Biomarkers in Barrett's esophagus

Why do we need biomarkers in Barrett's esophagus?

Gastroesophageal reflux disease (GERD), BE, and esophageal adenocarcinoma are diseases of modern times. Winkelstein [3] described the first cases of peptic esophagitis in 1935. Barrett [4] described the syndrome of the columnar-lined esophagus in 1950. Allison and Johnstone [5] linked GERD to BE in 1953. In their classic series in 1975, Naef et al [6] convincingly linked GERD to BE and esophageal adenocarcinoma. Once considered a rare diagnosis, the incidence of esophageal adenocarcinoma has

Biomarkers in Barrett's esophagus: how can clinicians recognize where they are in development?

Biomarkers arise from the understanding of the molecular biology of neoplastic evolution in BE, but not all molecular advances produce clinically useful biomarkers. Potential biomarkers may be discovered in basic science research or in areas outside of gastroenterology. The National Cancer Institute's Early Detection Research Network (EDRN) has recently proposed five phases in the validation of biomarkers [17]. All five phases are important for developing valid biomarkers and deserve

How do clinicians identify candidate predictive biomarkers among the nonrandom events?

There are five LOH events (9p, 17p, 18q, 5q, and 13q) that are candidate biomarkers. How do clinicians identify those that have the best potential to be predictive biomarkers, which can be used for risk stratification for progression from BE to esophageal adenocarcinoma? There are four ways any biomarker (referred to as “B” in Fig. 2) may be related to cancer [35], [36]:

• 1.

  The biomarker can consistently occur before cancer.

• 2.

  It can consistently occur after cancer develops.

• 3.

  It may occur independently

Are there other biomarker investigations that clinicians should be aware of?

As biomarkers become validated, they may be used as surrogate endpoints for cancer in clinical trials and epidemiologic studies of risk and protective factors to evaluate hypotheses that may eventually be used to prevent cancer in BE. Several examples of this type of study have been published recently and are summarized in the following paragraphs.

In one study, proliferation—in this case, S phase as assessed by PCNA immunhistochemistry—was used as an endpoint in a study that hypothesized that

What biomarkers can be used clinically?

There are no set answers to the phase at which clinicians can begin to use biomarkers for risk stratification. The current authors believe that a minimum requirement is at least two phase 4 studies—with at least one large-scale study involving hundreds of patients—with comparable results and demonstrated reproducibility of the biomarker across experienced centers. A large-scale multicenter phase 4 study would be ideal, but such a study would take 5 to 10 years to perform even if there were

Summary

This article provides a framework for clinicians who are attempting the difficult task of interpreting the Barrett's biomarker literature with the goal of improving care for their patients. Although many articles, including more that 60 proposed biomarkers, have been published on this subject, only a few describe phase 3 and 4 studies that are of interest to the clinical gastroenterologist (Table 1). For years, dysplasia grade has been the sole means of risk stratification for patients with BE,

Acknowledgements

The authors received support from the National Institutes of Health (Grants R01 CA61202 and P01 CA91955).