- Email: [email protected]
Treatment of Barrett's esophagus with early neoplasia: a comparison of endoscopic therapy and esophagectomy
ORIGINAL ARTICLE: Clinical Endoscopy
Treatment of Barrett’s esophagus with early neoplasia: a comparison of endoscopic therapy and esophagectomy Drew B. Schembre, MD, Jasmine L. Huang, MD, Otto S. Lin, MD, MSc, Nico Cantone, BS, Donald E. Low, MD Seattle, Washington, USA
Background: Endoscopic therapies for early neoplasia in Barrett’s esophagus may be a viable alternative to esophagectomy. Objective: Our purpose was to compare endoscopic therapy and esophagectomy. Design: Retrospective review from a single institution. Setting: A medium-sized tertiary referral center. Patients and Interventions: All patients with Barrett’s esophagus and dysplasia or intramucosal carcinoma treated by photodynamic therapy (PDT), EMR, or argon plasma coagulation (APC) or esophagectomy with curative intent from May 1998 until November 2005. Main Outcome Measurements: Survival, progression to cancer, eradication of dysplasia and Barrett’s esophagus, major and minor complications, and costs were compared. Results: Sixty-two patients who underwent endoscopic therapy (2 APC alone, 18 EMR þ APC, 20 PDT þ APC, and 22 EMR þ PDT þ APC) and 32 patients who underwent esophagectomy met the inclusion criteria. The 30-day mortality rate included 1 patient in the endotherapy group (2%) and none in the surgical group (P Z.49). No deaths from esophageal cancer occurred in either group. Cancer developed in 6% of endotherapy patients and in none in the surgical cohort (P!.05). Major and minor complications occurred in 8% and 31% of endotherapy patients, respectively, and 13% and 63% of surgery patients (P Z.50, P!.001). Median cost to date was $40,079 for endotherapy and $66,060 for esophagectomy (P!.001). Limitations: Retrospective study, relatively short follow-up, small numbers. Conclusions: Both endotherapy and esophagectomy can effectively treat high-grade dysplasia and intramucosal carcinoma associated with Barrett’s esophagus. Endotherapy is associated with a higher risk of tumor progression, although this is uncommon. Esophagectomy incurs higher initial costs and results in more frequent minor complications but is usually curative. (Gastrointest Endosc 2008;67:595-601.)
Copyright ª 2008 by the American Society for Gastrointestinal Endoscopy 0016-5107/$32.00 doi:10.1016/j.gie.2007.08.042
programs.2 Once HGD has been found, management has included close endoscopic surveillance, esophagectomy, and more recently, endoscopic ablative therapies (ET). Endoscopic therapies have included photodynamic therapy (PDT), EMR, endoscopic submucosal dissection (ESMD), thermal ablation, radiofrequency ablation, and cryotherapy.3-7 Although debate exists whether surveillance or treatment of HGD is more appropriate, surgical esophagectomy remains the standard intervention for fit patients with the disorder in most parts of the United States. However, esophagectomy for esophageal cancer has been associated with substantial morbidity and mortality rates, especially at low-volume centers.8 The risk of
www.giejournal.org
Volume 67, No. 4 : 2008 GASTROINTESTINAL ENDOSCOPY 595
The majority of adenocarcinomas of the esophagus arise within Barrett’s esophagus (BE).1 Identification of the immediate histologic precursor to cancer, high-grade dysplasia (HGD), is the end point of current Barrett’s surveillance Abbreviations: APC, argon plasma coagulation; ASA, American Society of Anesthesiologists; BE, Barrett’s esophagus; ESMD, endoscopic submucosal dissection; ET, endoscopic ablative therapies; HGD, high-grade dysplasia; IMC, intramucosal carcinoma; IRB, institutional review board; LGD, low-grade dysplasia; PDT, photodynamic therapy; VMMC, Virginia Mason Medical Center.
Treatment of Barrett’s esophagus with early neoplasia
esophagectomy for HGD appears to be somewhat less, but it is still significant.9 Over the last 10 years, numerous studies have shown that ET can effectively treat HGD and even intramucosal carcinoma (IMC), possibly with lower complication rates than esophagectomy.10-12 Studies directly comparing ET and surgical therapy are limited.13,14 This study compares patients who have undergone either esophagectomy or ET for BE with HGD or IMC at one high-volume center.
Schembre et al
Capsule Summary What is already known on this topic d
What this study adds to our knowledge d
METHODS We performed a retrospective analysis of 2 institutional review board (IRB)–approved prospective databases: an esophageal resection database that has been in place since 1991 and an endoscopic therapy database that has been in place since 1998. To reduce variability between groups, we included only those patients who underwent esophagectomy or ET with curative intent between May 1, 1998, and November 30, 2005. Database information included patient age, sex, weight, and American Society of Anesthesiologists (ASA) classification, endotherapy and surgical data (type of therapy), pathologic findings (degree of dysplasia, complete resection, presence of invasive cancer, length of Barrett’s segment), posttreatment complications, total hospital charges, and long-term survival. For ET patients, the database also included number of endoscopies, number and types of therapies, and histologic diagnosis at most recent biopsy. All appropriate patients received both surgical and gastroenterology consultations before treatment. All pretreatment and posttreatment biopsies were reviewed by one or more pathologists at Virginia Mason Medical Center (VMMC), with equivocal samples reviewed additionally by nationally recognized experts in Barrett’s histology at the University of Washington in Seattle, or the Cleveland Clinic in Cleveland, Ohio. ET included PDT, EMR, argon plasma coagulation (APC), or any combination of the techniques. ET patients underwent follow-up endoscopy with 4-quadrant jumbo biopsies over the entire length of the pretreatment Barrett’s segment of the esophagus at 3-month intervals for 1 year and then at 6-month intervals for 1 year and then yearly if no dysplasia was identified. Patients found to have invasive cancer during initial staging and those who had less than 6 months of follow-up were excluded. From 1998 to 2000, PDT with porfimer sodium (Photofrin, Axcan Pharma, Mont-Saint-Hilaire, Quebec) 2 mg/kg þ 200 j/cm light exposure at 620 nm was the primary treatment for BE with HGD or IMC. From 2000 to 2003, we used EMR for focal nodular disease followed by PDT for any residual Barrett’s tissue. After 2003, EMR was used for large-area BE less than 5 cm with PDT reserved for Barrett’s tissue more than 5 cm in length. APC at 90 watts was used to treat small-area primary or residual disease. All patients underwent initial endoscopy with EUS to help identify invasive disease. The goal 596 GASTROINTESTINAL ENDOSCOPY Volume 67, No. 4 : 2008
Despite its substantial morbidity and mortality rates, especially at low-volume centers, esophagectomy remains the standard intervention for patients with Barrett’s high-grade dysplasia.
d
In a retrospective review of 94 patients, both endoscopic ablation and esophagectomy effectively treated Barrett’s high-grade dysplasia and intramucosal carcinoma. Endotherapy was associated with a higher risk of tumor progression, and esophagectomy incurred higher initial costs and more frequent minor complications.
