It was a pleasure to receive a letter from Dr. Michael Zinner inviting us to write a prospective on the chapters from Drs. Daniel Tong and Simon Law from Hong Kong on cancer of the esophagus and Drs. Jon Wee, Shelby Steward, and Raphael Bueno from The Brigham and Women’s Hospital, Harvard Medical School, on surgical procedures to resect and replace the esophagus. We begin by stating both chapters are exceedingly well written, informative, and enlightening. Our reading identified areas where we were moved to comment based on our personal experience. The comments usually take the form of additional thoughts or alternatives. Occasionally, we raise a note of caution or take a controversial point of view.

The specific reason for the dramatic increase in adenocarcinoma of the esophagus continues to remain a mystery, even though it represents the largest epidemiologic change ever recorded for a solid cancer.¹ Reasons proposed for the significant and continuous increase in incidence of esophageal adenocarcinoma are the epidemics of obesity and gastroesophageal reflux disease (GERD).² The latter can result in the complication of Barrett esophagus and its subsequent development into high-grade dysplasia. As Tong and Law point out, high-grade dysplasia is the last preinvasive stage in the metaplasia–dysplasia–cancer sequence and has become the focus of efforts to combat the problem. Intensive surveillance of patients with Barrett esophagus is proposed as a strategy to identify high-grade dysplasia prior to the evolvement of cancer and treat it with endoscopic mucosal resection or ablation. To date, the results of intensive surveillance have been disappointing, and the work involved in ablation and its required unlimited follow-up is extremely expensive. Further, Tong and Law indicate a 20% recurrence rate after an average follow-up of 2.4 years, and the median length of Barrett mucosa in which the recurrent lesion occurred was only 0.6 cm, which is significantly shorter and more difficult to identify than the pretreatment length.

Disappointment with intensive surveillance as a solution to the Barrett esophagus problem has stimulated an interest in the possibility of surgically preventing the development of Barrett esophagus. The progression of GERD to Barrett esophagus in patients who are receiving treatment with proton pump inhibitors (PPIs) has been reported by several investigators.^3-5 This has led to concerns that PPI therapy does not address all aspects of the disease and patients who are at risk of progression need to be identified early in the course of their disease in order to prevent the development of visible Barrett esophagus. Recently, it has been reported that biopsies of the squamocolumnar junction that show microscopic intestinalized metaplastic cardiac mucosa in endoscopically normal patients with GERD are predictive of the evolvement of visible Barrett esophagus.⁶

A further indicator of progression toward visible Barrett esophagus is a permanent structural alteration of the lower esophageal sphincter (LES) on manometry, resulting in uncontrollable reflux.^5,7 Acid suppression therapy is notably effective in patients who have a normal LES but less so in those who have a structurally defective LES. Permanent structural alterations of the LES are difficult to correct without surgical intervention. The likelihood of symptom control and prevention of progression to Barrett esophagus in patients with persistent symptoms on PPI therapy is greater if surgical correction of a compromised LES is carried out earlier rather than later.⁷

A proposed solution to the rising incidence of esophageal adenocarcinoma is to apply the principle: Where there is no Barrett esophagus, there is no cancer. A competent Nissen fundoplication has been shown to prevent Barrett esophagus if performed before it develops.⁸ Early fundoplication in patients with microscopic intestinal metaplasia just below the squamocolumnar junction, in the absence of endoscopically visible Barrett esophagus, results in complete regression of the intestinal metaplasia and no evidence of visible Barrett esophagus in 74% of patients.⁹ In a separate control group of patients with endoscopically visible Barrett esophagus, only 4% of patients showed regression of the Barrett esophagus after fundoplication.⁹

Despite the ability of the fundoplication to induce regression of microscopic intestinal metaplasia and prevent development of visible Barrett esophagus, the side effects of the procedure (ie, dysphagia, bloating, and the inability to burp and vomit) have discouraged the use of this approach to avert the evolvement of the precancerous lesion. Over the past decade, minimally invasive outpatient LES augmentation procedures have been developed. These procedures avoid the side effects associated with Nissen fundoplication and may therefore be appropriate for early surgical intervention.¹⁰ The effectiveness of these procedures to induce regression of microscopic intestinal metaplasia at the squamocolumnar junction and avert the development of Barrett esophagus and a subsequent adenocarcinoma should be investigated in future clinical studies.

The burden of esophageal adenocarcinoma is predicted to continue to rise dramatically in high-income countries.¹¹ It is incumbent on both gastroenterologists and surgeons to work together to stop the rising incidence of Barrett esophagus, the premalignant lesion of esophageal adenocarcinoma. By 2030, it is predicted that 1 out of every 100 European men will be diagnosed with esophageal adenocarcinoma before the age of 75.¹¹

Tong and Law have rightly pointed out that “accurate staging allows stage-directed therapies and quality control for clinical trials.” They could not be more correct; however, there is an emerging problem. Clinical studies based on clinical staging have up to an 84% error rate for specific staging of a patient¹² and a 24% error rate for staging a patient above or below a specified stage.¹³ Such a degree of error could affect proper randomization between the arms of a clinical study if the size of the study population is insufficient. As a consequence, the results of randomized studies between different treatment regimens that are based on clinical staging become suspect.

Superficial esophageal adenocarcinomas, particularly T1a lesions, are nearly always staged correctly and cured by endoscopic mucosal resection. Similarly, advanced locoregional disease (cT2-3N2-3 in the updated staging system) is usually staged correctly, and although neoadjuvant therapy has been shown to be of benefit, cure of this advanced disease is rare. The problem with clinical staging arises in patients with limited locoregional disease (cT2-3N0-1 in the updated staging system) who fall between these 2 extremes. The accuracy of staging cT2-3N0-1 disease correctly by clinical staging is between 13% and 25%.^12,13 In such patients, the assessment of nodal metastases remains problematic and is culpable for the staging error.

Complicating things further, it is now recognized that the number of involved lymph nodes is a critical factor in survival.¹⁴ The updated seventh edition of the American Joint Committee on Cancer (AJCC) staging system has N0 (no positive nodes), N1 (1-2 positive nodes), N2 (3-6 involved nodes), and N3 (7 or more involved nodes) categories. Currently, there is little information on the accuracy of clinical N staging using the updated staging system. Treatment algorithms would be simplified and clinical staging would be of little significance if neoadjuvant therapy was beneficial for all resected patients other than those with superficial tumors amenable to treatment by endoscopic resection alone. However, multiple studies have failed to show a benefit for neoadjuvant therapy in patients with limited locoregional disease.^12,15-18

A recent study by Worrell et al reported the accuracy of clinical staging and survival of patients treated with primary esophagectomy for limited locoregional disease defined as cT1-3 and N0-1 (in the updated staging system, disease limited to 0-2 involved nodes).¹³ Using endoscopic resection, endoscopic ultrasound, and/or computed tomography/positron emission tomography scan, the final pathology confirmed accurate clinical staging (≤T3N1) in 76% of patients with limited locoregional disease. The remaining 24% of patients were understaged (>T3N1), and in all cases, understaging was based on advanced nodal (N2 or N3) disease. Overall, 65% of the patients were node negative (T1-2N0), and the survival rate in these patients was 82% at 5 years and would likely not have been improved with induction therapy. These results compare favorably to the 64% 5-year survival rate for similarly clinically staged patients who received neoadjuvant chemoradiotherapy before esophagectomy in a series reported from the MD Anderson Cancer Center.¹⁹ This comparison and other reported experiences are providing accumulating evidence that neoadjuvant therapy is not beneficial in node-negative patients.