Novel Endoscopic Imaging Methods for the Evaluation of Blood Vessels in Gastrointestinal Cancers (IM-ANG)

Study of Minimally Invasive Endoscopic IMaging Methods for the Evaluation of neoANGiogenesis in Gastrointestinal Cancers

The aim of the project is to study the role of minimally invasive imaging methods, such as magnification endoscopy with narrow-band imaging (M-NBI) combined with confocal laser endomicroscopy (CLE), in correlation with immunohistochemical analysis, for assessing the angiogenesis status of patients with gastrointestinal tumors, in particular with colorectal and gastric cancer. Angiogenesis, i.e. the process of forming new blood vessels, represents an essential event for tumor growth and metastasis and the importance of its understanding stems from potential applications for diagnosis, prognosis stratification and mainly from the possibility of developing and improving targeted therapies. While current methods for evaluating tumor vascularity are based on immunohistochemistry techniques with microvascular density (MVD) calculations, these imply repeated tissue sampling and are not feasible in the context of clinical practice. Imaging techniques might overcome limitations associated with MDV measuring, obtaining both functional and morphological information and enabling repeated evaluations that are necessary for the assessment of a dynamic process as angiogenesis during follow-up of targeted therapies. NBI is a digitally enhanced endoscopic imaging technique that uses optical filters to illuminate tissue with light at blue and green wavelengths. These are selectively absorbed by hemoglobin and, as a result superficial vascular networks are highlighted and morphological changes in capillary patterns can be described for different lesions. CLE represents a revolutionary technology that enables endoscopists to collect real-time in vivo histological images or "virtual biopsies" of the gastrointestinal mucosa during endoscopy, and has raised significant interest for the potential clinical applications and numerous research possibilities. After intravenous administration of fluorescein as a contrast agent, CLE enables real-time visualization of the tumor vasculature, which is structurally and functionally altered compared to the normal vascular networks. Therefore M-NBI will be used for enhanced visualization of morphological changes of the superficial capillaries, while CLE will be directed towards vascular regions of interest for characterization of these changes at the microscopic level. Furthermore, imaging studies will be backed by MVD calculation using immunohistochemical methods, based on tissue samples harvested during endoscopic procedures.

The key objectives of the project are: Real-time imaging of angiogenesis in GI tumors by using minimally invasive methods: Endoscopic magnification with narrow-band imaging (M-NBI) for macroscopic assessment of the vascular pattern In vivo probe-based confocal laser endomicroscopy (pCLE) for characterization of the microvascular pattern Offline analysis of M-NBI and pCLE sequences with additional image processing software Immunohistochemical evaluation of angiogenesis in GI cancers based on MVD calculation Comparing imaging (M-NBI and pCLE) and pathological (immunohistochemistry) data using computer aided diagnosis (CAD) The study will include patients with gastric (n=20) and colorectal (n=30) cancers that will be prospectively evaluated according to the predefined protocols. Magnification endoscopy with narrow band imaging (M-NBI) will be performed for each patient after the identification of the lesion with conventional white light endoscopy. With only a push of a button optical filters are applied to reduce the illuminating light to 415 nm (blue) and 540 nm (green) wavelengths for enhanced visualization of capillary changes. Classification of the vascular pattern will be made on-site by the examiner (DIR) and multiple images will be saved for off-site later analysis. While interpretation of endoscopic images can be subjective and operator-dependent we will also perform an objective evaluation with computer-aided analysis based on image processing software. A vascular region of interest will be selected in NBI mode for targeted microscopic examination with pCLE and tissue sampling for pathology and IHC assessment. M-NBI examination will also be performed for normal mucosa for comparative analysis of the vascular pattern. Confocal laser endomicroscopy (CLE) will be performed for visualization and characterization of vascular changes inside the tumor, using the probe-based CLE system (pCLE). The faster image acquisition rate of the pCLE system (12 frames per second) enables direct in vivo visualization of blood flow. The main advantage of the pCLE system is that it can be easily integrated into the imaging protocol as its flexible catheter probes can be passed through the working channels of the endoscopes once the vascular area of interest is selected with M-NBI. For pCLE examinations we will use fluorescein as a contrast agent which is already approved for in vivo human examinations. Fluorescein is administered intravenously, highlighting the vessels and surrounding epithelial structures. The vascular pattern will be assessed in real-time as well as off-site based on objective measurements of the stored sequences that will include different vascular parameters (vessel diameter, vascular density). These will be determined for both tumor tissue and normal adjacent mucosa as control, using dedicated software. Immunohistochemistry (IHC) analysis with MVD calculations will be assessed and correlated to the imaging findings. Paired biopsies of tumors and normal GI mucosa obtained during endoscopic procedures (ensuring co-registration with M-NBI and pCLE examined regions) will be processed for fixation in 10% neutral buffered formalin (NBF), paraffin embedding and sectioning. For MVD analysis, the blocks will be completely cut in serial sections 4-25 μm-thick, with one thick section for every 10 thin sections. In order to assess the total vascular architecture we will utilize non-selective endothelial cell markers, such as anti-CD31 and anti-CD34 antibodies.

