scopus (12) Google Scholar (Figure 3 illustrates a T1b lesion on EUS.) Computed tomography scan with positron emission tomography for early stage cancer does not appear to be needed, although it is often recommended as part of EC staging and work-up. In particular, benign lesions that are found during initial staging often can be confounders, and the evaluation of these lesions can delay needed therapy.13Cuellar S.L. Carter B.W. Macapinlac H.A. et al.Clinical staging of patients with early esophageal adenocarcinoma: does FDG-PET/CT have a role?.J Thorac Oncol. 2014; 9: 1202-1206Abstract Full Text Full Text PDF PubMed Scopus (37) Google Scholar At the current time, imaging of the submucosa is clinically best done with EUS, although there are suggestions that volume laser endomicroscopy using infrared light appears to be capable to discern the submucosal presence of disease and may be more accurate in determining submucosal invasion.14Swager A.F. Tearney G.J. Leggett C.L. et al.Identification of volumetric laser endomicroscopy features predictive for early neoplasia in Barrett's esophagus using high-quality histological correlation.Gastrointest Endosc. 2017; 85: 918-926 e7Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar The risk of lymph node metastasis (LNM) depends on the depth of invasion, histologic type, and molecular characterization of the tumor. Table 1 summarize predictors of higher risk for lymph node metastasis in esophageal cancer. For T1a EAC, lymph node involvement is rarely seen regardless of the depth of invasion in the mucosa. Early ESCC limited to the epithelium (M1) or to the lamina propria (M2) has a negligible risk for LNM. However, there is a slight increase in LNM in ESCC invading into deep mucularis mucosa (M3 lesions), which is estimated to be ∼10% in published literature.15Li B. Chen H. Xiang J. et al.Prevalence of lymph node metastases in superficial esophageal squamous cell carcinoma.J Thorac Cardiovasc Surg. 2013; 146: 1198-1203Abstract Full Text Full Text PDF PubMed Scopus (45) Google Scholar, 16Siewert J.R. Stein H.J. Feith M. et al.Histologic tumor type is an independent prognostic parameter in esophageal cancer: lessons from more than 1,000 consecutive resections at a single center in the Western world.Ann Surg. 2001; 234 (discussion 368–369): 360-367Crossref PubMed Scopus (440) Google Scholar Therefore, the majority of T1a tumors are considered curable by endoscopic resection, with a favorable 5-year survival rate as seen in multiple Japanese studies. In ESCC, T1b lesions with SM1 invasion have a relatively low risk of LNM (8%–26.5%), whereas T1b lesions with SM2 invasions have a significantly higher LNM risk, up to 60%.10Oyama T. Inoue H. Arima M. et al.Prediction of the invasion depth of superficial squamous cell carcinoma based on microvessel morphology: magnifying endoscopic classification of the Japan Esophageal Society.Esophagus. 2017; 14: 105-112Crossref PubMed Scopus (3) Google Scholar Well-differentiated T1b EAC with tumor size smaller than 2 cm had a low risk of LNM (4.2%) in 782 patients who underwent esophagectomy in the United States. Tumors that were poorly differentiated and/or larger than 2 cm had a higher rate of LNM in the same cohort.17Newton A.D. Predina J.D. Xia L. et al.Surgical management of early-stage esophageal adenocarcinoma based on lymph node metastasis risk.Ann Surg Oncol. 2018; 25: 318-325Crossref PubMed Scopus (24) Google Scholar In a similar cohort of 90 patients who underwent esophagectomy for T1 EAC at M. D. Anderson, Houston, TX, LNM was noted in 8% of patients who had tumor invading into the superficial submucosa. When lesions were stratified by tumor size, 75% of T1b tumors larger than 1.2 cm had LNM.18Liu L. Hofstetter W.L. Rashid A. et al.Significance of the depth of tumor invasion and lymph node metastasis in superficially invasive (T1) esophageal adenocarcinoma.Am J Surg Pathol. 