F or patients found to have low-grade dysplasia (LGD) in Barrett’s esophagus, two recent studies suggest that the decision between treating with radiofrequency ablation (RFA) or simply continuing endoscopic surveillance is a slam dunk for RFA.1,2 In a European, randomized trial of surveillance versus radio frequency ablation (SURF) for 136 patients who had LGD confirmed by expert pathologists, progression to high-grade dysplasia (HGD) or cancer was observed at three years in 26.5 percent of the surveillance group, but in only 1.5 percent of the RFA group (P<0.001).1 These results favoring RFA so impressed the study’s data and safety monitoring board that they terminated the trial early. A retrospective study of patients with LGD diagnosed by pathologists at three academic medical centers in the U.S. found that progression to HGD or cancer occurred during a median follow-up of approximately 29 months in 36 (28.8 percent) of 125 patients who had endoscopic surveillance alone, but in only 1 (2.2 percent) of 45 patients treated with RFA (adjusted HR=0.06; 95 percent CI 0.008-0.48).2 As straightforward as these results seem for RFA, the devil is in the details.
One bedeviling detail is the remarkably high rate of neoplastic progression in the surveillance groups of both studies. Patients in the SURF trial’s surveillance group developed cancer at a rate some 14-fold higher than that described in other large cohorts of patients with LGD.3 The reasons underlying the enormous disparities among studies on the natural history of LGD are not entirely clear, but a major factor appears to be differences in how study pathologists diagnose LGD. A number of investigations have documented poor inter-pathologist agreement on that diagnosis, 4,5 and other studies suggest that “expert” pathologists will usually downgrade a diagnosis of LGD made by community pathologists.6,7 Studies of patientschosen by pathologists who eliminate the mild or equivocal cases of LGD likely will be skewed to include only those at the highest risk for neoplastic progression.
Because of the notoriously poor inter-observer agreement in grading dysplasia, all major GI society guidelines recommend that any diagnosis of dysplasia should be confirmed by an expert pathologist. While I have made this recommendation myself, I consider it untenable for several reasons. First, there are no consensus criteria to establish a pathologist as an “expert” in this area. Furthermore, some studies document poor inter-observer agreement in diagnosing LGD even among pathologists deemed experts in Barrett’s esophagus.8,9Since a diagnosis of dysplasia is meant to indicate high cancer risk, the only unequivocal way to establish a pathologist’s expertise in this area is with a study documenting that patients whom that pathologist has diagnosed with dysplasia frequently develop cancer. By that criterion, there are very few “experts” in the world. Finally, even though the pathologists of the SURF trial and the aforementioned American study fulfill that stringent criterion for expertise in diagnosing LGD, approximately 30 percent of patients in the surveillance group of both studies had no dysplasia found on subsequent endoscopies.1,2 It is not clear whether that dysplasia truly regressed or simply was missed due to biopsy sampling error. Nevertheless, had those patients been treated with RFA, this finding would have been considered “complete eradication of dysplasia” (CE-D).Thus, spontaneous CE-D occurs frequently, even in LGD diagnosed by expert pathologists.
Another detail calling into question the benefit of RFA for LGD is the lack of a clear demonstration that RFA improves patient outcomes. In the SURF trial, RFA significantly decreased the rate of neoplastic progression.1 However, that neoplastic progression involved submucosal invasion for only one patient in the surveillance group who was treated with esophagectomy and remained tumor-free at follow-up 37 months later. For the other 17 surveillance patients who exhibited neoplastic progression, it was to an intramucosal neoplasm that was both detected by surveillance and amenable to endoscopic treatment. In the retrospective American study, two patients (1.6 percent) in the surveillance group developed submucosal neoplasms, but the report provides no information on their outcomes.sup>2 Furthermore, that study found that RFA was associated with a significant reduction in the neoplastic progression of LGD, but not in progression to adenocarcinoma specifically, and the authors acknowledged that selection bias might have influenced their findings. Thus, neither study has established that RFA is better than endoscopic surveillance (with treatment when neoplastic progression is documented) for preventing incurable cancers for patients with LGD.
