Examining the Issues in Barrett's Esophagus: An Expert Interview With Prateek Sharma, MD

Editor's Note:

Barrett's esophagus is a metaplastic lesion that is generally confined to the lower region of the esophagus, and which considerably increases the risk for esophageal adenocarcinoma. The risk for progression to malignancy has been estimated at 0.5% to 1% per year. The strongest associated risk factor for the development of this precancerous lesion is long-standing gastroesophageal reflux disease. Frequent exposure to such caustic refluxate erodes the regular squamous epithelium, which may subsequently be replaced with the specialized intestinal metaplasia known as Barrett's esophagus. Medscape spoke with Prateek Sharma, MD, Professor of Medicine, University of Kansas School of Medicine, Kansas City, Missouri, to explore the latest advances in our understanding of this clinically important condition, with a focus on the implications for clinical practice, as framed by data presented at Digestive Disease Week (DDW) 2008.

Medscape: The American College of Gastroenterology (ACG) recently issued updated guidelines on the diagnosis, surveillance, and treatment of Barrett's esophagus.^[1] What are the significant changes in these guidelines since they were last updated in 2002, and how will these new recommendations translate into clinical practice?

Dr. Sharma: The significant changes in the ACG updated guidelines pertain to approaches to patients with dysplasia in Barrett's esophagus. The diagnosis of low-grade dysplasia poses a significant challenge to clinicians because this diagnosis drives the frequency of surveillance endoscopy. On the other hand, the majority of patients with low-grade dysplasia have a rather benign outcome, with regression to nondysplastic tissue on subsequent examinations. Given this transient nature of low-grade dysplasia, the updated guidelines recommend that this diagnosis be confirmed by an expert gastrointestinal pathologist. For high-grade dysplasia and intramucosal cancer, the updated guidelines emphasize the role of endoscopic therapies as a treatment modality. Although esophagectomy has been considered as the treatment of choice for patients with intramucosal cancer in Barrett's esophagus, recent emerging data support the use of endoscopic therapies as a very viable alternative in this group of patients. Given the high morbidity and mortality associated with esophagectomy, the clinical utility of endoscopic therapies in patients with high-grade dysplasia and intramucosal adenocarcinoma will continue to increase. However, endoscopic ablation is not recommended in patients with nondysplastic Barrett's esophagus or even in those with low-grade dysplasia given the low risk for cancer development in this patient population.

Medscape: Adherence to guidelines for endoscopic surveillance in Barrett's esophagus in the community is suboptimal. During this year's DDW meeting, investigators conducted an analysis using a large community-based pathology database to identify patient-related factors predictive of adequate biopsy sampling in patients with Barrett's esophagus.^[2] What can you tell us about this study and what were the key findings?

Dr. Sharma: The current guidelines and standard practice in patients with Barrett's esophagus involve obtaining target biopsies from any abnormalities within the columnar-lined segment followed by 4-quadrant biopsies every 2 cm. In fact, a recent study from the United Kingdom^[3] showed that such a biopsy protocol was significantly more likely to detect dysplasia compared with random biopsies in patients with Barrett's esophagus. In the present study, Abrams and colleagues^[2] conducted an analysis using a large community-based pathology database to assess whether adequate biopsy sampling was performed in the community and the factors associated with sampling. They defined "adequate sampling" as ≥ 2 biopsy specimens per centimeter of Barrett's esophagus; data were obtained from a total of 1386 surveillance cases. The mean age of the patients undergoing surveillance endoscopy for Barrett's esophagus was 63 years and the majority were men (68%) with ≥ 3 cm of columnar-lined esophagus (51%). The study authors determined that only 48% of these patients undergoing surveillance had adequate sampling of their Barrett's segment, and longer segments and older age were associated with such inadequate sampling. This study highlights the need for better sampling in patients with Barrett's esophagus and for identifying potential barriers that may impede adherence to practice guidelines.

Medscape: Chronic exposure of the esophagus to refluxate can erode the esophageal mucosa. Mucosal injury in Barrett's esophagus depends on the pH of the refluxate as well on the duration of contact with the esophageal mucosa, and prolonged exposure can promote inflammatory cell infiltrate. T cells trigger mucosal inflammation. In this context, Berndt and colleagues^[4] presented the results of a study that investigated changes in T-cell-related mucosal combinatorial molecular protein patterns (CMPs) in biopsies taken from Barrett's esophagus, esophageal adenocarcinoma, and normal control tissue. Can you briefly discuss the key findings of this study, with a view toward the implications for clinical practice?

