Endoscopy to endocytoscopy to endopathology: Are we ready?

”One faces the future with one’s past.“ - Pearl S. Buck.

The 21st century has ushered in an astonishing, even dizzying, number of new endoscopic imaging modalities - high magnification endoscopy, high definition endoscopy, autofluorescence endoscopy, endoscopic optical coherence tomography, narrow band imaging, confocal endoscopy, and now endocytoscopy. In the United States and Western Europe, Barrett’s esophagus has become the ”model disease” for the application of these developing technologies.

Interest in Barrett’s esophagus has been heightened by the alarming increase in adenocarcinoma of the esophagus during the past three decades [1]. Our current clinical strategy of performing screening endoscopy in older adults with chronic symptoms of gastroesophageal reflux disease (GERD) followed by periodic endoscopic biopsy surveillance in those detected with Barrett’s esophagus is expensive and woefully suboptimal. Deficiencies in this strategy are related to a number of facts: a major proportion, nearly 40 %, of patients with Barrett’s esophagus have no significant GERD symptoms; compliance with screening endoscopy in those who have chronic GERD symptoms is poor; although columnar esophageal mucosa is easy to recognize, specialized intestinal metaplasia is patchy and can only be identified by histology; dysplastic changes in Barrett’s esophagus are multifocal and cannot be readily identified by endoscopy; and the interpretation of dysplasia is quite observer dependent [2]. Advances in endoscopic imaging can address some of these problems if they can reliably identify intestinal metaplasia and dysplasia and reduce sampling error associated with random biopsies.

A novel endoscopic imaging technique for Barrett’s esophagus will replace our current technique of white light endoscopy and random biopsy only if it has a very high sensitivity (95 % or higher) for intestinal metaplasia and/or dysplasia, a moderate specificity (85 % or higher) for dysplasia, and the ability to examine the entire affected esophagus. For it to be clinically applicable, it will also need to be relatively inexpensive. Most importantly, it will have to be interpretable by endoscopists, with an interobserver agreement that is at least as good as the agreement among pathologists interpreting dysplasia.

In this issue of Endoscopy, two groups of European investigators report on their initial experience with endocytoscopy [3] [4]. This novel technique basically tries to approach in vivo microscopy. It requires the passage of a 3.4 mm endoscope through the accessory channel of a therapeutic gastroscope. A cap is attached to the tip of the therapeutic endoscope to maintain stability, mucus is cleaned off the epithelium with N-acetylcysteine if necessary, and the epithelium is stained with methylene blue dye for contrast. This technology enables magnifications of up to × 450 or × 1125, essentially allowing the identification of single cells and nuclei in a limited sampling area that is less than half a millimeter in diameter.

In vitro microscopy using a stable microscopic stage with a thinly sectioned, fixed and stained slice of biologic tissue mounted on a glass slide is difficult enough. The performance of in vivo endocytoscopy is clearly a technology tour de force. But the question is whether this technologic advancement will eventually result in clinical advancement. The answer, if one reads the study by Pohl et al. [3], is that the glass is half empty, whereas the study by Eberl et al. [4] suggests that the glass is half full. Given the difficulty of maintaining a moving organ in focus using a flexible instrument that is over 100 cm long, it is not surprising that no more than 23 % of the images obtained by the endocytoscope in the Pohl study were interpretable [3]. It is also not surprising that a pathologist was required to interpret the histologic image obtained by endocytoscopy in both studies. In the Pohl study, at high magnification, the sensitivity and specificity were 43 % and 85 %, respectively, for diagnosing dysplasia or cancer in Barrett’s esophagus. In the Eberl study, the sensitivity and specificity were 91 % and 100 %, respectively, for esophageal lesions but 75 % and 83 %, respectively, for gastric lesions. Unlike the Pohl study, which only examined dysplasia in Barrett’s esophagus, the Eberl study evaluated all pathology and did not specify the type of lesions that were diagnosed. The Pohl study also showed that interobserver agreement for interpreting the images was marginal.

Both studies demonstrate that it is possible to obtain in vivo microscopic resolution at endoscopy. These studies also point out the many challenges that lie ahead if any of these novel technologies are to become clinically applicable. Even a sensitivity of 90 % may not be sufficient to replace current endoscopy and random biopsy techniques, and certainly a sensitivity of 43 % will put a technology back on the shelf. Gastrointestinal motility will continue to make it difficult to obtain interpretable images, and microscopic examination of the esophagus will likely require great skill and patience. These endocytoscopic or confocal endoscopic ”point and shoot” techniques suffer from the problem of limited sampling that is inherent in our current technique of obtaining small random biopsies (i. e. less than 1 % of the area of interest is sampled). Will there be any real advantage of obtaining a diagnosis at the time of endoscopy over waiting for the pathology report? Finally, perhaps the biggest hurdle for these techniques will be that they will require endoscopists to interpret histopathology. These studies show that visualization of cellular detail at endoscopy is feasible but the question that must be answered is ”Will endoscopists be willing to evolve from endoscopists to endocytoscopists to bona fide endopathologists?”

Competing interests: None

References

• 1 Devesa S S, Blot W J, Fraumeni J F. Changing patterns in the incidence of esophageal and gastric carcinoma in the United States.  Cancer. 1998;  83 2049-2053
• 2 Sharma P, McQuaid K, Dent J. et al . A critical review of the diagnosis and management of Barrett’s esophagus: the AGA Chicago Workshop.  Gastroenterology. 2004;  127 310-330
• 3 Pohl H, Koch M, Khalifa A. et al . Evaluation of endocytoscopy in the surveillance of patients with Barrett’s esophagus.  Endoscopy. 2007;  39 492-496
• 4 Eberl T, Jechart G, Golczyk M. et al . Can endo-cytoscope system (ECS) predict histology in neoplastic lesions?.  Endoscopy. 2007;  39 497-501