Consecutive patients with gastric cancer will be examined using magnification endoscopy with narrow band imaging and probe based confocal laser endomicroscopy.

Consecutive patients with colorectal cancer will be examined using magnification endoscopy with narrow band imaging and probe based confocal laser endomicroscopy.

Magnification endoscopy with narrow band imaging is an optically enhanced endoscopic imaging technique used for better characterization of lesions as compared to white light endoscopy alone. With only a push of a button optical filters are applied to reduce the illuminating light to 415 nm (blue) and 540 nm (green) wavelengths, which are selectively absorbed by haemoglobin. As a result the superficial vascular networks are highlighted and morphological changes in capillary patterns can be described for different gastrointestinal lesions.

Confocal laser endomicroscopy enables in vivo microscopic analysis during ongoing endoscopy and has shown good accuracy for predicting the histopathological diagnosis in lesions of both the upper and lower GI tract. During the examination fluorescein is administered intravenously as a contrast agent, highlighting the vessels and surrounding epithelial structures. The probe based confocal laser endomicroscopy system can be easily integrated into the imaging protocol as it uses flexible catheter probes that can be passed through the working channels of the endoscopes once the area of interest has been identified.

Real-time imaging of angiogenesis in gastric and colorectal tumors by using M-NBI and pCLE examinations

After identification of the lesion with conventional white light endoscopy, endoscopic magnification with narrow-band imaging (M-NBI) will be performed for enhanced visualization of the capillary changes. A vascular region of interest will be selected in the M-NBI mode for targeted microscopic examination with probe based confocal laser endomicroscopy (pCLE) and tissue sampling for pathology and immunohistochemistry assessment. The vascular pattern will be assessed in real-time by the examiner as well as off-site based on objective measurements of the stored sequences that will include different vascular parameters (vessel diameter, vascular density). These will be determined for both tumors and normal adjacent mucosa as control, using dedicated image processing software.

Paired biopsies of GI tumors and normal mucosa obtained during endoscopic procedures (ensuring co-registration with M-NBI and pCLE examined areas) will be processed for immunohistochemical analysis by utilizing anti-CD31 and anti-CD34 antibodies as endothelial cell markers. Microvascular density (MVD) calculations will be performed by using the 'hot-spot' method and the results will be compared to the vascular parameters as assessed from M-NBI and pCLE stored images.

Inclusion Criteria: Patients with gastric and colorectal cancer Age 18 to 90 years old, men or women Signed informed consent for M-NBI and pCLE examinations with tissue sampling. Exclusion Criteria: Failure to provide informed consent Prior or ongoing chemo- and/or radiotherapy Patients with a contraindication for GI endoscopic procedures Known allergy to fluorescein

Fox SB, Harris AL. Histological quantitation of tumour angiogenesis. APMIS. 2004 Jul-Aug;112(7-8):413-30. doi: 10.1111/j.1600-0463.2004.apm11207-0803.x.

Aihara H, Saito S, Tajiri H. Rationale for and clinical benefits of colonoscopy with narrow band imaging: pathological prediction and colorectal screening. Int J Colorectal Dis. 2013 Jan;28(1):1-7. doi: 10.1007/s00384-012-1591-7. Epub 2012 Oct 9.

Hirata I, Nakagawa Y, Ohkubo M, Yahagi N, Yao K. Usefulness of magnifying narrow-band imaging endoscopy for the diagnosis of gastric and colorectal lesions. Digestion. 2012;85(2):74-9. doi: 10.1159/000334642. Epub 2012 Jan 19.

Gheonea DI, Cartana T, Ciurea T, Popescu C, Badarau A, Saftoiu A. Confocal laser endomicroscopy and immunoendoscopy for real-time assessment of vascularization in gastrointestinal malignancies. World J Gastroenterol. 2011 Jan 7;17(1):21-7. doi: 10.3748/wjg.v17.i1.21.

Sanduleanu S, Driessen A, Gomez-Garcia E, Hameeteman W, de Bruine A, Masclee A. In vivo diagnosis and classification of colorectal neoplasia by chromoendoscopy-guided confocal laser endomicroscopy. Clin Gastroenterol Hepatol. 2010 Apr;8(4):371-8. doi: 10.1016/j.cgh.2009.08.006. Epub 2009 Aug 13.

Bok GH, Jeon SR, Cho JY, Cho JH, Lee WC, Jin SY, Choi IH, Kim HG, Lee TH, Park EJ. The accuracy of probe-based confocal endomicroscopy versus conventional endoscopic biopsies for the diagnosis of superficial gastric neoplasia (with videos). Gastrointest Endosc. 2013 Jun;77(6):899-908. doi: 10.1016/j.gie.2013.01.018. Epub 2013 Mar 6.

Wang SF, Yang YS, Wei LX, Lu ZS, Guo MZ, Huang J, Peng LH, Sun G, Ling-Hu EQ, Meng JY. Diagnosis of gastric intraepithelial neoplasia by narrow-band imaging and confocal laser endomicroscopy. World J Gastroenterol. 2012 Sep 14;18(34):4771-80. doi: 10.3748/wjg.v18.i34.4771.