Dr. Sharma: Barrett's esophagus, a complication of gastroesophageal reflux disease, is associated with chronic inflammation, but why only a minority (approximately 10%) of individuals with reflux disease develop this premalignant condition is not clearly understood. Because mucosal inflammation is triggered by T cells, it may be helpful to evaluate changes in T-cell-related mucosal CMPs between controls and patients with Barrett's esophagus and esophageal adenocarcinoma. Using a unique robotic whole-cell imaging technology that allows the integration of cell biology and mathematical tools to simultaneously visualize multiple proteins and correlate cellular location of proteins with function, Berndt and colleagues^[4] studied biopsy samples from controls and patients with Barrett's esophagus and esophageal cancer. Using a search depth to 5 antibody combinations, these investigators showed for the first time that control and Barrett's tissue could be differentiated by 63, controls and cancer by 3227, and Barrett's esophagus and cancer could be differentiated by 1521 distinct protein combinations. Thus, using proteomic analysis, it was shown that CMPs were distinct in patients with Barrett's esophagus and esophageal adenocarcinoma, whereas some were comparably changed in both groups of patients, suggesting that proteomics may help improve our understanding of the various pathogenic mechanisms involved in Barrett's esophagus and its progression to cancer.

Medscape: Endoscopic therapy represents a potential alternative option to surgical therapy for patients with low-risk early esophageal adenocarcinoma (mucosal [T1a] disease); however, there are limited data on the long-term outcomes of patients treated endoscopically. During DDW 2008, Prasad and colleagues^[5] reported their findings on the long-term survival of cohorts of patients with mucosal esophageal carcinoma treated endoscopically and surgically. Can you briefly discuss these findings and their clinical relevance?

Dr. Sharma: Surgical therapy has been considered the gold standard for the treatment of patients with resectable esophageal adenocarcinoma. However, given the significant mortality and morbidity associated with esophagectomy and the older age and comorbidities present in such patients, endoscopic therapies have been increasingly used to treat mucosal esophageal cancer. To determine the long-term survival of patients with mucosal esophageal adenocarcinoma treated with either surgical or endoscopic therapy, Prasad and colleagues^[5] performed a retrospective analysis of such patients treated between 1995 and 2007. They identified 172 patients treated with endoscopic and 50 who had undergone surgery during this time period. The mean follow-up of patients in these 2 groups was 42 and 76 months, respectively. Although patients in the endoscopic group were significantly older compared with the surgical group (mean age, 71 vs 66 years; P = .015), there was no significant difference in the overall survival between the 2 groups (cumulative mortality, 14.5% vs 14%; P = .92). Overall survival in these patients with mucosal adenocarcinoma was significantly associated with increasing age, but not treatment modality. These data provide reassurance that the overall mortality and long-term survival in Barrett's esophagus patients undergoing endoscopic therapy for mucosal adenocarcinoma are comparable to esophagectomy.

Medscape: Few studies have compared the esophagectomy findings in Barrett's esophagus patients with high-grade dysplasia vs intramucosal (T1a) carcinoma. In a study presented during DDW 2008, investigators addressed the question of whether the rate of submucosal invasion at esophagectomy in such patients could help guide selection of endoscopic vs surgical strategies of management.^[6] What can you tell us about these study findings, and what are the potential implications?

Dr. Sharma: It has been suggested in previously published surgical series that up to 40% of Barrett's esophagus patients with high-grade dysplasia undergoing esophagectomy have unsuspected invasive adenocarcinoma in the resected specimen; this has been one of the major reasons to advocate surgery in this patient group. With the advent of endoscopic therapies and their increasing use in patients with high-grade dysplasia and intramucosal cancer, it is imperative to define the true prevalence of invasive cancer in these patients. Wang and colleagues^[6] reviewed records of 60 patients (41 with high-grade dysplasia and 19 with intramucosal cancer) who had undergone esophagectomy. Only 4 patients (6.7%) had invasive adenocarcinoma in the resected specimen (ie, submucosal invasion with the potential for lymph node metastasis), and all of these patients had nodular Barrett's mucosa at the time of preoperative endoscopy. Of the remaining patients, 45% had intramucosal cancer; 45% had high-grade dysplasia; and 3.3% had nondysplastic Barrett's esophagus -- and there were no differences between these patient groups with respect to their preoperative findings. This study reports a low prevalence (6.7%) of unsuspected invasive cancer in patients with high-grade dysplasia and/or intramucosal cancer undergoing esophagectomy and emphasizes the importance of a meticulous examination and biopsy protocol during upper endoscopy in these patients.