Advertisement

Random biopsy of the neosquamous epithelium is a waste of time and money

      Abbreviations:

      BE (Barrett’s esophagus), CEIM (complete eradication of intestinal metaplasia), EET (endoscopic eradication therapy), GEJ (gastroesophageal junction), IM (intestinal metaplasia), RFA (radiofrequency ablation), SCJ (squamocolumnar junction)
      “You can observe a lot just by watching.”–Yogi Berra
      The past decade has demonstrated that endoscopic eradication therapy (EET), combining radiofrequency ablation (RFA) and endoscopic resection (ER), is the standard of care for patients with Barrett’s esophagus (BE)–related neoplasia. Although this approach boasts high rates of reversion to squamous epithelium and a low risk of subsequent death resulting from adenocarcinoma,
      • Cotton C.C.
      • Wolf W.A.
      • Overholt B.F.
      • et al.
      Late recurrence of Barrett's esophagus after complete eradication of intestinal metaplasia is rare: final report from ablation in intestinal metaplasia containing dysplasia trial.
      the biggest stumbling block is the recurrence of BE after complete eradication of intestinal metaplasia (CE-IM). The rates of BE recurrence are approximately 10% per patient-year of follow-up.
      • Cotton C.C.
      • Wolf W.A.
      • Overholt B.F.
      • et al.
      Late recurrence of Barrett's esophagus after complete eradication of intestinal metaplasia is rare: final report from ablation in intestinal metaplasia containing dysplasia trial.
      • Sami S.S.
      • Ravindran A.
      • Kahn A.
      • et al.
      Timeline and location of recurrence following successful ablation in Barrett’s oesophagus: an international multicentre study.
      Our naïve hopes that ablative therapy might be a definitive approach to this problem have been dashed, and periodic endoscopic surveillance of the neosquamous epithelium is clearly essential to detect recurrent, sometimes dangerous, disease. The logistics and timing of postablation surveillance are informed only by expert opinion. Because current guidelines reflect practices carried forward from a time when the efficacy of EET was unclear, and because they mirror surveillance practices used before the ablation era, they are almost certainly too aggressive. Their presence in guidelines is a product of the lack of data to guide surveillance practices, and the authors’ wish to “do no harm” in prescribing postablation surveillance.
      Against this background, in the current issue of Gastrointestinal Endoscopy, Omar et al
      • Omar M.
      • Thaker A.M.
      • Wani S.
      • et al.
      Anatomic location of Barrett's esophagus recurrence after endoscopic eradication therapy: development of a simplified surveillance biopsy strategy.
      report the results of a retrospective study of the timing and location of recurrence after initially successful EET. In their cohort, 50 of 443 patients (11.3%) experienced recurrence of BE after CE-IM was initially achieved.
      • Omar M.
      • Thaker A.M.
      • Wani S.
      • et al.
      Anatomic location of Barrett's esophagus recurrence after endoscopic eradication therapy: development of a simplified surveillance biopsy strategy.
      Of the 50 patients with BE recurrence, 17 (34%) experienced nonvisible recurrence, 16 (94%) of which were identified on biopsy specimens within 2 cm of the squamous-columnar junction (SCJ). Overall, 49 of the 50 BE recurrences (98%) occurred either within 2 cm of the SCJ or at the site of a visible lesion. Only a single case of BE recurrence was both not endoscopically visible and located >2 cm from the gastroesophageal junction (GEJ). Recurrence in this patient occurred between 2 and 3 cm from the SCJ, with histologic analysis revealing nondysplastic BE. There were no reported cases of subsquamous IM. Given these findings, the authors propose a new biopsy protocol, which includes targeted biopsy of any visible lesions, along with random 4-quadrant biopsies of the SCJ, 1 cm proximal to the SCJ and 2 cm proximal to the SCJ. They postulate that this new biopsy strategy would identify 98% of all BE recurrences and would be associated with a 57% reduction in the number of biopsies needed (assuming a hypothetical BE of 6 cm).
      This study provides welcome data confirming and extending previous observations about recurrence of BE after CE-IM. The data presented are strikingly similar to those in previous studies assessing outcomes after CE-IM, in both the timing and pattern of disease recurrence. In a study by Cotton et al,
      • Cotton C.C.
      • Wolf W.A.
      • Pasricha S.
      • et al.
      Recurrent intestinal metaplasia after radiofrequency ablation for Barrett’s esophagus: endoscopic findings and anatomic location.
      among 198 patients with dysplastic BE undergoing surveillance after achieving CE-IM, 32 patients (16.2%) experienced BE recurrence after a mean follow-up time of 3 years. In this study, all recurrences more than 1 cm proximal to the GEJ were visible endoscopically, and random biopsies of the neosquamous epithelium in these areas had no incremental yield. Only 5 cases (15.6% of recurrence; 2.5% of all patients) had recurrence with histologic features more severe than the pretreatment histology.
      • Cotton C.C.
      • Wolf W.A.
      • Pasricha S.
      • et al.
      Recurrent intestinal metaplasia after radiofrequency ablation for Barrett’s esophagus: endoscopic findings and anatomic location.
      In a recent multicenter retrospective cohort study, Sami et al
      • Sami S.S.
      • Ravindran A.
      • Kahn A.
      • et al.
      Timeline and location of recurrence following successful ablation in Barrett’s oesophagus: an international multicentre study.
      assessed the recurrence of BE among 594 patients in whom CE-IM was achieved, confirmed by 2 consecutive negative endoscopic results. Of those 594 patients, 151 (25.4%) experienced BE recurrence after a median follow-up time of 2.8 years.
      • Sami S.S.
      • Ravindran A.
      • Kahn A.
      • et al.
      Timeline and location of recurrence following successful ablation in Barrett’s oesophagus: an international multicentre study.
      Among those, only 4 of the 151 recurrences (2.6%; 3 NDBE and 1 LGD) were found by random biopsy specimens taken >2 cm from the GEJ.