2005; 29: 1079-1085Crossref PubMed Google Scholar In a meta-analysis of 7645 patients who had T1b early EC, the risk of LNM was dependent on the depth of invasion within the submucosa and the histology of the tumor. EAC patients with SM1 tumor had the lowest LNM (6%), whereas ESCC patients with SM3 had the highest LNM (55%). Overall, the rate of LNM was higher in SM1 and SM2 ESCC patients compared with that of EAC patients. In addition, micrometer-based assessment for depth of invasion can be more difficult to obtain in U.S. centers and should be discussed with pathologists when beginning endoscopic resections.19Gockel I. Sgourakis G. Lyros O. et al.Risk of lymph node metastasis in submucosal esophageal cancer: a review of surgically resected patients.Expert Rev Gastroenterol Hepatol. 2011; 5: 371-384Crossref PubMed Scopus (67) Google Scholar Emerging studies showed that several immunohistochemistry markers are associated with higher LNM such as E-cadherin and cyclin D1.20Shiozaki H. Doki Y. Yamana H. et al.A multi-institutional study of immunohistochemical investigation for the roles of cyclin D1 and E-cadherin in superficial squamous cell carcinoma of the esophagus.J Surg Oncol. 2002; 79: 166-173Crossref PubMed Scopus (28) Google Scholar It is worth noting that the muscularis mucosa could be duplicated or trifurcated in Barrett’s esophagus. This may result in overstaging of tumor by EUS or when evaluating specimens removed by endoscopic mucosal resection (EMR). Early studies suggested that the space between the duplicated muscularis mucosa could be as rich in lymphatic channels as submucosal space.21Lewis J.T. Wang K.K. Abraham S.C. Muscularis mucosae duplication and the musculo-fibrous anomaly in endoscopic mucosal resections for barrett esophagus: implications for staging of adenocarcinoma.Am J Surg Pathol. 2008; 32: 566-571Crossref PubMed Scopus (58) Google Scholar However, more recent analysis of the duplicated area suggests that the lamina propria of this region is similar to the mucosa. In retrospective analysis of 99 patients who underwent esophagectomy for early EAC, 41 patients had duplicated muscularis mucosa. In this cohort, the risk of LNM was similar between tumor invading into the space between the duplicated muscularis mucosa and tumor limited to mucosa (T1a). This finding suggests that tumor invading into the duplicated muscularis mucosa space should be classified as T1a tumor rather than T1b tumor.22Estrella J.S. Hofstetter W.L. Correa A.M. et al.Duplicated muscularis mucosae invasion has similar risk of lymph node metastasis and recurrence-free survival as intramucosal esophageal adenocarcinoma.Am J Surg Pathol. 2011; 35: 1045-1053Crossref PubMed Scopus (30) Google Scholar Overall, the data mentioned above suggest that well-differentiated, small (less than 2 cm) T1b early EC localized to SM1 and without lymphovascular invasion on the resected specimen is the ideal scenario for endoscopic treatment alone. Table 2 summarize risk of LNM in submucosal esophageal cancer based on cancer type. As noted by the American Joint Committee on Cancer (AJCC), current guidelines would suggest that the early stage cancers are T1a or T1b cancers, depending on their grade, which has become a new modifier in the AJCC. The grading system is categorized as GX for tumors in which differentiation cannot be assessed, G1 for well-differentiated tumors, G2 for moderately differentiated tumors, and G3 for poorly differentiated tumors. Differentiation is assessed by the percentage of well-formed gland in the tumor (>95% in differentiated tumors and <50% in poorly differentiated tumors).23Rice T.W. Patil D.T. Blackstone E.H. 8th edition AJCC/UICC staging of cancers of the esophagus and esophagogastric junction: application to clinical practice.Ann Cardiothorac Surg. 2017; 6: 119-130Crossref PubMed Scopus (221) Google Scholar The grade of the cancer as determined by histopathology also now impacts overall staging. Higher-grade tumors including grade 2 to 3 cancers that are T1a would be staged as stage 1b cancer. Grade 1 cancers that are T1a would continue to