Pancreatic cancer screening for high risk individuals with familial or genetic susceptibility is cost-effective and potentially lifesaving, according to an economic analysis published in the April issue of Pancreas. A multi-institutional international initiative led by researchers from the Mayo Clinic in Jacksonville, FL, determined that any person with a lifetime risk of more than 5% or a five-fold relative risk of developing pancreatic cancer should have abdominal imaging exams starting at the age of 40, with magnetic resonance imaging (MRI) recommended as the most cost-effective strategy.1
Published reviews of pancreatic cancer screening studies have shown that both MRI and endoscopic ultrasound (EUS) can detect early pancreatic cancer or dysplastic lesions.2,3 International consensus guidelines developed in 2011 by the International Cancer of the Pancreas Screening (CAPS) Consortium recommend screening with EUS and MRI every 6 to 12 months for patients diagnosed as having stable pancreatic lesions.4
Physicians led by Michael B. Wallace, MD, of the Division of Gastroenterology and Hepatology, performed an economic analysis using a Markov model to identify different clinical and cost determinants of pancreatic cancer surveillance in high risk individuals. They used data from the Surveillance, Epidemiology, and End Results (SEER) and Medicare databases to determine risks for developing pancreatic cancer, survival, and costs, applying these to three strategies that included no screening, EUS screening, and MRI screening for an asymptomatic high-risk, 40-year-old patient.
The Markov model transitioned the patient from being healthy, to developing a low-risk lesion, then a high-risk lesion, followed by pancreatic cancer within 12 months, post-pancreatectomy diabetes, and death. Surveillance continued until the lifetime horizon of the cohort because no consensus recommendation has been established on when to end surveillance for high-risk individuals without pancreatic lesions. The model also assumed that EUS-guided fine-needle aspiration would be performed on low-risk lesions; if the biopsy were positive, a pancreatectomy would be performed. If negative, EUS would be repeated in six months. A pancreatectomy would be performed for any patient identified with a high-risk lesion.
The authors calculated direct costs based on the 2017 U.S. dollar, and made the assumption that cost-effectiveness would be limited to a $100,000 threshold. They compared this with up to two years of treatment costs for pancreatic cancer.
They determined that screening every three years, with MRI for lesser-risk individuals and with EUS only for individuals with a 20-fold risk or higher, was more cost-effective than annual screening. Screening was most cost-effective up to the age of 75. They also determined that most conditions should be screened with MRI, with EUS being used only on patients with Peutz-Jeghers syndrome, hereditary pancreatitis, and familial pancreatic cancer.
Monte Carlo simulation determined that in the absence of screening, 443 patients out of a hypothetical population of 10,000 would develop pancreatic cancer, compared to 130 patients who regularly underwent MRI screening and 127 patients EUS screening. Only 12% of the no-screening cohort had resectable cancer, compared to 40% and 51% respectively of the two screening groups.
The authors estimated that 32 high-risk individuals would need to be screened to detect one high-risk lesion. They noted that the United States Prevention Services Task force recommends against surveillance for pancreatic cancer in the general population, but does not address the high-risk population of their study.
Pancreatic cancer screening with abdominal imaging is cost-effective. Appl Radiol.