of ET in every patient was complete eradication of all Barrett’s mucosa. All esophagectomy patients underwent operation by a single surgeon (D.L.). Preoperative investigations in operative candidates included CT scan of the chest, abdomen, and pelvis, along with endoscopy and EUS. All patients over 50 years of age underwent objective cardiac testing. In addition, all patients were reviewed at our institution’s thoracic oncology tumor board. Resectional approach was individualized according to patient physiology and characteristics. The primary end points are the number and percentage of patients surviving, those who are cancer free, and those who are free of HGD. Secondary end points include eradication of all Barrett’s mucosa, frequency and severity of complications, and cost of treatment. Treatment costs are represented by total hospital and outpatient charges related to esophageal disease and subsequent treatment accrued at VMMC from the time of the initial consultation until December 1, 2005. Comparative analysis was done with use of the unpaired Student t test for continuous variables and the c2 test for categorical variables. All tests were 2-tailed. P values !.05 were considered significant. Statistical tests were performed with SPSS 13.0 (SPSS, Chicago, Ill) or Microsoft Excel (Microsoft, Redmond, Wash). All patients were U.S. citizens and survival data were checked against the Social Security Administration Death Index.
RESULTS From 1998 to 2005, 117 patients were referred to our institution for treatment of presumptive BE with HGD or IMC. Twenty-three patients were excluded from analysis because disease discovered at initial examination was T1sm or greater (3), low-grade dysplasia (LGD) (4), or no dysplasia (8) or follow-up has been less than 6 months www.giejournal.org
Schembre et al
Treatment of Barrett’s esophagus with early neoplasia
TABLE 1. Patient characteristics Endotherapy
Esophagectomy
Significance
No. of patients
62
32
Median age (y)
70
64
P Z .003
Percent male
73
93
P Z .007
Average ASA level
2.6
2.5
P Z .62
Average length of Barrett’s segment (cm)
5.1
7.2
P Z .024
TABLE 2. Results of treatment Endotherapy
Esophagectomy
Significance
Progression to cancer
4 (6%)
0
P Z .14
Persistent dysplasia
8 (13%)
1 (3%)
P Z .12
27 (44%)
1 (3%)
P ! .001
89%
93%
P Z .49
Procedure-related mortality
1 (2%)
0
P Z .47
Major complications
5 (8%)
4 (13%)
P Z .50
Minor complications
20 (32%)
26 (66%)
P ! .001
Hospital charges
$40,079
$66,060
P ! .001
Residual Barrett’s segment Overall 4-year adjusted survival rate
(8). Sixty-two patients (48 with HGD and 13 with IMC) underwent ET: 2 with APC alone, 18 with EMR alone, 20 with PDT alone, and 22 with EMR and PDT. All ET patients have had some focal treatment with APC during follow-up examinations. Thirty-two patients (15 with HGD and 17 with IMC) underwent esophagectomy with the following surgical approaches: 4 transhiatal, 10 Ivor-Lewis, and 18 left thoracoabdominal. Additional characteristics of the 2 groups are shown in Table 1. The median follow-up was 20 months for the endotherapy group (range 6 to 84 months) and 48 months for surgical patients (range 6 to 88 months). Two ET patients were lost to endoscopic follow-up after 1 year, 3 patients at 2 years, and 3 patients after 3 years. No surgical patients were lost to follow-up. Survival data are complete on all patients. One patient in the endotherapy group died of sepsis resulting from diverticulitis 11 days after treatment. There was no 30-day mortality in the surgical group (P not significant). Adjusted survival at 4 years was 93% for the surgical group and 89% for the endotherapy group (P Z.49). None of these deaths were due to esophageal cancer. Cancer developed in 4 (6%) of the ET patients. Two cases of intramucosal cancer and 2 cases of invasive cancer developed in patients with a pretreatment diagnosis of HGD. One patient with invasive cancer underwent chemoradiation and one underwent radiation alone followed by esophageal stenting. One patient with IMC is
still undergoing ET, and one elected to undergo esophagectomy. Three additional patients in the ET group opted for surgical resection for persistent HGD after an average of 2 treatment sessions. No distant cancers developed in the surgical cohort (P ! .05), despite identifying previously unrecognized cancer in 8 (25%) of the resected esophagi. One patient who underwent esophagectomy after EMR for IMC had no identifiable dysplasia or BE at surgery. All surgical patients had R0 resections for cancer. One patient with a very long segment of BE had persistent HGD proximal to a high anastamosis and underwent successful ET of the residual area of dysplasia. Median overall hospital charges (to date) were $40,079 for the endotherapy group and $66,060 for the surgical group. Additional results are shown in Table 2. Specific posttreatment complications are presented in Tables 3 and 4. There was no statistically significant difference in the incidence of major complications between the 2 treatment groups. Minor complications were more common in the surgical group (P ! .001). Histologic examination among 44 ET patients with HGD who did not progress to cancer has shown 8 patients with residual focal HGD. These patients are undergoing further therapy. No ET patient among the 13 with intramucosal cancer has shown any residual cancer or HGD on the most recent biopsy. After ET, the average length of
www.giejournal.org
Volume 67, No. 4 : 2008 GASTROINTESTINAL ENDOSCOPY 597
Treatment of Barrett’s esophagus with early neoplasia
Schembre et al
TABLE 3. Complications of endotherapy No. (%) Major complications (8%)* Death
1 (2%)
Bleeding
2 (3%)
Prolonged hospitalization
3 (5%)
Minor complications (31%)* Stricture
13 (21%)
Photosensitivity
4 (6%)
Pneumonia
2 (3%)
*The sum of individual complications may not equal the total complication rate because one individual may have more than one event.
Figure 1. Number of patients with endotherapy histologic features: initial and at most recent follow-up.
TABLE 5. Results by type of endotherapy APC alone
EMR
Progression to cancer
0
1 (6%)
Persistent dysplasia
0
5 (28%) 1 (5%)
TABLE 4. Complications of surgery No. (%) Major complications (13%)* Anastomotic leaky
1 (3%)
Chyle leaky
2 (6%)
Deep venous thrombosis
1 (3%)
Minor complications (63%)*
Residual Barrett’s segment
PDT
EMR D PDT
2 (10%) 1 (5%) 2 (9%)
1 (50%) 10 (56%) 8 (40%) 8 (36%)
Procedure-related mortality
0
0
0
1 (5%)
Major complications
0
1 (6%)
1 (5%)
3 (14%)
0
6 (33%) 5 (25%) 8 (36%)
Pneumonia
2 (6%)
Minor complications
Atrial fibrillation
6 (19%)
Wound infection
3 (9%)
Small numbers in multiple groups and variance in degree of disease at presentation precludes meaningful statistical comparisons among groups.
