Detection of buried Barrett glands after Radiofrequency Ablation (RFA) with Volumetric Laser Endomicroscopy (VLE); a histopathology correlation study. vs 2.0

Barrett*s esophagus (BE) is a preneoplastic condition in which the normal
lining of the esophagus is replaced by columnar epithelium with intestinal
metaplasia. Malignant progression in BE develops through a step-wise process
from non-dysplastic intestinal metaplasia (IM), low-grade intraepithelial
neoplasia (LGIN), high-grade intraepithelial neoplasia (HGIN) to early
carcinoma (EC). Patients with HGIN/EC are eligible for endoscopic treatment.
Visible lesions, if estimated to be intramucosal, should be removed by
endoscopic resection (ER). Residual Barrett*s epithelium and flat lesions
containing HGIN can subsequently be treated with radiofrequency ablation (RFA).
RFA has shown to be an effective treatment with high rates of complete
eradication of dysplasia and IM with a favorable safety profile. A known
disadvantage of ablation techniques in Barrett*s esophagus is the possible
occurrence of *buried Barrett*s* (BB): the occurrence of residual Barrett
glands hidden under neosquamous epithelium (NSE). These glands may lead to
recurrence of Barrett*s epithelium after treatment or may progress to dysplasia
or cancer while hidden under the NSE. Presence of BB has been described in
multiple ablation techniques. Limiting factors in most studies on this subject
are the lack of reporting frequency of biopsy sampling of the NSE and that in
general most biopsies do not incorporate the lamina propria and BB may,
therefore, be missed. For this reason our group performed in 2008 ER and
keyhole biopsies from NSE in 22 patients two months after RFA treatment showing
no BB in any of the biopsies or endoscopic resection specimens (see the file
number of the protocol of this study under E9a). White light endoscopy (WLE)
and/or narrow band imaging (NBI) only image the superficial surface and are not
able to detect BB. Optical frequency-domain imaging (OFDI), also known as
Volumetric Laser Endomicroscopy (VLE) is a new Optical Coherence Tomography
(OCT) technique that is able to visualize underlying layers of the NSE. This
technique utilizes optical scattering based on differences in tissue
composition to perform high-speed acquisition of large luminal surfaces
creating three-dimensional images. OFDI is capable of generating
cross-sectional images of the entire circumference of the esophagus over a
length of 6 cm with an axial-resolution of up to 10 *m, which is comparable to
low-power microscopy. OCT has shown to be a feasible method to differentiate
between normal squamous mucosa, Barrett*s epithelium and HGIN/EC, with
acceptable sensitivity and specificity. In addition, also subsquamous
structures such as glands, crypts, cysts and bloodvessels can be distinguished.
The mucosa can be investigated in full thickness and over a large area. Due to
its high-resolution and high-acquisition rates, VLE is theoretically the ideal
technique for assessing the prevalence of buried glands after RFA. It may be a
promising tool for investigating the presence of BB and thus for future
follow-up of patients treated with RFA. Direct correlation of VLE images with
histopathology is needed in order to validate the technique.

The aim of this study is to investigate the feasibility of VLE to detect the
presence of glandular structures under the NSE after RFA and direct correlation
of VLE images with histopathology.

This is a single centre, single arm, observational trial which will take place
at the department of Gastroenterology of the AMC which is a tertiary-care
referral centre for patients with a Barrett*s esophagus and the detection and
treatment of early Barrett*s neoplasia.
In the completed pilot phase the feasibility of the protocol was tested next to
the primary outcome measurements. In this continuation phase of the study, it
will be tested if the results of the pilot phase can be reproduced. The same
outcome measurements will be used, adding one: 'Consensus confidence level of
BB detection in targeted suspicious area* (page 5, 7, 8, 9 protocol).
Endoscopic procedure: the esophagus will first be examined in overview with
white light endoscopy (WLE). In order to localize possible BB underneath the
NSE, 4 reference markers will be made with an electrocoagulation device. Two
markers 5 mm above the GEJ, in neutral position at the 12 o*clock position (1
dot) and at the 6 o*clock position (2 dots), subsequently at 4 cm above the GEJ
an additional 2 mark sets will be made at the 12 o*clock position (1 dot) and
at the 6 o*clock position (2 dots). The Nvision guidesheath and probe will be
introduced through the working channel of a therapeutic endoscope and
positioned in the distal esophagus including the GEJ. Once the correct position
is obtained, a full scan will be performed. Based on this scan the most likely
area containing OCT-structures suspicious for buried glands will be identified
and localized in the esophagus according to 4 reference markers. Subsequently,
the probe and balloon are removed from the endoscope. The area of interest with
a maximum diameter of 15 mm will be delineated with an additional set of
electrocoagulation markers according to the following protocol: The proximal
margin (top most) of the area will be marked by two proximal reference markers,
in the 12 o*clock position of the lesion and single markers at 3, 6 and 9
o*clock positions. When no structures are seen suspicious for buried glands,
the second set of coagulation markers will be placed at a random location at
the discretion of the endoscopist including the GEJ. After placement of the
second set of coagulationmarkers an additional VLE-scan of the distal esophagus
will be performed to allow optimal correlation between the in-vivo VLE-image
and the area of interest delineated with the second coagulationmarker set.
After imaging is completed the selected area will be resected en-bloc including
the coagulationmarkers per standard clinical practice, using the ER-cap method.
After ER, the ER-specimen will be obtained and additional ex-vivo VLE imaging
will be performed. After ER and VLE imaging four random biopsies will be
obtained under the NSE in the GEJ (<5mm) and repeated in all 4 quadrants every
2 cm, according to standard follow-up protocol.

Endoscopic follow-up is standard policy in patients that underwent RFA
treatment for Barrett's esophagus with an early neoplastic lesion. General
risks associated with a gastroscopy are: mild irritation of the throat due to
introduction of the endoscope, difficulties swallowing and retrosternal pain.
The ER, performed using the ER cap-technique, is an additional procedure that
is not routinely done during standard follow-up in patients post RFA treatment.
The ER cap-technique is widely used and a safe technique, serious complications
such as severe bleeding and perforation are rare (<3%). Perforations when occur
can mostly be handled conservatively without the need of surgery. Minor
complications, such as minor bleeding, occur in 9% of the cases and usually are
easily managed with endoscopic hemostatic techniques. Yielding an additional ER
specimen for diagnostic and histology correlation purposes was done before in
studies by our group in 2008 (See protocol with file number NL21978.018.08, ABR
form 21978 and also in another protocol: NL35326.018.11, ABR form 35326).
During this study patients will undergo an nVision pVLE procedure before and
after endoscopic resection. The resection specimen will also be imaged in a
standardized ex-vivo set-up with pVLE. The regular diagnostic and therapeutic
proces is not influenced by these extra procedures. The endoscopy will take 15
minutes longer compared to the standard endoscopy. The nVision pVLE System is
non-invasive in nature. The type of light delivered by the optical probe is
equivalent in intensity to the standard light source used and delivered by a
standard endoscope; the excitation of tissue by the light energy delivered by
the optical biopsy system is non-damaging and does not result in any thermal
effects on tissue. The additional risk associated with the application of
nVision VLE is mucosal laceration due to the extension of the balloon. In very
rare occasions perforation of the esophagus can happen when inflated above
maximum allowed pressure. A safety valve is included in the design of the
balloon to avoid this problem. In patients with an esophageal stricture, the
risk of laceration increases, and these patients are therefore excluded from
participation in this trial.