Our goal is to find the best ways to prevent pancreatic cancer deaths in patients with pancreatic cysts. Recent advances in imaging have led to the detection of innumerable pancreatic cysts that could never be seen before, now visible in >10% of patients who have an MRI and in >2% who have a CT scan for an unrelated reason. When such cysts are unexpected and asymptomatic, they are considered ?incidental.? The majority represent intraductal papillary mucinous neoplasms (IPMNs), which are indolent precursors to Pancreatic Ductal AdenoCarcinoma (PDAC), the most common type of pancreatic cancer. Given the poor prognosis and survival of patients with PDAC, pancreatic cysts have become a primary target of early PDAC detection. Imaging surveillance is advised for most patients who are diagnosed with an incidental pancreatic cyst, but key factors that define surveillance ? e.g., the frequency, modality, and duration of imaging, and when to pursue biopsy or surgery ? are highly controversial. Critics of intensive surveillance raise concerns about overtesting and overtreatment, particularly given that many patients with such cysts are older and have comorbidities. Advocates emphasize the singular opportunity for early PDAC detection that arises from close monitoring. To address this problem, we will build a computer-based simulation model that replicates the natural history of incidental pancreatic cysts, and use it to formulate a precision management approach. Our research plan will draw from our team?s existing simulation model of PDAC, which is calibrated to data from the National Cancer Institute?s Surveillance, Epidemiology, and End Results (SEER) Program and published studies. First, we will extend this model to replicate the natural history of incidentally detected pancreatic cysts (Aim 1). We will then use the model to identify effective (Aim 2) and cost-effective (Aim 3) management strategies that are tailored to both cyst features (size, complexity) and patient characteristics (age, comorbidity status). Finally, we will evaluate the potential for emerging blood and cyst-fluid biomarkers to further improve management (Aim 4). The proposed research is innovative because it applies an advanced modeling approach to a controversial problem that will be difficult to solve with observational studies or clinical trials alone. The research team is well-suited, with an established track record in pancreatic cancer care and incidental pancreatic cysts, and with substantial experience in developing mathematical models that have been used to inform health policy at national levels. The results will be threefold: 1) a detailed natural history model of incidental pancreatic cysts; 2) a tailored approach to their management, based on cyst features and patient characteristics; and 3) a roadmap for advancing future research in cystic precursors to pancreatic cancer in the coming years.
Recent advances in imaging have led to the detection of innumerable pancreatic cysts that could never be seen before, now visible in >10% of patients who have an MRI and in >2% who have a CT scan for an unrelated reason. The majority of these cysts represent low-risk precursors of pancreatic cancer, a small percentage of which will progress to invasive cancer over months to years. In the proposed research, we will build a mathematical model that simulates the natural history of pancreatic cysts, and use it to formulate a management approach that optimally balances the benefits of cancer control with the risks and costs associated with overtesting and overtreatment.
