By Vicki Glaser
November 18, 2008 | Miniaturized probe technology and real-time image analysis are enabling “virtual biopsies,” eliminating the need to collect colon polyp samples for pathologic analysis and then re-scope patients to remove any premalignant lesions.
Endoscopic digital imaging technology can determine in situ whether a polyp detected on visual inspection of the colon is harmless or premalignant, and the technology will improve the efficiency and reduce the cost and risk associated with colonoscopic polyp removal.
This technology, developed by Mauna Kea Technologies (Paris, France; Cellvizio, United States), took a major step forward with the recent findings of a study from the Mayo Clinic-Florida, in which a probe-based confocal laser endomicroscopy (pCLE) system was able to differentiate between harmless and premalignant, benign polyps with 90% accuracy. Michael Wallace, professor of medicine at the Mayo Clinic and lead investigator of the study, expressed confidence that with a bit of tweaking pCLE could reach nearly 100% accuracy.
In essence, pCLE miniaturizes traditional benchtop microscopy. Comprised of a 2mm diameter fiberoptic probe that fits inside a conventional endoscope, the pCLE system can image single cells and structures as small as 1 micron. Imaging with a confocal microscope allows for sectioning of specific levels of tissue and visualization at high resolution of cells situated 50 to 250 microns below the surface. The probe scans the tissue and transmits the images to a device outside the patient for analysis.
With this proximal scanning approach, probe diameter can be reduced to well below 1 mm, expanding the potential for using pCLE to image difficult to access tissues such as bladder or kidney via cystoscopy, the upper gastrointestinal tract using laparoscopy, as well as lung and other tissues.
In addition to ongoing studies on colon polyps, Wallace and colleagues are using the pCLE system to detect early esophageal lesions that occur in patients with chronic gastroesophageal acid reflux and a condition called Barrett’s esophagus. These neoplastic changes are not readily visualized on endoscopy, so the current strategy involves taking random biopsy samples at 2-cm intervals along the esophagus, an approach compromised by sampling error.
“pCLE has the potential to target biopsies to abnormal areas that are most likely to be pre-malignant,” says Wallace, “and can eliminate the 95% of benign biopsies. This has large cost implications,” he adds, as each biopsy sample has to be evaluated by a pathologist.
Although the pCLE system is an FDA-approved device, “it is not yet widely used in routine clinical practice.” Wallace’s group is using the system as part of routine care for patients that have very large polyps, in which removal of those determined to be premalignant is challenging. Endoscopic mucosal resection enables real-time imaging of the polyp margins to ensure removal of all the precancerous tissue.
Wallace also describes an exciting future role for pCLE in the diagnosis of solid cancers such as breast, liver, or pancreas. With further miniaturization of the probe it will be possible to image inside these solid organs, and even individual lymph nodes. For example, a micron-scale probe that could fit inside a biopsy needle could be combined with fine needle aspiration of suspicious breast lesions to enable in situ virtual biopsies and to help guide biopsy sample collection to minimize sampling error and missed diagnoses.