Stricture Pneumothorax
15 (47%) 1 (3%)
In the United States, the incidence of adenocarcinoma of the esophagus is rising at one of the highest rates.15 The current approach to BE with HGD includes surveillance, esophagectomy, or endoscopic ablation. In theory, surveillance identifies those individuals who progress to cancer early, allowing intervention while it
may still be curative and sparing those who never progress from the risks of therapy. However, invasive cancer may be difficult to recognize endoscopically, and even jumbo biopsy specimens at close intervals may miss established malignancies.16 Further, the rate of progression of HGD to cancer can vary widely, with patients with multifocal HGD and aneuploidy at particularly high risk for malignant progression.17 Moreover, delaying intervention in a younger patient leads to years of surveillance endoscopy and necessitates that the patient endure the anxiety of harboring a precancerous condition. The current study shows that in a tertiary referral center both esophagectomy and endotherapy can be effective and safe methods for treating BE with HGD or IMC. We chose to include both HGD and IMC in our analysis because the 2 histologic types behave similarly, with very low risk of nodal spread. In addition, intraobserver agreement among pathologists differentiating them is poor.18 Esophagectomy has been identified historically as a procedure with high morbidity and mortality rates.
598 GASTROINTESTINAL ENDOSCOPY Volume 67, No. 4 : 2008
www.giejournal.org
*The sum of individual complications may not equal the total complication rate because one individual may have more than one event. yTreated conservatively.
persistent Barrett’s tissue was 1 cm. These results are presented in Figure 1. A breakdown of ET patients by treatment modality is shown in Table 5.
DISCUSSION
Schembre et al
Treatment of Barrett’s esophagus with early neoplasia
Mortality rates of 8% in high-volume centers and 20% in low-volume centers have been reported as recently as 2002.7 International databases such as the British ASCOT Report demonstrates a mortality rate of 12%,19 whereas a comprehensive literature review estimates a mortality rate of 8.8% in centers publishing their outcomes.11 Esophagectomy-related morbidity rates have been reported to range between 30% and 60%.2,4 Our center has previously demonstrated that large series of esophageal resections can be done with very low or no mortality13 as was the case in this series. Lower major complication rates after surgery were also noted in the current study. Limitations within this study as well as similarities in outcomes make it difficult to reach any broad conclusions about whether esophagectomy or endotherapy is ultimately ‘‘better.’’ The number of patients enrolled limits the power of the study to detect subtle differences in outcomes. The nonrandomized, retrospective nature of the study prevents controlling many variables. The 2 groups are similar but clearly not equivalent. Older (although not necessarily sicker) patients resided in the endotherapy group, whereas the extent of BE was greater in surgical patients. This is not surprising because endotherapy was initially reserved for poor surgical candidates, and only more recently have younger, healthier patients been referred for consideration of nonsurgical treatment. Therefore, it is difficult to accurately compare survival curves of groups with significantly different ages, except to note that the 4-year survival rate in both groups is high. It is also important that the one treatment-associated death and 3 of the 4 cases of progression to cancer in the endotherapy group occurred in patients who were deemed not to be surgical candidates. If we consider only surgical candidates who underwent ET, 97% are cancer free and 95% show no HGD at most recent follow-up, which is very similar to the surgical group. Among surgical candidates, ET resulted in no major complications and a 27% minor complication rate, significantly fewer than among the surgical group. Among the 8 patients who demonstrated HGD on most recent biopsy specimens, only 4 had been followed for more than 2 years, and it is anticipated that the residual HGD in these individuals will clear with additional treatments. In fact, biopsy specimens obtained after the close of the study period showed that HGD had been cleared in 5 of the original 8. Six of the 8 were treated only with EMR, reflecting the evolution of ET over the last decade away from broad ablative therapy (ie, PDT alone) to a more focused excision-based therapy, or excision for high-risk areas followed by ablation.20 Since the close of this study we have begun using EMR plus radiofrequency ablation for longer areas of BE. Unfortunately, small numbers in the ET subgroups and differences in disease severity and duration of follow-up prevent us from drawing conclusions about one method
of ablation versus another. For instance, individuals with short-segment BE may have been treated with only APC or EMR, whereas those with long segments or advanced histologic grades were more likely to have undergone EMR plus PDT or PDT alone. Median follow-up among members of the ET group was only 20 months and it is possible that with time, additional cancers will develop. However, no cancers have developed in patients in whom endotherapy had initially cleared all dysplasia. This suggests that treatment failures result from progression of persistent HGD rather than deterioration of healthy new tissue or ‘‘buried’’ HGD. Small areas of Barrett’s tissue beneath new squamous tissue were discovered on biopsy specimens from 5 PDT patients within 6 months of treatment, but on subsequent biopsy specimens, no further buried glands were found. It is possible that buried glands with no connection to surface BE may atrophy with time.21 This absence, however, may also have resulted from sampling error caused by the random nature of surveillance biopsies. Previously undiscovered cancer was identified in 25% of the surgery patients. It is unclear whether any undetected invasive cancer existed in the endotherapy group. Three invasive cancers were identified after EMR in patients from the initial cohort who were anticipating endotherapy. They were directed to surgical therapy and were not included in this analysis. Two early patients who did not undergo initial EMR and were treated only with PDT did progress to invasive cancer. It is conceivable that they harbored undetected invasive cancer at the time of treatment. The study reveals potential pitfalls unique to endotherapy. Because the success of ET often depends on multiple applications of therapy to recurrent or persistent BE, close follow-up is essential. At 3 years, 8 patients (13%) had failed to return for follow-up. Although several of these patients were elderly and all were free of dysplasia, 2 were under 60 years old and refused to return for follow-up despite frequent requests from their physicians. We found that the cost of both therapies was high; however, these costs are comparable to charges incurred for treatment of pancreatic, ovarian, and lung cancers.22 Median charges to date show 50% greater charges for the esophagectomy group compared with the ET group. However, these data may be deceptive. After esophagectomy and the immediate follow-up, surgical patients are unlikely to generate additional charges directly related to BE. Endotherapy patients continue to require long-term surveillance and possibly additional therapy. The exact duration and interval for endoscopy after complete ablation of BE has not been established. Lifelong charges for ET for a 50-year-old man will likely exceed those of esophagectomy. In an analysis of cost-effectiveness of PDT for treatment of BE with HGD, Hur et al23 showed that PDTwould cost $7100 more than esophagectomy. However, they did conclude that quality-adjusted life expectancy was greater in the PDT-treated patients.
www.giejournal.org
Volume 67, No. 4 : 2008 GASTROINTESTINAL ENDOSCOPY 599
Treatment of Barrett’s esophagus with early neoplasia
Schembre et al
1. Devesa SS, Blot WJ, Fraumeni JF Jr. Changing patterns in the incidence of esophageal and gastric carcinoma in the United States. Cancer 1998;83:2049-53.
