Lorna J. Dunn

Consensus Statements for Management of Barrett's Dysplasia and Early-Stage Esophageal Adenocarcinoma, Based on a Delphi Process

BACKGROUND & AIMS Esophageal adenocarcinoma (EA) is increasingly common among patients with Barretts esophagus (BE). We aimed to provide consensus recommendations based on the medical literature that clinicians could use to manage patients with BE and low-grade dysplasia, high-grade dysplasia (HGD), or early-stage EA. METHODS We performed an international, multidisciplinary, systematic, evidence-based review of different management strategies for patients with BE and dysplasia or early-stage EA. We used a Delphi process to develop consensus statements. The results of literature searches were screened using a unique, interactive, Web-based data-sifting platform; we used 11,904 papers to inform the choice of statements selected. An a priori threshold of 80% agreement was used to establish consensus for each statement. RESULTS Eighty-one of the 91 statements achieved consensus despite generally low quality of evidence, including 8 clinical statements: (1) specimens from endoscopic resection are better than biopsies for staging lesions, (2) it is important to carefully map the size of the dysplastic areas, (3) patients that receive ablative or surgical therapy require endoscopic follow-up, (4) high-resolution endoscopy is necessary for accurate diagnosis, (5) endoscopic therapy for HGD is preferred to surveillance, (6) endoscopic therapy for HGD is preferred to surgery, (7) the combination of endoscopic resection and radiofrequency ablation is the most effective therapy, and (8) after endoscopic removal of lesions from patients with HGD, all areas of BE should be ablated. CONCLUSIONS We developed a data-sifting platform and used the Delphi process to create evidence-based consensus statements for the management of patients with BE and early-stage EA. This approach identified important clinical features of the diseases and areas for future studies.

Barrett's Adenocarcinoma 52 Years After Subtotal Esophagectomy for Pediatric Peptic Stricture

Barretts esophagus results from the long-term effects of both acid and bile reflux. After subtotal esophagectomy and reconstruction with a gastric tube, many patients experience profound reflux. Development of Barretts epithelium in the esophageal remnant has been reported. Here we report the case of a man who was diagnosed with adenocarcinoma in his esophageal remnant on a background of Barretts change 52 years after undergoing one of the first esophageal resections for benign disease as a child.

Columnar Metaplasia in the Esophageal Remnant After Esophagectomy A Common Occurrence and a Valuable Insight Into the Development of Barrett Esophagus

Objective: This study aimed to establish the incidence of postesophagectomy columnar metaplasia and dysplasia, and the timescale over which it develops. It also aimed to assess if this epithelium is molecularly similar to sporadic Barrett esophagus, thereby confirming suitability as a research model. Background: Metaplasia in the esophageal remnant after esophagectomy is well described, but incidence and the potential for dysplasia are uncertain, and the clinical relevance unclear. Although proposed as a model for Barrett esophagus, no large studies have examined the molecular phenotype of postesophagectomy metaplasia. Methods: Patients underwent prospective endoscopic evaluation having previously undergone esophagectomy. The macroscopic appearance of the esophageal remnant was noted and biopsies taken. Specimens were stained using hematoxylin and eosin and by immunohistochemistry for cytokeratins 7 and 20, and Chromogranin A—proteins which have a well described expression pattern in sporadic Barrett esophagus. Results: Of the 126 eligible patients, 45 (36%) had evidence of metaplasia. There were no cases of dysplasia. Nonintestinalized columnar epithelium occurred earlier than specialized intestinal metaplasia (median 4.8 vs 8.1 yr; P = 0.025). Thirty-seven samples underwent immunohistochemical analysis. A classic cytokeratin 7/20 staining pattern was present in 23 cases (62%), within the prevalence range reported for sporadic Barrett. Conclusions: Columnar metaplasia is common following esophagectomy, but the absence of dysplasia in this large cohort is reassuring. Presence of specialized intestinal metaplasia is associated with increased time from surgery, suggesting this represents later disease. Immunohistochemistry staining is similar to sporadic Barrett, suggesting that this group of patients represent an accurate human model for the development of Barrett.

Randomized clinical trial of laparoscopic total (Nissen) versus posterior partial (Toupet) fundoplication for gastro-oesophageal reflux disease based on preoperative oesophageal manometry (Br J Surg 2008; 95: 57–63)

The Editors welcome topical correspondence from readers relating to articles published in the Journal. Responses should be sent electronically via the BJS website (www.bjs.co.uk). All letters will be reviewed and,if approved,appear on the website. A selection of these will be edited and published in the Journal. Letters must be no more than 250 words in length. Copyright

Chemoprevention of gastrointestinal cancer.