be staged as stage 1a tumors. If they are T1b lesions that extend into the submucosa, they are all staged as stage 1b lesions. This is not location specific even in the setting of squamous cancers. Endoscopic resection can be accomplished by either EMR or ESD. EMR has an inherent limitation of not being able to provide an accurate assessment of the lateral margins in piecemeal resection specimens, whereas ESD can provide en bloc specimens regardless of tumor size with lower residual and local recurrences. In addition to its therapeutic role, ESD is considered as a diagnostic tool. After obtaining a pathologic diagnosis with endoscopy and biopsy, ESD can accurately upstage or downstage the tumor pathology on the basis of careful examination of the entire resected tumor.24Coman R.M. Gotoda T. Forsmark C.E. et al.Prospective evaluation of the clinical utility of endoscopic submucosal dissection (ESD) in patients with Barrett's esophagus: a Western center experience.Endosc Int Open. 2016; 4: E715-E721Crossref PubMed Google Scholar For T1b EC, however, esophagectomy remains the mainstay treatment because the prevalence of LNM in these patients is not negligible. A study specifically reported outcomes on clinical T1bN0M0 ESCC in 173 patients where 102 of 173 were treated with radical esophagectomy and the rest with definitive chemoradiation (CRT). The 3- and 5-year survival rates were 87.0% and 77.7% for the surgery group and 77.8% and 68.6% for the CRT group, respectively. The surgery group had disease recurrence in 12 patients (11.8%) and the CRT group in 20 patients (28.2%).25Motoori M. Yano M. Ishihara R. et al.Comparison between radical esophagectomy and definitive chemoradiotherapy in patients with clinical T1bN0M0 esophageal cancer.Ann Surg Oncol. 2012; 19: 2135-2141Crossref PubMed Scopus (61) Google Scholar ESD data on adenocarcinoma of the esophagus is sparse. A comparative retrospective study involving 79 patients with Siewert type II adenocarcinoma showed a 5-year overall survival rate of 93.9% for ESD group (40) and 97.3% for the surgery group (39) (P = .376). The treatment-related adverse events were similar in the 2 groups (10.0% vs 17.9%, P = .308), with no mortality in either group. Of all the tumors, 46 (58.2%) were well-differentiated, and 33 (41.8%) were moderately differentiated adenocarcinomas, with 64 (81.0%) T1a and 15 (19.0%) T1b. Synchronous or metachronous lesions developed in 4 patients in ESD group.26Gong E.J. Kim D.H. Ahn J.Y. et al.Comparison of long-term outcomes of endoscopic submucosal dissection and surgery for esophagogastric junction adenocarcinoma.Gastric Cancer. 2017; 20: 84-91Crossref PubMed Scopus (25) Google Scholar (Figure 4 highlights post-ESD resection of T1 b lesions with central fibrosis and exposure of few muscularis propria fibers.) Endoscopic treatment of T1b EC followed by CRT in patients who decline or are unfit for surgery has been reported. In a retrospective study of 66 patients with ESCC (T1b lesions, 65%) who underwent endoscopic resection followed by CRT, the 3- and 5-year survival rates were 87% and 75%, respectively, which is similar to survival rates after esophagectomy. Six of 36 patients with LNM had metastatic recurrence, whereas none of the 30 patients without LNM had metastatic recurrence.27Hamada K. Ishihara R. Yamasaki Y. et al.Efficacy and safety of endoscopic resection followed by chemoradiotherapy for superficial esophageal squamous cell carcinoma: a retrospective study.Clin Transl Gastroenterol. 2017; 8: e110Crossref PubMed Scopus (22) Google Scholar Another retrospective study compared ESD with CRT with CRT only in 47 patients with early ESCC. The 3-year overall survival rate in ESD with CRT group was 90% compared with 63.2% in the CRT only group. Recurrence was seen in 6.3% of patients in ESD CRT group compared with 29.0% for CRT group alone.28Kawaguchi G. Sasamoto R. Abe E. et al.The effectiveness of endoscopic submucosal dissection followed by chemoradiotherapy for superficial esophageal cancer.Radiat Oncol. 