Use of PET Imaging to Distinguish Malignant From Benign IPMN

• Positive and Negative Predictive Value of PET imaging for Identifying Malignant IPMN [ Time Frame: 1 month ] [ Designated as safety issue: No ]
  The primary outcome will be to determine the positive and negative predictive values of [18F]-FDG PET imaging for identifying malignant IPMN lesions in patients who are to undergo surgical resection. We will determine the mean SUV that would provide optimal positive predictive value for malignant IPMN. IPMN lesions will be classified categorically as benign (adenoma or borderline ) or malignant (in situ or invasive carcinoma) and PET imaging will be classified categorically as negative or positive, with focal FDG uptake corresponding to pancreatic lesion.
  
  

• Change in lesion characteristics to assess benefit of PET scans [ Time Frame: 1 month ] [ Designated as safety issue: No ]
  The secondary outcome that we are interested in studying is the benefit of PET scan as compared to other imaging modalities, such as CT, MRI, and EUS. Thus, Patients will also have CT, MRI, and EUS imaging. We will look at how size, location, and branch of the IPMN lesion on PET compare to these other imaging modalities. The location, size, and pathology results of the actual surgical specimen will serve as the gold standard.
  
  

Intraductal papillary mucinous neoplasm (IPMN) is a cystic pancreatic neoplasm that is a precursor to invasive pancreatic cancer. Differentiating whether an IPMN lesion is benign or malignant is critical, as the prognosis and management differs drastically, varying from surgical resection to observation. However, despite attempts to characterize features concerning for malignancy, it is difficult to determine the likelihood of malignancy with conventional imaging techniques, including CT, MRI, and EUS. An accurate, non-invasive test to identify malignant IPMN is needed.

The investigators' hypothesis is that [18F]-FDG PET may be a superior modality for differentiating between benign and malignant IPMN lesions. We are planning a prospective pilot study of ten consecutive patients with IPMN from the Columbia University Pancreas Center who are undergoing surgical resection for their disease. These patients will undergo [18F]-FDG PET imaging, as well as CT, MRI, and EUS as clinically indicated. All scans will be reviewed by two experienced nuclear medicine radiologists who will be blinded to the clinical characteristics of study patients and who will reach a consensus. Areas of focally increased [18F]-FDG intake will be identified. Side-by-side reading with CT scan will be performed to evaluate whether the increased [18F]-FDG uptake corresponds to a pancreatic lesion. Mean and maximal SUV values, as well as differences in intensity between the region of interest and the remaining pancreas, will be calculated.Surgical pathology will be utilized as the gold standard for histological determination. Standard post-operative histological interpretation of each IPMN lesion will be recorded, including size, duct involvement (main, side, or mixed), ductal dilatation, lesion location (head, neck, body, tail), and histologic grade (adenoma, borderline, carcinoma in situ, invasive carcinoma). In addition, any associated pancreatitis or any other non-IPMN neoplastic change will also be noted.

Using PET scan results and surgical pathology information, we will evaluate the positive and negative predictive values of [18F]-FDG PET for malignancy within IPMN lesions.