2. Sampliner RE. The Practice Parameters Committee of the American College of Gastroenterology: updated guidelines for the diagnosis, surveillance and therapy of Barrett’s esophagus. Am J Gastroenterol 2002;97:1888-95. 3. Overholt BF, Lightdale CJ, Wang KK, et al. Photodynamic therapy with porfimer sodium for ablation of high-grade dysplasia in Barrett’s esophagus: international, partially blinded, randomized phase III trial. Gastrointest Endosc 2005;62:488-98. 4. Sharma VK, Fleischer DE, Wang KK, et al. A randomized multi-center trial of radiofrequency (RF) ablation of specialized intestinal metaplasia (SIM) of the esophagus using a balloon-based bipolar electrode array: preliminary results [abstract]. Gastrointest Endosc 2004;59:AB113. 5. Jung KW, Lee OJ. Extensive spontaneous submucosal dissection of the esophagus: Long-term sequential endoscopic observation and treatment. Gastrointest Endosc 2002;55:262-5. 6. Johnston MH. Cryotherapy and other newer techniques [review]. Gastrointest Endosc Clin North Am 2003;13:491-504. 7. Dulai GS, Jensen DM, Cortina G, et al. Randomized trial of argon plasma coagulation vs. multipolar electrocoagulation for ablation of Barrett’s esophagus. Gastrointest Endosc 2005;61:232-40. 8. Birkmeyer JD, Siewers AE, Finlayson EV, et al. Hospital volume and surgical mortality in the United States. N Engl J Med 2002;346: 1128-37. 9. Headrick JR, Nichols FC 3rd, Miller DL, et al. . High-grade esophageal dysplasia: long-term survival and quality of life after esophagectomy. Ann Thorac Surg 2002;73:1697-703. 10. Buttar NS, Wang KK, Lutzke LS, et al. Combined endoscopic mucosal resection and photodynamic therapy for esophageal neoplasia within Barrett’s esophagus. Gastrointest Endosc 2001;54:682-8. 11. Overholt BF, Panjehpour M, Halberg DL. Photodynamic therapy for Barrett’s esophagus with dysplasia and/or early stage carcinoma: long-term results. Gastrointest Endosc 2003;58:183-8. 12. May A, Gossner L, Pech O, et al. Intraepithelial high-grade neoplasia and early adenocarcinoma in short-segment Barrett’s esophagus (SSBE): curative treatment using local endoscopic treatment techniques. Endoscopy 2002;34:604-10. 13. Pacifico RJ, Wang KK, Wongkeesong LM, et al. Combined endoscopic mucosal resection and photodynamic therapy versus esophagectomy for management of early adenocarcinoma in Barrett’s esophagus. Clin Gastroenterol Hepatol 2003;1:252-7. 14. Reed MF, Tolis G Jr, Edil BH, et al. Surgical treatment of esophageal high-grade dysplasia. Ann Thorac Surg 2005;79:1110-5. 15. National Cancer Institute. 2003. Available at: www.cancer.gov. Accessed Dec 2, 2006. 16. Falk GW, Rice TW, Goldblum JR, et al. Jumbo biopsy forceps protocol still misses unsuspected cancer in Barrett’s esophagus with high-grade dysplasia. Gastrointest Endosc 1999;49:170-6. 17. Reid BJ, Blount PL, Rabinovitch PS. Biomarkers in Barrett’s esophagus. Gastrointest Endosc Clin North Am 2003;13:369-97. 18. Ormsby AH, Petras RE, Henricks WH, et al. Observer variation in the diagnosis of superficial oesophageal adenocarcinoma. Gut 2002;51: 671-6. 19. McCulloch P, Ward J, Tekkis PP. Mortality and morbidity in gastrooesophageal cancer surgery: initial results of ASCOT multicentre prospective cohort study. BMJ 2003;327:1192-7. 20. Bergman JJ. Endoscopic resection for treatment of mucosal Barrett’s cancer: time to swing the pendulum. Gastrointest Endosc 2007;65: 11-3. 21. Hornick JL, Blount PL, Sanchez CA, et al. Biologic properties of columnar epithelium underneath reepithelialized squamous mucosa in Barrett’s esophagus. Am J Surg Pathol 2005;29:372-80. 22. Chang S, Long SR, Kutikova L, et al. Estimating the cost of cancer: results on the basis of claims data analyses for cancer patients diagnosed with seven types of cancer during 1999 to 2000. J Clin Oncol 2004;22:3524-30.
600 GASTROINTESTINAL ENDOSCOPY Volume 67, No. 4 : 2008
www.giejournal.org
Unfortunately, subgroup analysis did not clearly define which patients might benefit more from each therapy. Two of 4 ET failures progressing to cancer occurred in very overweight, older men with large hiatal hernias and long-segment BE. One of these patients had his postPDT omeprazole changed to a histamine-2 blocker by an uninformed central pharmacy, which very likely compromised repithelialization. Wolfsen et al24 have shown that individuals with long-segment BE are at higher risk of ET failure after PDT and that advanced age and the presence of large hiatal hernias can lead to persistent esophageal acid exposure despite high doses of antisecretory medications.25 Individuals with these characteristics may be at higher risk for ET failure; however, these same individuals are likely to be at higher risk for surgical complications as well. Larger series will be needed to better define selection criteria for determining when endotherapy or surgery is more appropriate. Because we have had excellent results with both treatment options, we do not usually push one therapy over the other. After an informed discussion, we allow patients to choose their therapy. Although most opt for an initial trial of ET, many still choose esophagectomy because of the perception that it affords a complete and immediate cure. In conclusion, both esophagectomy and endotherapy represent appropriate and highly successful therapies for treating patients with BE with HGD or IMC when performed at a large-volume center. Both treatments can provide high, rates of clearance of dysplastic tissue; however, these treatments can result in morbidity and mortality. Esophagectomy is associated with less risk of persistence of dysplasia or progression to cancer, whereas endotherapy results in fewer short-term complications. Until carefully controlled prospective trials comparing surgical and endoscopic therapies are performed, the decision of how to treat BE with HGD or IMC must be made on a case-by-case basis and should hinge on patient preference after thorough discussions with surgical and endoscopic specialists. DISCLOSURE The following authors report that they have no disclosures relevant to this publication: J. L. Huang, O. S. Lin, N. Cantone, D. E. Low. The following author has disclosed an actual or potential conflict: D. B. Schembre has received an unrestricted investigator research grant from TAP Pharmaceutical Products Inc, Lake Forrest, Illinois, which assisted in the preparation of the document.
REFERENCES
Schembre et al 23. Hur C, Nishioka NS, Gazelle GS. Cost-effectiveness of photodynamic therapy for treatment of Barrett’s esophagus with high grade dysplasia. Dig Dis Sci 2003;48:1273-83. 24. Wolfsen HC, Woodward TA, Raimondo M. Photodynamic therapy for dysplastic Barrett esophagus and early esophageal adenocarcinoma. Mayo Clin Proc 2002;77:1176-81. 25. Fass R, Sampliner RE, Malagon IB, et al. Failure of oesophageal acid control in candidates for Barrett’s oesophagus reversal on a very high dose of proton pump inhibitor. Aliment Pharmacol Ther 2000;14:597-602.
www.giejournal.org
Treatment of Barrett’s esophagus with early neoplasia
Received February 22, 2007. Accepted August 20, 2007. Current affiliations: Division of Gastroenterology (D.B.S., O.L., N.C.) and Department of Surgery (J.H., D.L.), Virginia Mason Medical Center, Seattle, Washington, USA. Presented at Digestive Disease Week, Los Angeles, May 21-24, 2006 (Gastrointest Endosc 2006;63:AB83). Reprint requests: Drew B. Schembre, MD, Division of Gastroenterology, Virginia Mason Medical Center, 1100 9th Ave. Seattle, WA 98101.