      Placing these data in the context of the previous studies, what can we conclude regarding endoscopic surveillance after successful EET for dysplastic BE? First, it is necessary! Although this point seems trivial, the high rates of recurrent disease argue that loss to follow-up after CE-IM puts the patient at risk for death from adenocarcinoma. Second, our currently advocated surveillance strategy is needlessly aggressive, assuming extra cost and some risk, for negligible benefit. Biopsies of any visible abnormalities in the neosquamous segment, coupled with limited sampling of the distal 2 cm of the esophagus, and 4-quadrant sampling of the high cardia, will find all or almost all recurrences. Any random sampling in areas of neosquamous epithelium that are >2 cm from the GEJ is associated with yields at or near zero. The data on this question are now sufficient to enable amendment of the current guidelines to reflect a change in recommended biopsy protocols.
      The handling of IM of the cardia after otherwise successful ablation is a matter of some debate. Some investigators, such as the present group, consider IM of the cardia to be recurrence. Others, including the previous study by our group,
      • Cotton C.C.
      • Wolf W.A.
      • Pasricha S.
      • et al.
      Recurrent intestinal metaplasia after radiofrequency ablation for Barrett’s esophagus: endoscopic findings and anatomic location.
      do not. No matter how this finding is handled, it is somewhat intellectually unsatisfying. If IM of the cardia is a recurrence, it implies both that the area of the cardia where it was found was initially treated, and that IM of the cardia has some malignant potential—otherwise, why would we care about it? However, we know that treatment of the cardia is inconsistently performed and, when it is performed, variable in its thoroughness and how far into the cardia the treatment is extended. Therefore, calling IM of the cardia a recurrence is likely often wrong because it is probably often just persistence of the previous histologic changes. We also know that this lesion, as an isolated entity, confers little if any increased risk of cancer.
      • Horwhat J.D.
      • Baroni D.
      • Maydonovitch C.
      • et al.
      Normalization of intestinal metaplasia in the esophagus and esophagogastric junction: incidence and clinical data.
      • Sharma P.
      • Weston A.P.
      • Morales T.
      • et al.
      Relative risk of dysplasia for patients with intestinal metaplasia in the distal oesophagus and in the gastric cardia.
      • Jung K.W.
      • Talley N.J.
      • Romero Y.
      • et al.
      Epidemiology and natural history of intestinal metaplasia of the gastroesophageal junction and Barrett's esophagus: a population-based study.
      Finally, we know that large proportions of patients with chronic GERD have IM of the cardia.
      • Spechler S.J.
      • Zeroogian J.M.
      • Antonioli D.A.
      • et al.
      Prevalence of metaplasia at the gastro-oesophageal junction.
      Calling persistence of this lesion after otherwise successful ablation a recurrence greatly inflates recurrence rates and enriches the recurrent pool with patients who may have no documented increased risk of cancer compared with a general GERD population.
      On the other hand, deciding not to consider IM of the cardia a recurrence comes with its own set of implications, the most important of which being that the finding is not a lesion that we should worry about. The truth is that we still poorly understand the outcomes associated with nondysplastic IM in the cardia after a patient has undergone successful ablation for dysplastic BE. It is possible that in this patient population, IM of the cardia may be associated with a more aggressive course than when this finding is seen in isolation. After all, these patients had the genetics and environmental milieu to generate dysplastic BE in the first place.
      Either way, subjecting such patients, with normal esophagus and spotty IM in the cardia, to recurrent endoscopic therapy to rid them of the IM below the Z line has not been proved to change outcomes, and it clearly comes with associated expense and potential adverse events. For these reasons, our group and others currently are not routinely treating nondysplastic cardiac IM found on an otherwise normal surveillance examination. Instead, we monitor it with routine 4-quadrant biopsies of the cardia and intervene if, and when, dysplasia of the cardia develops.
      Another interesting feature of the present study is that it joins the significant list of ablation studies analyzing their data using multiple definitions of CE-IM. The lack of a consensus definition of CE-IM continues to plague the field. To avoid misclassification due to sampling error, some studies advocate an approach requiring 2 consecutive negative endoscopic and biopsy results for IM to declare CE-IM.
      • Sami S.S.
      • Ravindran A.
      • Kahn A.
      • et al.
      Timeline and location of recurrence following successful ablation in Barrett’s oesophagus: an international multicentre study.
      • Cotton C.C.
      • Wolf W.A.
      • Pasricha S.
      • et al.
      Recurrent intestinal metaplasia after radiofrequency ablation for Barrett’s esophagus: endoscopic findings and anatomic location.
      The present study required only a single negative endoscopy and biopsy result to define CE-IM, but that demonstrates that changing the definition of CE-IM from 1 to 2 negative endoscopic results did not have a significant effect on the rate of recurrence, nor would it have affected the performance of their proposed new biopsy protocol. Given the cumulative data on the issue, it is time to arrive at a consensus definition of CE-IM. This definition should be a single biopsy session demonstrating no endoscopic or histologic evidence of IM. Acknowledging that any definition of CE-IM is subject to sampling error, this definition allows maximal use of the available data and yields study results consistent with more stringent definitions.
      In conclusion, given our better understanding of the biology of BE and its recurrence after EET, targeted sampling of any visible lesions with random sampling of only the distal 2 cm of BE and the high cardia is appropriate surveillance of the post–CE-IM patient. Random sampling of more proximal areas of the neosquamous epithelium is associated with negligible incremental diagnostic yield. Intestinal metaplasia of the cardia is a lesion of uncertain significance. Until more is known about the outcomes associated with it, routine therapy directed at it in an otherwise normal patient has the potential to do harm without demonstrated benefit. Finally, the time has come for a consensus definition of CE-IM; given the available data, the most desirable definition is a single endoscopy session documenting lack of disease.

