EUS-guided Through-the-needle Microforceps Biopsy Outcomes

Recently, a EUS-guided through-the-needle microforceps device (Moray Micro Forceps, US Endoscopy, Mentor, Ohio, United States), witch can be passed through the lumen of a 19-gauge FNA needle, has been introduced as a novel approach for PCLs tissue acquisition. It has been shown that this device can drastically changed the diagnosis, providing diagnoses otherwise not suggested by cytology or cyst fluid CEA levels. The possibility of obtaining histological specimens from the cyst wall with this microforceps may change the management os PCLs, by obtening enough material for supplementary immunohistochemical analysis, allowing for even more accurate histological diagnosis, and in the case of IPMNs, possibly histological subtyping. The aims of this study is to assess: 1) the increment in diagnostic yield, by using this device, when compared to standard evaluation with cross-sectional imaging, EUS-FNA with cyst fluid analysis, 2) to evaluate the technical success, clinical success and safety, 3) Finally inter-observer accuracy in the assessment of PCLs using the different methods will be mesure, by the participation of all the endosonographers of Argentina, in the context of the annual EUS meeting.

• MATERIALS AND METHODS Study design and Setting: It is multicenter, prospective, interventional, non-controlled or randomized and double blind study, performed at the Institute of Gastroenterology and Advanced Endoscopy - IGEA (Bahía Blanca, Argentina), in conjunction with five high volume centers of all over Argentina (La Plata, Buenos Aires x 2, Neuquen and Mendoza), with patients included from July 2019 to July 2020. The study protocol and consent form has been approved by the institutional review board, will be registered at ClinicalTrials.gov and will be conducted according to the declaration of Helsinki. Written informed consent will be obtained from all subjects. No financial support will be received from the manufacturer of the microforceps, and none of the authors have any conflicts of interest. Data collection: Patient demographics, clinical history, cross-sectional imaging and EUS cyst findings like number of cysts, location, size and features (unilocular/septated, mural nodules, thickened wall, communication with pancreatic duct, dilated pancreatic duct and solid component) will be review. Fluid characteristics (viscosity, color, volume), amylase, glucose, CEA level and citology will be analyzed. The CEA level will be classified as <5, 6-192, >192 ng/mL and glucose level as > 50 and ≤ 50 mg/dl. ( ) EUS-FNA type of needle used, biopsy route (stomach/duodenum) and number of passes will be recorded as well as TTNB number of biopsies. Adverse events, technical and clinical success of EUS-FNA and TTNB will be recorded. Results of both procedures will be expressed in terms of cyst histotype, grade of dysplasia and mucinous vs. nonmucinous. Finally surgery result and follow up (12 month) will be used as gold standards. Follow-up data will be obtained from clinical encounters and/or telephone calls. Adverse events will be recorded per published American Society for Gastrointestinal Endoscopy criteria. Intervention: EUS-TTNB procedure technique. All procedures will be performed in a hospital-based interventional endoscopy suite, under intravenous sedation in a standardized manner and under antibiotics prophylactic. In cases of intracystic bleeding or impossibility of emptying the cyst, antibiotic therapy will be prolonged for 5 days. An endosonographer of each institution will perform the EUS procedures, but all the endosonographers of Argentina will analyze the results. EUS procedures will be performed with a linear echo-endoscope (EG-3870 UTK - Pentax, Tokyo, Japan; Hitachi Medical Systems, Tokyo, Japan). Cysts morphology will be recorded as was previously described and then punctured under Doppler guidance by using a 19-gauge EUS-FNA needle (Expect needle; Boston Scientific, Marlborough, Mass or EchoTip Ultra needle; Cook Medical, Bloomington, Ind). The stylet will be removed and the through-the-needle microbiopsy forceps (Moray; US Endoscopy, Mentor, Ohio, USA), with a sheath diameter of 0.8 mm and a jaw opening width of 4.3 mm, will be inserted through the needle into the cystic lumen. Microbiopsies will be obtained from the cyst wall randomly and from mural nodules or septae when observed. The open jaws of the forceps will be pushed gently onto the cyst wall, closed and pulled back to cause visible tenting. Two "bites" per pass of the microforceps will be obtained. Then the forceps will be removed, leaving the needle tip inside the cyst, the specimen will be placed in formalin vials and the forceps will be generously washed with normal saline solution before repeat biopsy. In the absence of intracystic bleeding, the procedure will be repeated 2 times, based on previous information with this device, or until 2 visible specimens are obtained. After completion of biopsies, the cyst fluid will be completely aspirated and sent for biochemistry (CEA, amylase, glucose) and cytology analyses, as was previously mentioned. If the cyst fluid is scarce, analysis of CEA concentration will be preferred over cytology. If a mural nodule is visualized on EUS, it will be accessed separately by using the 19-gauge needle for EUS-FNA cytology. No on-site cytopathological examination will be performed and all specimens will be evaluated by experienced gastrointestinal cytopathologists. Microbiopsies will be fixed immediately in 10% formalin solution and send to be stained with standard hematoxylin and eosin (H and E) and supplementary immunohistochemical (IHC) staining when indicated. After the procedure, patients will be observed for 2 hours in the recovery room before being discharged.

