according to a recent study. However, screening individuals under the age of 50 years, even those with risk factors, is a low-yield strategy.
In a retrospective analysis of 86 asymptomatic adult patients with high-risk germline mutations followed at a single center, screening by a variety of imaging modalities showed a pancreatic abnormality in about one-quarter of patients (23 of 86, 26.7%). No cancers were detected on initial screening or during a median 29.8-month follow-up period, though the investigators saw more abnormalities in patients over the age of 60 years (P = .043).
The mutations were detected through genetic screening at the University of Texas MD Anderson Cancer Center, Houston, where the study was conducted.
“At our institution, based on oncology and/or medical genetics assessments, patients with a personal history of breast cancer and/or family history of [pancreatic cancer] are screened for BRCAs and BRCA1 mutations. Patients with other types of cancers in the family are screened for P53, STK11, MSH2, ATM, and APC mutations,” explained first author, of the division of gastroenterology, hepatology, and nutrition at the University of Texas, Houston, and his collaborators.
Patients were considered to have a family history of pancreatic cancer (PC) if any first-, second-, or third-degree relative had PC; overall, this amounted to 37 patients (43%). For the study, familial PC was defined as having either two first-degree or three or more relatives of any degree diagnosed with PC.
Patients, who were a median age of 48.5 years, were included in the study if they had any of the germline mutations, which are associated with an increase in relative risk of PC ranging from 2.26 for BRCA1 to 132 for STK11/LKB1, the mutation that causes Peutz-Jeghers syndrome. Patients who had a family history alone, without high-risk germline mutations, were excluded from the study. Most participants (79.8%) were women, and most of the participants were BRCA2 or BRCA1 positive (58.1% and 16.3%, respectively).
Of the abnormalities found, almost half (47.8%) were cysts, and almost as many (43.5%) were hyperechoic strands and foci. Two patients (8.7% of abnormalities) had mild pancreatic duct dilation. Patients who had only fatty infiltration or pancreas divisum were classified as having normal variations rather than abnormalities.
The pancreatic abnormalities had been found through endoscopic ultrasound (EUS), CT, or MRI, though the study’s primary aim was to look at outcomes and diagnostic yield for high-risk germline mutation patients receiving EUS. A secondary aim of the study, said the authors, was to compare EUS with both MRI and CT as methods to screen for early PC.
In this regard, Dr. DaVee and his coauthors wrote that “EUS-based screening conferred a higher yield in detecting abnormal pancreatic findings, compared with MRI [P = .007].” For MRI in comparison with EUS as the criterion standard, sensitivity and specificity were 55.6% and 93.8% for detecting pancreatic abnormalities. CT fared a little worse, with sensitivity of 50% and specificity of 89.5% for detecting pancreatic abnormalities.
Overall, the investigators noted that “the frequency of [pancreatic abnormalities] in our study is lower than reported in some studies using alternate criteria.” The relatively young age of the study cohort may be partly responsible for this finding, they said, adding that “our data further support evidence that abnormal pancreatic findings increase with age ... in asymptomatic [high-risk individuals], including the subtle finding of hyperechoic strands and foci.”
To improve survival, “premalignant and early PC identification is key,” wrote Dr. DaVee and his coauthors, pointing out that, of the 10% of patients diagnosed with PC when it is still localized, the 5-year survival rate is about 31.5%, compared with the overall 8.2% of individuals with PC who are still alive 5 years after their diagnosis. The latter figure, they noted, is a small improvement over the 3% 5-year survival figure from 1975.
Since population-wide screening for PC would not have a favorable cost-benefit profile, the challenge is identifying which individuals might benefit from targeted screening for PC. Dr. DaVee and his colleagues suggested that, as a general rule, high-risk individuals aged younger than 50 years need not be screened, and that EUS be used for the index screening, with MRI used to follow individuals with unremarkable initial screenings.
None of the authors reported relevant conflicts of interest, and no outside source of funding was reported.
SOURCE: DaVee RT et al. .