| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
EDITORIALS |
From the Department of Surgery, University of Southern California, Los Angeles, California.
Correspondence: Address correspondence to: Jeffrey H. Peters, MD, Professor of Surgery, University of Southern California, 1510 San Pablo St., Los Angeles, CA 90033; Fax: 323-442-5833; E-mail: jhpeters{at}surgery.usc.edu
Aside from its prevention, the most profound impact a medical professional may have on cancer is to detect it early. Diagnostic technology has advanced to the point where many solid tumors, including breast, prostate, colon, and stomach, can be detected at such an early stage that, somewhat paradoxically, management decisions are difficult and controversial. Esophageal cancer can be added to this list. Its detection at a premalignant or early stage, once a decidedly rare clinical occurrence, is now common. Thirty percent of patients with Barrett’s esophagus undergoing resection at the University of Southern California have high-grade dysplasia and/or an early-stage curable lesion.1 The clear link between Barrett’s esophagus and esophageal adenocarcinoma, the liberal use of flexible endoscopy, and the widespread adoption of surveillance programs have all contributed to this fact. The article by Al-kasspooles et al.2 appearing in this issue of Annals of Surgical Oncology outlines four key issues in the management of patients with highly dysplastic Barrett’s esophagus. Do we know enough about the natural history of high-grade dysplasia to make rational treatment decisions? Is esophagectomy too "radical" of a procedure given such an early lesion? If so, is there a role for nonsurgical ablative therapies? Finally, can we reliably detect cancer among dysplastic cells? Given the rising prevalence of Barrett’s esophagus, an answer to these questions becomes more important than ever.
Al-kasspooles et al.2 correctly state that there is much we do not know about the natural history of high-grade dysplasia, although they underemphasize what we have learned from several recent studies. Large cohorts of patients with high-grade dysplasia have now been prospectively followed at The University of Washington3,4 and the University of Kansas5 and retrospectively reviewed at the Hines Veterans Hospital in Chicago.6 These data clearly show that cancer will be identified (identified is a more appropriate term than develop, as many of these patients may have had carcinoma for some time before its detection) in approximately 25% of patients at 1.5 years,3 50% at 3 years,5 and up to 80% 8 years later.4 The 80% figure should be interpreted in light of the fact that there is a 20% or so error rate in the pathologic diagnosis of high-grade dysplasia.7 Thus the natural history of high-grade dysplasia is becoming clear. Most patients will have an invasive adenocarcinoma identified during a 5- to 10-year surveillance period, although a significant minority may not. Although far from perfect, these facts, particularly when taken in association with p53 and cell cycle (flow cytometry) abnormalities, give the clinician significant information on which to base clinical decisions.
The authors also review the morbidity associated with esophagectomy. Without question, removing the esophagus is a major undertaking that is often fraught with significant morbidity and mortality. What is often underestimated is the intensity of resources and emotional burden associated with the decision to pursue surveillance every 3 months. When given the option, many patients prefer to eliminate the possibility of developing esophageal adenocarcinoma, even if esophagectomy is the price to do so. Our challenge is to improve the state of the art such that this can be accomplished with as little morbidity as possible. After esophagectomy, average mortality has steadily decreased over the past two to three decades from more than 25% to 2% to 4% in most centers. It approaches zero in large series of resection for benign disease8 (in which patients with high-grade dysplasia can be included) and in units that have specifically focused on preventing death from esophageal resection, such as that in Hong Kong.9 That being said, as the authors aptly point out, it is arguably the most sensitive surgical procedure to volume-outcome relationships. With fewer than 4000 esophagectomies per year being performed in the United States, the time has come to limit esophageal resection to high-volume centers.
Underscoring these points, a decision analysis study, testing whether esophagectomy or continued surveillance is the optimal treatment for patients with high-grade dysplasia, was recently reported in abstract form.10 Seven strategies were tested, the first being immediate esophagectomy and the remaining six, surveillance for 3, 6, 12, 18, and 24 months and esophagectomy if cancer was identified, and finally, no cancer ever identified. The simulation continued until all patients died from cancer or other causes. A 5-year estimate of the development of cancer in high-grade dysplasia of 20% to 50% was used (quite reasonable, considering the data presented above) and included operative mortality and both short- and long-term disability associated with esophagotomy. Immediate esophagectomy was the preferred treatment for all values of cancer risk anywhere from 10% to 50%! Further immediate esophagectomy had the greatest gain in quality-adjusted life-years. Esophagectomy remained the preferred treatment unless the cancer incidence fell below 3% at 5 years, the operative mortality rose to above 64%, or the quality-adjusted life-years after esophagectomy declined to <0.5 (zero dead, one normal). These rather surprising data lend further credence to the decision for esophagectomy in patients with high-grade dysplasia.
Although efforts at effective ablation of dysplastic Barrett’s esophagus have been ongoing for more than a decade, major obstacles remain. Ablation of large segments of Barrett’s epithelium is compromised by the fact that residual Barrett’s remains in as much as half of the patients and that 25% to 30% will develop severe complications such as stricture or motility disturbances. Effective ablation of small areas of dysplasia, whether by mucosal resection, thermal, or photodynamic energy, requires accurate localization. Localization of a nonvisible area containing high-grade dysplasia is presently not possible, although technologies looming on the horizon such as optical coherence tomography may make this a clinical reality.11 Finally, investigators at the Mayo Clinic in Rochester have found that despite the histologic absence of dysplasia after ablation, genetic abnormalities characterizing a premalignant epithelium remain12! Ablation is still an elusive goal.
Well-designed prospective studies are sorely needed to answer many of the questions outlined by this review. It has often struck me that the state of knowledge and management of premalignant and malignant esophageal disease lags far behind that of the breast and colon. Research efforts to eliminate this gap are well overdue.
Received for publication February 1, 2002. Accepted for publication February 5, 2002.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