Definitions: Technical success will be defined as successful puncture of the pancreatic cyst with the 19-gauge needle under EUS guidance, advance the microforceps into the cyst, and successful tissue acquisition, that will be confirmed by gross visualization of the specimen from the jaws of the forceps. Clinical success will be defined as the ability to obtain a histopathological diagnosis of the PCL, by cytological or histological evaluation, using the microforceps biopsies. TTNB specimen diagnostic accuracy will be calculated at 3 different diagnostic levels: 1) Provide a specific diagnosis of the cyst histotype based on the characteristics of the epithelium and/or the cells composing the cyst wall. 2) Define the grade of dysplasia. 3) Differentiate mucinous versus nonmucinous cysts based on: The presence of a mucinous epithelium with cytoplasmic mucin and the presence of colloid-like extracellular mucin. The TTNB sampling diagnostic reliability (sensitivity and specificity) will be assessed by the evaluation of the correlation between the TTNB sampling results and surgical specimens (Gold Standard) in patients who underwent surgery or 12 months follow up in case of non-surgical PCL. Cysts with characteristics consistent with mucinous pancreatic cystic (mucinous epithelium with cytoplasmic mucin and/or colloid-like extracellular mucin) on FNA cytology, microforceps histology and/or surgical histology (when available) will be determined as mucinous pancreatic cystic. Otherwise, a final diagnosis will be determined, based on available cross-section imaging, EUS characterization, cyst fluid markers (CEA≥ 192 ng/mL and glucose ≤ 50) and stable appearance on 12 months follow-up. Cyst fluid CEA will be used to initially categorize the lesion as likely non-mucinous (CEA< 192 ng/mL), likely mucinous (CEA≥ 192 ng/mL) or inconclusive (between 6 and 191 ng/mL). Adverse events (AEs) will be defined and classified according to American Society for Gastrointestinal Endoscopy guidelines. Interobserver Agreement: It was mentioned before that diagnostic evaluation of PCLs can be challenging, requiring a combination of different methods, that many of them may have a low sensitivity and specificity, being dependent most of the time, on the subjective interpretation of the operator, which can be very difficult sometimes and experience dependent. The increment in diagnostic yield of a new method is appreciated as the result of the comparison of the new and previous methods accuracy. It is rational to understand that, if the method accuracy is experience dependent, this difference between methods accuracy will be less in more experience work groups. So when the increment in diagnostic yield of a new procedure is been analyzed it is important to ensure that the results can be apply to most of the practitioners (external validity/inter-observer agreement) and not only one group of work since it will only reflect the experience of that group (internal validity/ intra-observer agreement). For that outcome is that inter-observer agreement will be evaluated in the context of the Argentinian annual EUS meeting, where different cases will be presented to all the endosonographers of Argentina. Cases will be presented in 3 stages where a diagnostic assessment will have to be made in each stage. Initially (1st stage) patient's demographic data (sex, age), clinical history and images (CT, MRI, ultrasound) will be shown. Then (2nd stage) EUS images and videos will be presented and finally (3rd stage) the cyst fluid analysis results biochemistry (CEA, glucose, amylase) and cytology will be reviewed. Only 10 cases will be analyzed, during the meeting, due to time limitation, and agreement will be continue beeing evaluated using an online platform where cases will be presented using the same methodology. An external assistant to the protocol will analyze all the result and the different institutes will provide the cases, in a double blind manner, where neither the participating physician nor the operators will know the final results (TTNB or surgery).

Statistical analysis: Summary data will be expressed as mean +/- standard deviation (SD), or median and range for continuous variables and as number with percentage for categorical variables. Fisher o Chi-square test for categorical variables and the t test for continuous variables will be performed when indicated. The sensitivity, specificity, predictive values and accuracy of TTNB will be calculated with the 95% Confidence Interval (95% CI). To examine inter-observer agreement, kappa values will be calculated. Kappa coefficients below 0.4 indicate "poor agreement," values between 0.4 and 0.8 represent "moderate to good agreement," and values greater than 0.8 indicate "excellent agreement." A P value < 0.05 will be considered to be statistically significant. All the statistical analysis will be performed using SPSS software suite v.22.