      Disclosure

      Dr Shaheen is the recipient of research funding from Medtronic, CSA Medical, Interpace Diagnostics, C2 Therapeutics, and CDx Medical and is a consultant for Boston Scientific, Cernostics, Ambu, and PAVMed. The other author disclosed no financial relationships relevant to this publication.

      References

        • Cotton C.C.
        • Wolf W.A.
        • Overholt B.F.
        • et al.
        Late recurrence of Barrett's esophagus after complete eradication of intestinal metaplasia is rare: final report from ablation in intestinal metaplasia containing dysplasia trial.
        Gastroenterology. 2017; 153: 681-688.e2
        • Sami S.S.
        • Ravindran A.
        • Kahn A.
        • et al.
        Timeline and location of recurrence following successful ablation in Barrett’s oesophagus: an international multicentre study.
        Gut. 2019; 68: 1379-1385
        • Omar M.
        • Thaker A.M.
        • Wani S.
        • et al.
        Anatomic location of Barrett's esophagus recurrence after endoscopic eradication therapy: development of a simplified surveillance biopsy strategy.
        Gastrointest Endosc. 2019; 90: 395-403
        • Cotton C.C.
        • Wolf W.A.
        • Pasricha S.
        • et al.
        Recurrent intestinal metaplasia after radiofrequency ablation for Barrett’s esophagus: endoscopic findings and anatomic location.
        Gastroint Endosc. 2015; 81: 1362-1369
        • Horwhat J.D.
        • Baroni D.
        • Maydonovitch C.
        • et al.
        Normalization of intestinal metaplasia in the esophagus and esophagogastric junction: incidence and clinical data.
        Am J Gastroenterol. 2007; 102: 497-506
        • Sharma P.
        • Weston A.P.
        • Morales T.
        • et al.
        Relative risk of dysplasia for patients with intestinal metaplasia in the distal oesophagus and in the gastric cardia.
        Gut. 2000; 46: 9-13
        • Jung K.W.
        • Talley N.J.
        • Romero Y.
        • et al.
        Epidemiology and natural history of intestinal metaplasia of the gastroesophageal junction and Barrett's esophagus: a population-based study.
        Am J Gastroenterol. 2011; 106 (quiz 56): 1447-1455
        • Spechler S.J.
        • Zeroogian J.M.
        • Antonioli D.A.
        • et al.
        Prevalence of metaplasia at the gastro-oesophageal junction.
        Lancet. 1994; 344: 1533-1536

      Linked Article