2015; 10: 31Crossref PubMed Scopus (56) Google Scholar The aforementioned data are promising, and more data are expected to emerge regarding the role of ESD with CRT in EAC. Combination of ESD with radiofrequency ablation (RFA) to achieve complete clearance of dysplasia in EAC has been reported in the literature. Subramaniam et al29Subramaniam S. Kandiah K. Chedgy F. et al.The safety and efficacy of radiofrequency ablation following endoscopic submucosal dissection for Barrett's neoplasia.Dis Esophagus. 2018; 31: 1-7Crossref Scopus (3) Google Scholar evaluated the efficacy of ESD followed by RFA for eradication of dysplasia in a retrospective trial of 27 patients with Barrett’s esophagus and high-grade dysplasia or EAC (2 of them had T1b tumor). Clearance of dysplasia was achieved in 26 patients (96%), and clearance of intestinal metaplasia was achieved in 23 patients (85%). Similar results were seen in a prospective single center trial in which eradication of intestinal metaplasia was achieved in 8 of 10 patients with EAC who underwent ESD followed by RFA. Generally, 1–2 sessions of RFA were sufficient for eradication of intestinal metaplasia.30Neuhaus H. Terheggen G. Rutz E.M. et al.Endoscopic submucosal dissection plus radiofrequency ablation of neoplastic Barrett's esophagus.Endoscopy. 2012; 44: 1105-1113Crossref PubMed Google Scholar At this time on the basis of a systematic review of the literature, we believe that ESD or EMR can be successfully applied to submucosal invasive cancers that have a low risk of metastatic potential. Overall, the risk of metastasis is higher in submucosal SCC in comparison with that of submucosal adenocarcinoma. Staging of these tumors should be conducted with EUS, although the accuracy of EUS in discerning the depth of invasion is limited. Additional treatment should be determined by factors such as tumor grade, status of lymphovascular invasion, and depth of tumor, which have a direct influence on metastatic potential. In those patients with submucosal invasion and risk factors, adjuvant chemotherapy and radiation may mitigate metastatic potential. Future research should focus on novel biological and immunohistochemistry markers that can aid in the prediction of tumor behavior and LNM in T1b esophageal cancer .Table 1Predictors of Higher Risk of Lymph Node Metastasis in Esophageal CancerFactorsTumor sizeTumors larger than 2 cmMorphologyParis type 0-I protruded lesionsParis type 0-III excavated lesionsNarrow-band imagingType B 3 IPCL pattern in ESCCAvascular areaHistologyPoorly differentiated tumorsLVI on resected specimenDepth of invasionESCC tumors with submucosal invasion >200 μmEAC tumors with submucosal invasion >500 μmBiological markersE-cadherinCyclin D1EAC, esophageal adenocarcinoma; ESCC, esophageal squamous cell carcinoma; IPCL, intrapapillary capillary loop; LVI, lymphovascular invasion. Open table in a new tab Table 2Risk of Lymph Node Metastasis in Submucosal Esophageal Cancer Based on Cancer Type19Gockel I. Sgourakis G. Lyros O. et al.Risk of lymph node metastasis in submucosal esophageal cancer: a review of surgically resected patients.Expert Rev Gastroenterol Hepatol. 2011; 5: 371-384Crossref PubMed Scopus (67) Google ScholarDepth of invasionESCC (%)EAC (%)SM1276SM23623SM35558EAC, esophageal adenocarcinoma; ESCC, esophageal squamous cell carcinoma; SM, submucosa. Open table in a new tab EAC, esophageal adenocarcinoma; ESCC, esophageal squamous cell carcinoma; IPCL, intrapapillary capillary loop; LVI, lymphovascular invasion. EAC, esophageal adenocarcinoma; ESCC, esophageal squamous cell carcinoma; SM, submucosa. This expert review was commissioned and approved by the AGA Institute Clinical Practice Updates Committee (CPUC) and the AGA Governing Board to provide timely guidance on a topic of high clinical importance to the AGA membership, and underwent internal peer review by the CPUC and external peer review through standard procedures of Clinical Gastroenterology and Hepatology.