Volume 67, No. 4 : 2008 GASTROINTESTINAL ENDOSCOPY 601
Treatment of Barrett’s esophagus with early neoplasia: a comparison of endoscopic therapy and esophagectomy Drew B. Schembre, MD, Jasmine L. Huang, MD, Otto S. Lin, MD, MSc, Nico Cantone, BS, Donald E. Low, MD Seattle, Washington, USA
Background: Endoscopic therapies for early neoplasia in Barrett’s esophagus may be a viable alternative to esophagectomy. Objective: Our purpose was to compare endoscopic therapy and esophagectomy. Design: Retrospective review from a single institution. Setting: A medium-sized tertiary referral center. Patients and Interventions: All patients with Barrett’s esophagus and dysplasia or intramucosal carcinoma treated by photodynamic therapy (PDT), EMR, or argon plasma coagulation (APC) or esophagectomy with curative intent from May 1998 until November 2005. Main Outcome Measurements: Survival, progression to cancer, eradication of dysplasia and Barrett’s esophagus, major and minor complications, and costs were compared. Results: Sixty-two patients who underwent endoscopic therapy (2 APC alone, 18 EMR þ APC, 20 PDT þ APC, and 22 EMR þ PDT þ APC) and 32 patients who underwent esophagectomy met the inclusion criteria. The 30-day mortality rate included 1 patient in the endotherapy group (2%) and none in the surgical group (P Z.49). No deaths from esophageal cancer occurred in either group. Cancer developed in 6% of endotherapy patients and in none in the surgical cohort (P!.05). Major and minor complications occurred in 8% and 31% of endotherapy patients, respectively, and 13% and 63% of surgery patients (P Z.50, P!.001). Median cost to date was $40,079 for endotherapy and $66,060 for esophagectomy (P!.001). Limitations: Retrospective study, relatively short follow-up, small numbers. Conclusions: Both endotherapy and esophagectomy can effectively treat high-grade dysplasia and intramucosal carcinoma associated with Barrett’s esophagus. Endotherapy is associated with a higher risk of tumor progression, although this is uncommon. Esophagectomy incurs higher initial costs and results in more frequent minor complications but is usually curative. (Gastrointest Endosc 2008;67:595-601.)
Copyright ª 2008 by the American Society for Gastrointestinal Endoscopy 0016-5107/$32.00 doi:10.1016/j.gie.2007.08.042
programs.2 Once HGD has been found, management has included close endoscopic surveillance, esophagectomy, and more recently, endoscopic ablative therapies (ET). Endoscopic therapies have included photodynamic therapy (PDT), EMR, endoscopic submucosal dissection (ESMD), thermal ablation, radiofrequency ablation, and cryotherapy.3-7 Although debate exists whether surveillance or treatment of HGD is more appropriate, surgical esophagectomy remains the standard intervention for fit patients with the disorder in most parts of the United States. However, esophagectomy for esophageal cancer has been associated with substantial morbidity and mortality rates, especially at low-volume centers.8 The risk of
www.giejournal.org
Volume 67, No. 4 : 2008 GASTROINTESTINAL ENDOSCOPY 595
The majority of adenocarcinomas of the esophagus arise within Barrett’s esophagus (BE).1 Identification of the immediate histologic precursor to cancer, high-grade dysplasia (HGD), is the end point of current Barrett’s surveillance Abbreviations: APC, argon plasma coagulation; ASA, American Society of Anesthesiologists; BE, Barrett’s esophagus; ESMD, endoscopic submucosal dissection; ET, endoscopic ablative therapies; HGD, high-grade dysplasia; IMC, intramucosal carcinoma; IRB, institutional review board; LGD, low-grade dysplasia; PDT, photodynamic therapy; VMMC, Virginia Mason Medical Center.
Treatment of Barrett’s esophagus with early neoplasia
esophagectomy for HGD appears to be somewhat less, but it is still significant.9 Over the last 10 years, numerous studies have shown that ET can effectively treat HGD and even intramucosal carcinoma (IMC), possibly with lower complication rates than esophagectomy.10-12 Studies directly comparing ET and surgical therapy are limited.13,14 This study compares patients who have undergone either esophagectomy or ET for BE with HGD or IMC at one high-volume center.
Schembre et al
Capsule Summary What is already known on this topic d
What this study adds to our knowledge d
METHODS We performed a retrospective analysis of 2 institutional review board (IRB)–approved prospective databases: an esophageal resection database that has been in place since 1991 and an endoscopic therapy database that has been in place since 1998. To reduce variability between groups, we included only those patients who underwent esophagectomy or ET with curative intent between May 1, 1998, and November 30, 2005. Database information included patient age, sex, weight, and American Society of Anesthesiologists (ASA) classification, endotherapy and surgical data (type of therapy), pathologic findings (degree of dysplasia, complete resection, presence of invasive cancer, length of Barrett’s segment), posttreatment complications, total hospital charges, and long-term survival. For ET patients, the database also included number of endoscopies, number and types of therapies, and histologic diagnosis at most recent biopsy. All appropriate patients received both surgical and gastroenterology consultations before treatment. All pretreatment and posttreatment biopsies were reviewed by one or more pathologists at Virginia Mason Medical Center (VMMC), with equivocal samples reviewed additionally by nationally recognized experts in Barrett’s histology at the University of Washington in Seattle, or the Cleveland Clinic in Cleveland, Ohio. ET included PDT, EMR, argon plasma coagulation (APC), or any combination of the techniques. ET patients underwent follow-up endoscopy with 4-quadrant jumbo biopsies over the entire length of the pretreatment Barrett’s segment of the esophagus at 3-month intervals for 1 year and then at 6-month intervals for 1 year and then yearly if no dysplasia was identified. Patients found to have invasive cancer during initial staging and those who had less than 6 months of follow-up were excluded. From 1998 to 2000, PDT with porfimer sodium (Photofrin, Axcan Pharma, Mont-Saint-Hilaire, Quebec) 2 mg/kg þ 200 j/cm light exposure at 620 nm was the primary treatment for BE with HGD or IMC. From 2000 to 2003, we used EMR for focal nodular disease followed by PDT for any residual Barrett’s tissue. After 2003, EMR was used for large-area BE less than 5 cm with PDT reserved for Barrett’s tissue more than 5 cm in length. APC at 90 watts was used to treat small-area primary or residual disease. All patients underwent initial endoscopy with EUS to help identify invasive disease. The goal 596 GASTROINTESTINAL ENDOSCOPY Volume 67, No. 4 : 2008
Despite its substantial morbidity and mortality rates, especially at low-volume centers, esophagectomy remains the standard intervention for patients with Barrett’s high-grade dysplasia.
d
In a retrospective review of 94 patients, both endoscopic ablation and esophagectomy effectively treated Barrett’s high-grade dysplasia and intramucosal carcinoma. Endotherapy was associated with a higher risk of tumor progression, and esophagectomy incurred higher initial costs and more frequent minor complications.
of ET in every patient was complete eradication of all Barrett’s mucosa. All esophagectomy patients underwent operation by a single surgeon (D.L.). Preoperative investigations in operative candidates included CT scan of the chest, abdomen, and pelvis, along with endoscopy and EUS. All patients over 50 years of age underwent objective cardiac testing. In addition, all patients were reviewed at our institution’s thoracic oncology tumor board. Resectional approach was individualized according to patient physiology and characteristics. The primary end points are the number and percentage of patients surviving, those who are cancer free, and those who are free of HGD. Secondary end points include eradication of all Barrett’s mucosa, frequency and severity of complications, and cost of treatment. Treatment costs are represented by total hospital and outpatient charges related to esophageal disease and subsequent treatment accrued at VMMC from the time of the initial consultation until December 1, 2005. Comparative analysis was done with use of the unpaired Student t test for continuous variables and the c2 test for categorical variables. All tests were 2-tailed. P values !.05 were considered significant. Statistical tests were performed with SPSS 13.0 (SPSS, Chicago, Ill) or Microsoft Excel (Microsoft, Redmond, Wash). All patients were U.S. citizens and survival data were checked against the Social Security Administration Death Index.