Chemoprevention of gastrointestinal cancer uses natural or synthetic substances to reduce the risk of developing malignancy. Gastrointestinal cancers are a significant cause of morbidity and mortality throughout the world. Cancer Research UK’s latest figures (2004) reveal that oesophageal, gastric and bowel cancers together make up 18 per cent of new cancer diagnoses. The prospect of prevention is clearly appealing, especially for oesophageal cancer, which still carries a dismal 8 per cent 5-year survival rate. The ideal chemopreventive agent is effective, well tolerated and affordable. Individuals at very high risk may be willing to tolerate treatment with significant side-effects, but this is not acceptable in a population with a normal risk profile. Although the concept of chemoprevention has been around for over 30 years, progress has been slow for several reasons. More than 200 agents have been considered, yet none is officially licensed in the UK for gastrointestinal cancer1. The field lags far behind other areas, such as cardiology, where the emphasis has long since shifted from treating endstage events to risk factor modification. Trial design presents a problem. Cancer incidence is the ideal endpoint, but demonstrating a reduction in this variable would require large, long-term trials, which are not feasible logistically. Surrogate endpoints include biomarkers – phenotypic or genotypic characteristics that are altered during carcinogenesis. A widely used surrogate is colonic adenoma for colorectal cancer. The adenoma–carcinoma sequence is widely accepted, but potential problems remain. For instance, it may be that a chemopreventive agent exerts its effect after adenoma formation. The development of cancer is a complex process and, as a consequence, there are multiple targets in the carcinogenesis pathway for chemoprevention. Agents must have an established safety profile, and so far this has restricted research to existing drugs or nutritional supplements. Large prospective cohort studies may provide epidemiological evidence that a drug is associated with a decreased relative risk of developing a particular cancer. This can provide the basis for prospective trials. In the long term, greater understanding of the molecular mechanisms underlying neoplasia may allow the development of more specifically targeted chemopreventive agents. The most widely studied agents for chemoprevention are non-steroidal anti-inflammatory drugs (NSAIDs) and aspirin. Overexpression of cyclooxygenase (COX) 2 is a feature of both oesophageal and colorectal carcinomas, indicating a potential mechanism of action for the protective effects of these drugs. A randomized controlled trial has shown that 325 mg aspirin daily in 635 patients with a past history of colorectal cancer significantly decreased the risk of developing a new adenoma, compared with placebo2. Although a wealth of epidemiological data supports the use of aspirin, other randomized controlled trials have failed to confirm a benefit3. A landmark study by Giardiello and colleagues4 randomized 22 patients with familial adenomatous polyposis (FAP) to sulindac 150 mg (a non-selective NSAID) twice daily or placebo. Sulindac was associated with a significant decrease in both the number and the diameter of polyps. Subsequent studies have shown that both non-selective NSAIDs and COX-2 inhibitors reduce the incidence of colorectal adenoma and colorectal cancer. Celecoxib is now licensed in the USA for the reduction of polyp numbers in FAP, in conjunction with endoscopic surveillance or surgery. However, the side-effects of NSAIDs limit their value for chemoprevention in the average-risk population. NSAIDs have also been proposed as agents capable of preventing the progression of Barrett’s oesophagus5, but a recent randomized placebocontrolled study failed to show any benefit for celecoxib 200 mg twice daily over placebo6. Several COX-2 inhibitor trials have been disrupted because of concern over the cardiovascular risk profiles of the drugs, and it seems likely that such concern will limit their use in chemoprevention. Many pharmaceutical agents have moderately strong epidemiological evidence to support their use in chemoprevention. These include statins, ursodeoxycholic acid and hormone replacement therapy. Promising new drugs, including tyrosine

Stage-for-stage comparison of definitive chemoradiotherapy, surgery alone and neoadjuvant chemotherapy for oesophageal carcinoma (Br J Surg 2009; 96: 1300-1307).

Sir I read with interest the article by Neudecker and colleagues, which described a prospective randomized trial comparing laparoscopic and open surgery for colorectal cancer. The authors concluded that laparoscopic colorectal cancer surgery was associated with increased operating time but did not decrease morbidity even in a moderate-risk population. Besides the series supporting this conclusion, I still found some large series with pros and cons1 – 4. Nevertheless, I have several questions regarding this manuscript. In Table 4, it is noted that 13 patients (5·2 per cent) receiving a laparoscopic approach developed small bowel obstruction and seven (2·8 per cent) were treated surgically. Reviewing the other series1 – 4, I found that this problem was seldom mentioned. However, causes of this relatively rare complication were not discussed. Surgeons might gain from this to avoid the same complication. The authors mentioned that a body mass index of more than 26 kg/m2 increased the risk of general morbidity by 2·5-fold, but the opposite result was recorded in Table 8. Finally, the authors concluded that laparoscopic colorectal cancer surgery did not decrease morbidity even in a moderate-risk population. However their criteria for designating ‘moderate risk’ were not clearly defined. They did not analyse the morbidities of laparoscopic and open approaches based on individual American Society of Anesthesiologists grades. Hence, this conclusion seemed a little debatable. C.-C. Chiu Department of General Surgery, Chi Mei Medical Centre, Tainan County, Taiwan (e-mail: chiuchongchi@yahoo.com.tw) DOI: 10.1002/bjs.7064

Treatment of oesophageal anastomotic leaks by temporary stenting with self-expanding plastic stents (Br J Surg 2009; 96: 887-891).

The Editors welcome topical correspondence from readers relating to articles published in the Journal. Responses should be sent electronically via the BJS website (www.bjs.co.uk). All letters will be reviewed and, if approved, appear on the website. A selection of these will be edited and published in the Journal. Letters must be no more than 250 words in length. Copyright