A study presented at ASCO's ninth annual Gastrointestinal Cancers Symposium has identified new biomarkers that could eventually improve early detection of esophageal cancer in patients with Barrett's esophagus (BE), a condition that significantly increases esophageal cancer risk. Using a special type of high-powered microscope, researchers identified three "optical biomarkers," specific nano-scale changes in cells of the lining of the esophagus, which enabled them to stratify patients' risks of progressing to cancer.

"If subsequent testing proves successful, our approach could lead to simpler and more effective ways of monitoring for patients with Barrett's esophagus. Such a monitoring program would identify a subset of high-risk Barrett's patients who need more intensive surveillance, and who could also be candidates for therapy to destroy the precancerous tissue," said lead researcher Randall Brand, M.D., professor of medicine at the University of Pittsburgh.

The incidence of esophageal adenocarcinoma in the U.S. is increasing faster than any other type of cancer – approximately 400 percent in the last three decades. If detected early, 5-year survival can be as high as 90 percent; advanced esophageal adenocarcinoma has a 5-year survival rate of 13 percent.

Currently, patients with Barrett's esophagus are monitored for the presence of dysplasia or early stage cancers through the use of endoscopy and biopsies. The recommended surveillance method consists of multiple random biopsies of the esophagus every one to two centimeters along the length of the area of the esophagus affected by BE every three years, if no abnormality is identified. However, this practice may miss dysplastic cells that indicate unseen high-grade dysplasia or an undetected small esophageal adenocarcinoma. "Our ultimate goal is to identify all patients with high-grade dysplasia and early cancers to allow early treatment, when therapy is most effective and the least invasive," said Brand. "The problem is that sometimes intestinal metaplasia from a patient with high-grade dysplasia looks normal under a traditional microscope."

In the study, researchers retrospectively examined archived tissue from 60 patients with Barrett's esophagus who underwent biopsies. Of these, 33 were known to have only intestinal metaplasia, meaning no dysplasia or cancer, while 27 patients had high-grade dysplasia or cancer. The investigators examined cells from biopsy specimens that only had intestinal metaplasia from both groups of patients: those who had no dysplasia and those who had high-grade dysplasia or cancer were diagnosed on a different biopsy specimen.

Using spatial-domain low-coherence quantitative phase microscopy (SL-QPM), the researchers examined these samples for extremely small changes in the cell that cannot be seen by a conventional microscope. They found at least three new features that are based on the use of light to detect abnormalities in cells that distinguished patients with Barrett's esophagus who had no cancer from those who had high-grade dysplasia/cancer. These include:

• Average optical path length of the cell nucleus, or the density of the cell nucleus
• Intra-nuclear entropy, or the random structure of the cell nucleus
• Intra-nuclear uniformity, or the texture uniformity of the cell nucleus

Using these biomarkers, the researchers developed a prediction model that resulted in 89 percent sensitivity and 76 percent specificity in distinguishing Barrett's patients with high-grade dysplasia/adenocarcinoma from those without dysplasia by only looking at cells that under a traditional microscope are non-cancerous. According to Brand, if a standard pathology examination identifies any form of dysplasia, the patient will immediately undergo treatment. These patients would not require this special imaging test. However, this test may be used for the majority of patients with Barrett's esophagus whose pathology examination showed non-dysplastic intestinal metaplasia or was ambiguous for dysplasia. Ultimately, the researchers hope the new technique would be "a simpler, more sensitive and accurate approach," requiring only two or three random biopsies and "may also allow for the identification of a subset of Barrett's patients who require less frequent monitoring."

The researchers plan to expand the study population and test additional optical biomarkers as well to improve on the current sensitivity and specificity. In the long-term, they plan on a multi-center trial using previously collected specimens at other centers to validate their results.