RESULTS From 1998 to 2005, 117 patients were referred to our institution for treatment of presumptive BE with HGD or IMC. Twenty-three patients were excluded from analysis because disease discovered at initial examination was T1sm or greater (3), low-grade dysplasia (LGD) (4), or no dysplasia (8) or follow-up has been less than 6 months www.giejournal.org
Schembre et al
Treatment of Barrett’s esophagus with early neoplasia
TABLE 1. Patient characteristics Endotherapy
Esophagectomy
Significance
No. of patients
62
32
Median age (y)
70
64
P Z .003
Percent male
73
93
P Z .007
Average ASA level
2.6
2.5
P Z .62
Average length of Barrett’s segment (cm)
5.1
7.2
P Z .024
TABLE 2. Results of treatment Endotherapy
Esophagectomy
Significance
Progression to cancer
4 (6%)
0
P Z .14
Persistent dysplasia
8 (13%)
1 (3%)
P Z .12
27 (44%)
1 (3%)
P ! .001
89%
93%
P Z .49
Procedure-related mortality
1 (2%)
0
P Z .47
Major complications
5 (8%)
4 (13%)
P Z .50
Minor complications
20 (32%)
26 (66%)
P ! .001
Hospital charges
$40,079
$66,060
P ! .001
Residual Barrett’s segment Overall 4-year adjusted survival rate
(8). Sixty-two patients (48 with HGD and 13 with IMC) underwent ET: 2 with APC alone, 18 with EMR alone, 20 with PDT alone, and 22 with EMR and PDT. All ET patients have had some focal treatment with APC during follow-up examinations. Thirty-two patients (15 with HGD and 17 with IMC) underwent esophagectomy with the following surgical approaches: 4 transhiatal, 10 Ivor-Lewis, and 18 left thoracoabdominal. Additional characteristics of the 2 groups are shown in Table 1. The median follow-up was 20 months for the endotherapy group (range 6 to 84 months) and 48 months for surgical patients (range 6 to 88 months). Two ET patients were lost to endoscopic follow-up after 1 year, 3 patients at 2 years, and 3 patients after 3 years. No surgical patients were lost to follow-up. Survival data are complete on all patients. One patient in the endotherapy group died of sepsis resulting from diverticulitis 11 days after treatment. There was no 30-day mortality in the surgical group (P not significant). Adjusted survival at 4 years was 93% for the surgical group and 89% for the endotherapy group (P Z.49). None of these deaths were due to esophageal cancer. Cancer developed in 4 (6%) of the ET patients. Two cases of intramucosal cancer and 2 cases of invasive cancer developed in patients with a pretreatment diagnosis of HGD. One patient with invasive cancer underwent chemoradiation and one underwent radiation alone followed by esophageal stenting. One patient with IMC is
still undergoing ET, and one elected to undergo esophagectomy. Three additional patients in the ET group opted for surgical resection for persistent HGD after an average of 2 treatment sessions. No distant cancers developed in the surgical cohort (P ! .05), despite identifying previously unrecognized cancer in 8 (25%) of the resected esophagi. One patient who underwent esophagectomy after EMR for IMC had no identifiable dysplasia or BE at surgery. All surgical patients had R0 resections for cancer. One patient with a very long segment of BE had persistent HGD proximal to a high anastamosis and underwent successful ET of the residual area of dysplasia. Median overall hospital charges (to date) were $40,079 for the endotherapy group and $66,060 for the surgical group. Additional results are shown in Table 2. Specific posttreatment complications are presented in Tables 3 and 4. There was no statistically significant difference in the incidence of major complications between the 2 treatment groups. Minor complications were more common in the surgical group (P ! .001). Histologic examination among 44 ET patients with HGD who did not progress to cancer has shown 8 patients with residual focal HGD. These patients are undergoing further therapy. No ET patient among the 13 with intramucosal cancer has shown any residual cancer or HGD on the most recent biopsy. After ET, the average length of
www.giejournal.org
Volume 67, No. 4 : 2008 GASTROINTESTINAL ENDOSCOPY 597
Treatment of Barrett’s esophagus with early neoplasia
Schembre et al
TABLE 3. Complications of endotherapy No. (%) Major complications (8%)* Death
1 (2%)
Bleeding
2 (3%)
Prolonged hospitalization
3 (5%)
Minor complications (31%)* Stricture
13 (21%)
Photosensitivity
4 (6%)
Pneumonia
2 (3%)
*The sum of individual complications may not equal the total complication rate because one individual may have more than one event.
Figure 1. Number of patients with endotherapy histologic features: initial and at most recent follow-up.
TABLE 5. Results by type of endotherapy APC alone
EMR
Progression to cancer
0
1 (6%)
Persistent dysplasia
0
5 (28%) 1 (5%)
TABLE 4. Complications of surgery No. (%) Major complications (13%)* Anastomotic leaky
1 (3%)
Chyle leaky
2 (6%)
Deep venous thrombosis
1 (3%)
Minor complications (63%)*
Residual Barrett’s segment
PDT
EMR D PDT
2 (10%) 1 (5%) 2 (9%)
1 (50%) 10 (56%) 8 (40%) 8 (36%)
Procedure-related mortality
0
0
0
1 (5%)
Major complications
0
1 (6%)
1 (5%)
3 (14%)
0
6 (33%) 5 (25%) 8 (36%)
Pneumonia
2 (6%)
Minor complications
Atrial fibrillation
6 (19%)
Wound infection
3 (9%)
Small numbers in multiple groups and variance in degree of disease at presentation precludes meaningful statistical comparisons among groups.
