CE-IOUS for Patients With Pancreatic Cancer Undergoing Surgery: A Prospective, Proof-of-concept Trial

Overview

BACKGROUND: Pancreatic Cancer (PC) is one of the leading cancer-related causes of death worldwide, with the majority of patients undergoing potentially curative surgery. In this context, the implementation of an accurate imaging modality is crucial in order to facilitate the clinical decision-making on the basis of tumor resectability. The contrast enhanced intraoperative ultrasound (CE-IOUS) is a relatively new imaging modality that has been employed in the detection mainly of colorectal liver metastases, but not for those of pancreatic origin. AIM: The purpose of the present study is to validate the CE-IOUS in adult patients undergoing pancreatic surgery. METHODS: Prospective single-center analysis of all consecutive patients with PC undergoing pancreatic surgery from a single hepato-pancreato-biliary (HPB) surgery team between December 1st, 2020 and December 31st, 2022 will be performed. Baseline characteristics, type of surgery, intraoperative parameters, hospital length of stay (LOS), intensive care unit (ICU) stay, postoperative morbidity and 30-day mortality data will be obtained from the database. The primary outcome is the clinical utility, defining its ability to change surgical operation on the basis of its findings. LIMITATIONS: The key limitation is the inclusion of only one HPB surgery team from one center. STRENGTHS: This study will potentially be the first to evaluate EC-IOUS and to compare it with the IOUS, CT and MRI for pancreatic surgical patients.

1\. Introduction Pancreatic cancer is one of the most lethal types of cancer. In fact, is usually detected in an advanced stage, thus being associated with a poor prognosis. Therefore, every surgical intervention has to be precise. Furthermore, the staging of a tumor indicates the resectability of the tumor and the optimal extent of the surgery. In this context, spiral computer tomography (CT) and magnetic resonance tomography (MRI) are employed preoperatively to identify liver metastases but they demonstrate lower sensitivity and specificity than intraoperative ultrasound examination (IOUS) in a number of settings. IOUS was applied in pancreatic tumors during 1980s for the first time. At liver metastases detection IOUS significantly exceeds CT or MRI in sensitivity, reaching 94-96 % even at small focci with 3-5 mm dimensions. According to Kolesnik, 2015 study IOUS is a convenient method to use and demonstrates superiority over CT and MRI, such as obtaining dynamic image with possibility of correction and new data acquisition at any moment of operation. Recently, there is a growing trend of implementing contrast-enhanced IOUS (CE-IOUS) in order to identify potential liver metastases intraoperatively. Nonetheless, no real clinical studies on CE-IOUS have been published to evaluate its efficacy in pancreatic surgery. In fact, with CE-IOUS all small vessels and capillaries can be visualized using microbubble-based ultrasound contrast agents. This could benefit by dynamic real-time imaging modality for pancreatic metastases with abnormal vascularization, because isoechoic lesions are underappreciated on IOUS. Additionally, neoadjuvant chemotherapy could induce steatosis (which impairs ultrasound imaging) and cause a certain number of metastases to vanish posing diagnostic and therapeutic challenges, that require the surgeon to adapt the surgical procedure accordingly. To evaluate the clinical utility of CE-IOUS, we will conduct the first, to the best of our knowledge, prospective trial to evaluate the efficacy and safety of CE-IOUS in the context of pancreatic surgery. 3\. Material and methods Following approval by the Ethics Committee, the present study will be performed in the University Hospital of Larissa, Greece. Our primary endpoint is to assess the clinical utility (CU) of CE-IOUS regarding the identification of liver metastases in patients undergoing curative surgery for pancreatic cancer. CU is defined as the justified change in planned surgical strategy (conversion from therapeutic to palliative surgery) using CE-IOUS, based on the identification of new liver metastases that were not described by preoperative imaging modalities. Our secondary objectives are to outline technical modalities of CE-IOUS regarding the dosage and duration of contrast agent administration, compare CE-IOUS with CT, MRI, and IOUS, and to assess the CU of CE-IOUS in the subgroup of patients with missing metastases after neoadjuvant treatment. Patients Patients aged \> 18 years with resectable, borderline resectable or locally advanced+neoadjuvant-treatment pancreatic cancer undergoing surgery between December 2020 and December 2022 will be included. All patients will sign an informed consent. Patients allergic to contrast solution, patients with recent acute coronary syndrome or those with unstable ischemic heart disease, or severe arrhythmias, or pregnant or breastfeeding women, will be excluded. Preoperative staging For patients undergoing surgery for pancreatic cancer, the evaluation will include a CT, liver MRI (in case of neoadjuvant treatment), EUS in certain cases, CA19-9, biochemical assessment within 8 weeks. This evaluation will be validated by the multidisciplinary committee. Intraoperative ultrasound: conventional and contrast-enhanced We will use contrast agent which remains in the vascular compartment. The dual-screen mode allows checking whether the lesion is visible, suspected with difficulty or completely undetectable by conventional ultrasound. Following a laparotomy and liver mobilization, conventional IOUS will be performed to confirm the preoperative tumor staging and to search for new occult liver nodules. After IOUS, a bolus of perflubutane suspension will be injected through a peripheral vein at a microbubble doses of 1.5-4 μL. A systematic ultrasonic liver screening will then be performed using the harmonic mode at 10-15 min after contrast injection, with the focus point set at the bottom of the liver (CE-IOUS). Any hypoechoic liver nodule that was visible during CE-IOUS was considered to be a liver metastasis, except for the nodules that were diagnosed as cyst or hemangioma using the preoperative CT and MRI. All the IOUS and CE-IOUS procedures will be performed by a Resident Radiologist, supervised by an experienced Radiologist. Visualisation of lesions Following the contrast agent injection, healthy liver takes up the contrast uniformly while the metastases initially appear as a bright signal in an arterial phase ("wash in"). The metastases are then washed ("wash out"), which appear as gaps without contrast in venous phase. If the low signal disappears within 45 s, then the lesions are considered benign. Conversely, the low signal of malignant lesions lasts beyond 45 s and over several minutes. Indeterminate lesions are those that do not precisely correspond to either of the two previous definitions. Study design The planned surgery decision will be recorded following conventional IOUS, as well as with CE-IOUS during surgery. For a given patient, we will conclude that the CE-IOUS had CU if the two planned decisions are different and if the modification is justified. A justification could be: 1. pathological findings following CE-IOUS confirmed malignancy in case of larger resection, 2. pathological findings following CE-IOUS confirmed absence of malignancy in case of smaller resection in a given site that would have been treated without CE-IOUS, 3. for absence of liver resection, pathological exam of biopsy confirmed the absence or presence of malignancy of the lesions that were not removed, or 4. If biopsy is not feasible or interpretable, the justification of the surgical procedure has to be confirmed with the 3-month radiological follow-up (in the absence of post-surgery chemotherapy; otherwise in case of chemotherapy within 3 months, we considered that lesions were malignant). Statistical analysis The results will be analyzed using GraphPad Prism 8.0 for Mac (GraphPad Software, San Diego, CA). Normal distribution of the data will be performed by application of the D'Agostino and Pearson Omnibus normality test. Comparisons of continuous variables will be performed with two-tailed unpaired t-test for parametric data and Mann-Whitney U-test for nonparametric data. The categorical outcomes will be tabulated in 2x2 tables and will be assessed by employing the Chi square test. Differences will be deemed significant with a p ≤ 0.05.