There may be pros and cons for screening people who do not have symptoms of pancreatic cancer, but the evidence clearly supports screening those at a high risk for developing the disease, Vivek Kaul, MD, told his audience at the 2018 American College of Gastroenterology Annual Meeting.
There may be pros and cons for screening people who do not have symptoms of pancreatic cancer, but the evidence clearly supports screening those at a high risk for developing the disease, Vivek Kaul, MD, told his audience at the 2018 American College of Gastroenterology Annual Meeting in Philadelphia, Pennsylvania.
In 2017, there were 53,670 pancreatic cancer diagnoses and 43,090 deaths in the United States. Kaul, the Segal-Watson Professor of Medicine and chief of the Division of Gastroenterology and Hepatology at the University of Rochester Medical Center in New York, said that during his career, the frequency of new cases rose from an estimated 37,000 in 2008.1The disease now has a case fatality rate of 80%, and the 5-year survival is just 8%. “Almost everyone who gets it dies from it,” he said. “That’s a pretty morbid thought. Even those patients who undergo surgery with curative intent are not alive.”
Data show that individuals with a family history of pancreatic cancer are at an elevated risk of developing the disease, and high-risk individuals are often very motivated to be screened. Endoscopic ultrasound (EUS) and magnetic resonance imaging are effective screening modalities, especially when used in combination. The tests are not interchangeable, Kaul added.
Further, evidence suggests major gene involvement, and results from screening studies have uncovered precursor lesions. In 2011, Emmy Ludwig, MD, a gastroenterologist at Memorial Sloan Kettering (MSK) Cancer Center in New York, New York, and colleagues published results of a study exploring the feasibility and yield of screening in relatives of patients with pancreatic cancer.2
Ludwig et al enrolled 309 asymptomatic at-risk relatives into MSK’s Familial Pancreatic Tumor Registry. Participants were offered screening with magnetic resonance cholangiopancreato-graphy (MRCP), followed by EUS with fine needle aspiration if indicated. At the August 1, 2009, data cutoff, 109 people had been screened.
Eighteen patients had abnormal radiographic results. EUS confirmed a significant abnormality in 9 of 15 participants, 6 of whom underwent surgery.
The overall diagnostic yield was 8.3% (9 of 109). Yield was highest in relatives aged >65 years (35%) compared with those aged 55 to 65 years (3%) and <55 years (3% [2 of 61]).
“The greatest yield was in the seventh and eighth decade,” Kaul said. “The older patients with genetic abnormalities and family history were found to be significantly at risk.”
Prevention and/or early detection offer the only chance for cure, he said: “Clearly, this is a cancer, [and] intuitively, if we find it at an earlier stage, we could make a difference.”
The case for screening is not so easily made, however. Not all intraductal papillary mucinous neoplasms (IPMNs) or pancreatic intraepithelial neoplasia 1 lesions become carcinomas, which means patients are at risk of overtreatment. No national organization has issued formal screening guidelines, and there is no expert consensus on its value. Moreover, screening presents its own costs and potential complications and, so far, no study has shown that it is associated with improved survival.
“Does screening improve survival?” Kaul asked. “I don’t think we have a randomized, clinical, controlled trial and won’t have one for some time, but I think the most recent data support that if you [select] these people carefully, you’ll probably come out ahead. Remember that colonoscopy took about 35 years before we found that survival is [affected].”
Moreover, not all studies have demonstrated a high yield for neoplasia. Kaul cited results from a study conducted by Langer et al and published in 2009 in Gut that found a single IPMN among 76 patients with a history of familial pancreatic cancer (FPC).3
Generally defined as ≥2 first degree relatives with pancreatic cancer, FPC accounts for 4% to 10% of diagnoses.4The disease increases the risk of pancreatic cancer 18- to 57-fold.
Investigators evaluated high-risk patients in Germany from June 2002 to December 2007. Screening included EUS, MR angiography, and MRCP. EUS detected abnormalities in 25 patients, and MR/MRCP detected abnormalities in 12. Seven patients underwent fine needle aspiration cytology, and 7 underwent pancreatic explorations, 6 of whom proceeded to limited resections.
Langer et al concluded that screening could identify potential precursor pancreatic cancer lesions, but the yield for significant lesions was just 0.76%. “Taking into account the enormous psychological stress for the tested individual and the high costs, a general PC screening in high-risk individuals is not justified,” investigators wrote.
General screening might be unnecessary and infeasible, but Kaul said that high-risk populations should be screened. These include patients with hereditary chronic pancreatitis and those with the hereditary tumor syndrome Peutz-Jeghers, familial atypical multiple mole melanoma, hereditary nonpolyposis colorectal cancer (HNPCC), and FPC.
Kaul uses the 2018 International Cancer of the Pancreas Consortium consensus document on a daily basis when choosing whom to screen, he said. The guidelines recommend screening for first-degree relatives of patients with pancreatic cancer, those with a history of FPC, patients with Peutz-Jeghers syndrome, as well as p16, BRCA2, and HNPCC mutation carriers with at least 1 affected first-degree relative. Kaul said that the BRCA2 mutation is the most frequent genetic defect in FPC and emphasized that only patients who are fit for surgery should be screened.
“We know that there are very, very high-risk individual groups who definitely should be eligible for screening,” Kaul said. “There is a guideline in place to support that.”
He added that the field needs more robust survival data and updated international consensus guidelines.