Stricture Pneumothorax
15 (47%) 1 (3%)
In the United States, the incidence of adenocarcinoma of the esophagus is rising at one of the highest rates.15 The current approach to BE with HGD includes surveillance, esophagectomy, or endoscopic ablation. In theory, surveillance identifies those individuals who progress to cancer early, allowing intervention while it
may still be curative and sparing those who never progress from the risks of therapy. However, invasive cancer may be difficult to recognize endoscopically, and even jumbo biopsy specimens at close intervals may miss established malignancies.16 Further, the rate of progression of HGD to cancer can vary widely, with patients with multifocal HGD and aneuploidy at particularly high risk for malignant progression.17 Moreover, delaying intervention in a younger patient leads to years of surveillance endoscopy and necessitates that the patient endure the anxiety of harboring a precancerous condition. The current study shows that in a tertiary referral center both esophagectomy and endotherapy can be effective and safe methods for treating BE with HGD or IMC. We chose to include both HGD and IMC in our analysis because the 2 histologic types behave similarly, with very low risk of nodal spread. In addition, intraobserver agreement among pathologists differentiating them is poor.18 Esophagectomy has been identified historically as a procedure with high morbidity and mortality rates.
598 GASTROINTESTINAL ENDOSCOPY Volume 67, No. 4 : 2008
www.giejournal.org
*The sum of individual complications may not equal the total complication rate because one individual may have more than one event. yTreated conservatively.
persistent Barrett’s tissue was 1 cm. These results are presented in Figure 1. A breakdown of ET patients by treatment modality is shown in Table 5.
DISCUSSION
Schembre et al
Treatment of Barrett’s esophagus with early neoplasia
Mortality rates of 8% in high-volume centers and 20% in low-volume centers have been reported as recently as 2002.7 International databases such as the British ASCOT Report demonstrates a mortality rate of 12%,19 whereas a comprehensive literature review estimates a mortality rate of 8.8% in centers publishing their outcomes.11 Esophagectomy-related morbidity rates have been reported to range between 30% and 60%.2,4 Our center has previously demonstrated that large series of esophageal resections can be done with very low or no mortality13 as was the case in this series. Lower major complication rates after surgery were also noted in the current study. Limitations within this study as well as similarities in outcomes make it difficult to reach any broad conclusions about whether esophagectomy or endotherapy is ultimately ‘‘better.’’ The number of patients enrolled limits the power of the study to detect subtle differences in outcomes. The nonrandomized, retrospective nature of the study prevents controlling many variables. The 2 groups are similar but clearly not equivalent. Older (although not necessarily sicker) patients resided in the endotherapy group, whereas the extent of BE was greater in surgical patients. This is not surprising because endotherapy was initially reserved for poor surgical candidates, and only more recently have younger, healthier patients been referred for consideration of nonsurgical treatment. Therefore, it is difficult to accurately compare survival curves of groups with significantly different ages, except to note that the 4-year survival rate in both groups is high. It is also important that the one treatment-associated death and 3 of the 4 cases of progression to cancer in the endotherapy group occurred in patients who were deemed not to be surgical candidates. If we consider only surgical candidates who underwent ET, 97% are cancer free and 95% show no HGD at most recent follow-up, which is very similar to the surgical group. Among surgical candidates, ET resulted in no major complications and a 27% minor complication rate, significantly fewer than among the surgical group. Among the 8 patients who demonstrated HGD on most recent biopsy specimens, only 4 had been followed for more than 2 years, and it is anticipated that the residual HGD in these individuals will clear with additional treatments. In fact, biopsy specimens obtained after the close of the study period showed that HGD had been cleared in 5 of the original 8. Six of the 8 were treated only with EMR, reflecting the evolution of ET over the last decade away from broad ablative therapy (ie, PDT alone) to a more focused excision-based therapy, or excision for high-risk areas followed by ablation.20 Since the close of this study we have begun using EMR plus radiofrequency ablation for longer areas of BE. Unfortunately, small numbers in the ET subgroups and differences in disease severity and duration of follow-up prevent us from drawing conclusions about one method
of ablation versus another. For instance, individuals with short-segment BE may have been treated with only APC or EMR, whereas those with long segments or advanced histologic grades were more likely to have undergone EMR plus PDT or PDT alone. Median follow-up among members of the ET group was only 20 months and it is possible that with time, additional cancers will develop. However, no cancers have developed in patients in whom endotherapy had initially cleared all dysplasia. This suggests that treatment failures result from progression of persistent HGD rather than deterioration of healthy new tissue or ‘‘buried’’ HGD. Small areas of Barrett’s tissue beneath new squamous tissue were discovered on biopsy specimens from 5 PDT patients within 6 months of treatment, but on subsequent biopsy specimens, no further buried glands were found. It is possible that buried glands with no connection to surface BE may atrophy with time.21 This absence, however, may also have resulted from sampling error caused by the random nature of surveillance biopsies. Previously undiscovered cancer was identified in 25% of the surgery patients. It is unclear whether any undetected invasive cancer existed in the endotherapy group. Three invasive cancers were identified after EMR in patients from the initial cohort who were anticipating endotherapy. They were directed to surgical therapy and were not included in this analysis. Two early patients who did not undergo initial EMR and were treated only with PDT did progress to invasive cancer. It is conceivable that they harbored undetected invasive cancer at the time of treatment. The study reveals potential pitfalls unique to endotherapy. Because the success of ET often depends on multiple applications of therapy to recurrent or persistent BE, close follow-up is essential. At 3 years, 8 patients (13%) had failed to return for follow-up. Although several of these patients were elderly and all were free of dysplasia, 2 were under 60 years old and refused to return for follow-up despite frequent requests from their physicians. We found that the cost of both therapies was high; however, these costs are comparable to charges incurred for treatment of pancreatic, ovarian, and lung cancers.22 Median charges to date show 50% greater charges for the esophagectomy group compared with the ET group. However, these data may be deceptive. After esophagectomy and the immediate follow-up, surgical patients are unlikely to generate additional charges directly related to BE. Endotherapy patients continue to require long-term surveillance and possibly additional therapy. The exact duration and interval for endoscopy after complete ablation of BE has not been established. Lifelong charges for ET for a 50-year-old man will likely exceed those of esophagectomy. In an analysis of cost-effectiveness of PDT for treatment of BE with HGD, Hur et al23 showed that PDTwould cost $7100 more than esophagectomy. However, they did conclude that quality-adjusted life expectancy was greater in the PDT-treated patients.
www.giejournal.org
Volume 67, No. 4 : 2008 GASTROINTESTINAL ENDOSCOPY 599
Treatment of Barrett’s esophagus with early neoplasia
Schembre et al
1. Devesa SS, Blot WJ, Fraumeni JF Jr. Changing patterns in the incidence of esophageal and gastric carcinoma in the United States. Cancer 1998;83:2049-53.