Another disturbing detail is that RFA involves considerable expense and inconvenience, and some risk. Patients often require three or more expensive ablation sessions to eradicate their Barrett’s metaplasia. Adverse events caused by RFA were not rare in the SURF trial, in which 12 percent of patients developed esophageal strictures that required dilation.1 Furthermore, since recurrence of metaplasia occurs frequently after RFA, the procedure does not eliminate the need for regular endoscopic surveillance.10
In summary, the large majority of diagnoses of LGD in Barrett’s esophagus are made by community pathologists who often over-read dysplasia and, if the diagnosis is confirmed at all, it is often by a pathologist whose credentials as an expert are unestablished. The efficacy of RFA for preventing neoplastic progression of LGD has been demonstrated in only one randomized trial in which the rate of neoplastic progression in the control group was extraordinarily high, in which surveillance alone detected all progressors at a curable stage, and in which more than one-quarter of patients randomized to surveillance had no dysplasia detected on subsequent endoscopies. No study has established that RFA improves outcomes for patients with LGD, and RFA involves substantial expense, inconvenience and risk. RFA does not eliminate the need for subsequent endoscopic surveillance and does not entirely eliminate the risk of neoplastic progression. Despite all these caveats, I admit that I usually advise RFA for the treatment of LGD but I do not consider it a slam-dunk decision. Increased endoscopic surveillance is a very reasonable alternative.
Dr. Spechler serves as an ad hoc consultant for Ironwood Pharmaceuticals.
1. Phoa KN, van Vilsteren FG, Weusten BL, Bisschops R, Schoon EJ, Ragunath K, Fullarton G, Di Pietro M, Ravi N, Visser M, Offerhaus GJ, Seldenrijk CA, Meijer SL, ten Kate FJ, Tijssen JG, Bergman JJ. Radiofrequency ablation vs endoscopic surveillance for patients with Barrett esophagus and low-grade dysplasia: a randomized clinical trial. JAMA 2014; 311:1209-17
2.Small AJ, Araujo JL, Leggett CL, Mendelson AH, Agarwalla A, Abrams JA, Lightdale CJ, Wang TC, Iyer PG, Wang KK, Rustgi AK, Ginsberg GG, Forde KA, Gimotty PA, Lewis JD, Falk GW, Bewtra M. Radiofrequency ablation is associated with decreased neoplastic progression in patients with Barrett’s esophagus and confirmed low-grade dysplasia. Gastroenterology 2015; 149:567-76.
3. Rubenstein JH, Kwon RS. Radiofrequency ablation for Barrett’s esophagus with lowgrade dysplasia: a hammer looking for a nail. Gastroenterology 2014; 147:706-7.
4. Skacel M, Petras RE, Gramlich TL, Sigel JE, Richter JE, Goldblum JR. The diagnosis of low-grade dysplasia in Barrett’s esophagus and its implications for disease progression. Am J Gastroenterol 2000; 95:3383-7.
5. Montgomery E, Bronner MP, Goldblum JR, Greenson JK, Haber MM, Hart J, Lamps LW, Lauwers GY, Lazenby AJ, Lewin DN, Robert ME, Toledano AY, Shyr Y, Washington K. Reproducibility of the diagnosis of dysplasia in Barrett esophagus: a reaffirmation. Hum Pathol 2001; 32:368-78.
6. Alikhan M, Rex D, Khan A, Rahmani E, Cummings O, Ulbright TM. Variable pathologic interpretation of columnar lined esophagus by general pathologists in community practice. Gastrointest Endosc 1999; 50:23-6.
7. Curvers WL, ten Kate FJ, Krishnadath KK, Visser M, Elzer B, Baak LC, Bohmer C, Mallant- Hent RC, van Oijen A, Naber AH, Scholten P, Busch OR, Blaauwgeers HG, Meijer GA, Bergman JJ. Low-grade dysplasia in Barrett’s esophagus: overdiagnosed and underestimated. Am J Gastroenterol 2010; 105:1523-30.
8. Kerkhof M, van Dekken H, Steyerberg EW, Meijer GA, Mulder AH, de Bruïne A, Driessen A, ten Kate FJ, Kusters JG, Kuipers EJ, Siersema PD; CYBAR study group. Grading of dysplasia in Barrett’s oesophagus: substantial interobserver variation between general and gastrointestinal pathologists. Histopathology 2007; 50:920-7.
9. Wani S, Mathur SC, Curvers WL, Singh V, Alvarez Herrero L, Hall SB, Ulusarac O, Cherian R, McGregor DH, Bansal A, Rastogi A, Ahmed B, Singh M, Gaddam S, Ten Kate FJ, Bergman J, Sharma P. Greater interobserver agreement by endoscopic mucosal resection than biopsy samples in Barrett’s dysplasia. Clin Gastroenterol Hepatol 2010; 8:783-8.
10. Krishnamoorthi R, Singh S, Ragunathan K, A Katzka D, K Wang K, G Iyer P. Risk of recurrence of Barrett’s esophagus after successful endoscopic therapy. Gastrointest Endosc 2016; 83:1090-1106