2. Sampliner RE. The Practice Parameters Committee of the American College of Gastroenterology: updated guidelines for the diagnosis, surveillance and therapy of Barrett’s esophagus. Am J Gastroenterol 2002;97:1888-95. 3. Overholt BF, Lightdale CJ, Wang KK, et al. Photodynamic therapy with porfimer sodium for ablation of high-grade dysplasia in Barrett’s esophagus: international, partially blinded, randomized phase III trial. Gastrointest Endosc 2005;62:488-98. 4. Sharma VK, Fleischer DE, Wang KK, et al. A randomized multi-center trial of radiofrequency (RF) ablation of specialized intestinal metaplasia (SIM) of the esophagus using a balloon-based bipolar electrode array: preliminary results [abstract]. Gastrointest Endosc 2004;59:AB113. 5. Jung KW, Lee OJ. Extensive spontaneous submucosal dissection of the esophagus: Long-term sequential endoscopic observation and treatment. Gastrointest Endosc 2002;55:262-5. 6. Johnston MH. Cryotherapy and other newer techniques [review]. Gastrointest Endosc Clin North Am 2003;13:491-504. 7. Dulai GS, Jensen DM, Cortina G, et al. Randomized trial of argon plasma coagulation vs. multipolar electrocoagulation for ablation of Barrett’s esophagus. Gastrointest Endosc 2005;61:232-40. 8. Birkmeyer JD, Siewers AE, Finlayson EV, et al. Hospital volume and surgical mortality in the United States. N Engl J Med 2002;346: 1128-37. 9. Headrick JR, Nichols FC 3rd, Miller DL, et al. . High-grade esophageal dysplasia: long-term survival and quality of life after esophagectomy. Ann Thorac Surg 2002;73:1697-703. 10. Buttar NS, Wang KK, Lutzke LS, et al. Combined endoscopic mucosal resection and photodynamic therapy for esophageal neoplasia within Barrett’s esophagus. Gastrointest Endosc 2001;54:682-8. 11. Overholt BF, Panjehpour M, Halberg DL. Photodynamic therapy for Barrett’s esophagus with dysplasia and/or early stage carcinoma: long-term results. Gastrointest Endosc 2003;58:183-8. 12. May A, Gossner L, Pech O, et al. Intraepithelial high-grade neoplasia and early adenocarcinoma in short-segment Barrett’s esophagus (SSBE): curative treatment using local endoscopic treatment techniques. Endoscopy 2002;34:604-10. 13. Pacifico RJ, Wang KK, Wongkeesong LM, et al. Combined endoscopic mucosal resection and photodynamic therapy versus esophagectomy for management of early adenocarcinoma in Barrett’s esophagus. Clin Gastroenterol Hepatol 2003;1:252-7. 14. Reed MF, Tolis G Jr, Edil BH, et al. Surgical treatment of esophageal high-grade dysplasia. Ann Thorac Surg 2005;79:1110-5. 15. National Cancer Institute. 2003. Available at: www.cancer.gov. Accessed Dec 2, 2006. 16. Falk GW, Rice TW, Goldblum JR, et al. Jumbo biopsy forceps protocol still misses unsuspected cancer in Barrett’s esophagus with high-grade dysplasia. Gastrointest Endosc 1999;49:170-6. 17. Reid BJ, Blount PL, Rabinovitch PS. Biomarkers in Barrett’s esophagus. Gastrointest Endosc Clin North Am 2003;13:369-97. 18. Ormsby AH, Petras RE, Henricks WH, et al. Observer variation in the diagnosis of superficial oesophageal adenocarcinoma. Gut 2002;51: 671-6. 19. McCulloch P, Ward J, Tekkis PP. Mortality and morbidity in gastrooesophageal cancer surgery: initial results of ASCOT multicentre prospective cohort study. BMJ 2003;327:1192-7. 20. Bergman JJ. Endoscopic resection for treatment of mucosal Barrett’s cancer: time to swing the pendulum. Gastrointest Endosc 2007;65: 11-3. 21. Hornick JL, Blount PL, Sanchez CA, et al. Biologic properties of columnar epithelium underneath reepithelialized squamous mucosa in Barrett’s esophagus. Am J Surg Pathol 2005;29:372-80. 22. Chang S, Long SR, Kutikova L, et al. Estimating the cost of cancer: results on the basis of claims data analyses for cancer patients diagnosed with seven types of cancer during 1999 to 2000. J Clin Oncol 2004;22:3524-30.
600 GASTROINTESTINAL ENDOSCOPY Volume 67, No. 4 : 2008
www.giejournal.org
Unfortunately, subgroup analysis did not clearly define which patients might benefit more from each therapy. Two of 4 ET failures progressing to cancer occurred in very overweight, older men with large hiatal hernias and long-segment BE. One of these patients had his postPDT omeprazole changed to a histamine-2 blocker by an uninformed central pharmacy, which very likely compromised repithelialization. Wolfsen et al24 have shown that individuals with long-segment BE are at higher risk of ET failure after PDT and that advanced age and the presence of large hiatal hernias can lead to persistent esophageal acid exposure despite high doses of antisecretory medications.25 Individuals with these characteristics may be at higher risk for ET failure; however, these same individuals are likely to be at higher risk for surgical complications as well. Larger series will be needed to better define selection criteria for determining when endotherapy or surgery is more appropriate. Because we have had excellent results with both treatment options, we do not usually push one therapy over the other. After an informed discussion, we allow patients to choose their therapy. Although most opt for an initial trial of ET, many still choose esophagectomy because of the perception that it affords a complete and immediate cure. In conclusion, both esophagectomy and endotherapy represent appropriate and highly successful therapies for treating patients with BE with HGD or IMC when performed at a large-volume center. Both treatments can provide high, rates of clearance of dysplastic tissue; however, these treatments can result in morbidity and mortality. Esophagectomy is associated with less risk of persistence of dysplasia or progression to cancer, whereas endotherapy results in fewer short-term complications. Until carefully controlled prospective trials comparing surgical and endoscopic therapies are performed, the decision of how to treat BE with HGD or IMC must be made on a case-by-case basis and should hinge on patient preference after thorough discussions with surgical and endoscopic specialists. DISCLOSURE The following authors report that they have no disclosures relevant to this publication: J. L. Huang, O. S. Lin, N. Cantone, D. E. Low. The following author has disclosed an actual or potential conflict: D. B. Schembre has received an unrestricted investigator research grant from TAP Pharmaceutical Products Inc, Lake Forrest, Illinois, which assisted in the preparation of the document.
REFERENCES
Schembre et al 23. Hur C, Nishioka NS, Gazelle GS. Cost-effectiveness of photodynamic therapy for treatment of Barrett’s esophagus with high grade dysplasia. Dig Dis Sci 2003;48:1273-83. 24. Wolfsen HC, Woodward TA, Raimondo M. Photodynamic therapy for dysplastic Barrett esophagus and early esophageal adenocarcinoma. Mayo Clin Proc 2002;77:1176-81. 25. Fass R, Sampliner RE, Malagon IB, et al. Failure of oesophageal acid control in candidates for Barrett’s oesophagus reversal on a very high dose of proton pump inhibitor. Aliment Pharmacol Ther 2000;14:597-602.
www.giejournal.org
Treatment of Barrett’s esophagus with early neoplasia
Received February 22, 2007. Accepted August 20, 2007. Current affiliations: Division of Gastroenterology (D.B.S., O.L., N.C.) and Department of Surgery (J.H., D.L.), Virginia Mason Medical Center, Seattle, Washington, USA. Presented at Digestive Disease Week, Los Angeles, May 21-24, 2006 (Gastrointest Endosc 2006;63:AB83). Reprint requests: Drew B. Schembre, MD, Division of Gastroenterology, Virginia Mason Medical Center, 1100 9th Ave. Seattle, WA 98101.
Volume 67, No. 4 : 2008 GASTROINTESTINAL ENDOSCOPY 601