New blood test could identify early pancreatic cancer

View on the News

Encouraging results, but view with caution

The results of Kobayashi et al. are encouraging in many respects, but need to be viewed with caution. The first question is whether the findings are repeatable in other populations. If so, their test may prove to be more sensitive and specific than CA19-9 and CEA for smaller pancreatic cancers, especially in the setting of chronic pancreatitis. Replication in larger studies is important because this study was relatively small, with only nine cases of cancer stage 0-IIB. Second, is the assay more rapid, reliable, robust, and/or significantly less expensive than CA19-9 and CEA? These factors are important for physician and health care system acceptance. Third, this approach has not been adequately tested in other disease states, with the important exception of chronic pancreatitis. Fourth, the test was done in the serum of patients who were already diagnosed with pancreatic cancer by other methods. Even in this setting, the sensitivity and specificity were inferior to clinical suspicion and imaging techniques.

The study was not designed to evaluate this test for early detection of pancreatic cancer, and with the current sensitivity and specificity, this test will have little if any value in screening of otherwise healthy populations. Finally, this approach did identify specific metabolites that were different between pancreatic cancer, chronic pancreatitis, and controls, which may provide further insights into the metabolic derangements of pancreatic cancer. Thus, it is too early to embrace or reject this new approach in the average clinic, but it does provide new insights and perspectives – and is moving us in a positive direction.

David C. Whitcomb, M.D., Ph.D., is the Giant Eagle Professor of Cancer Genetics; professor of medicine, cell biology, and physiology and human genetics; and chief of the division of gastroenterology, hepatology, and nutrition at the University of Pittsburgh and University of Pittsburgh Medical Center. He reports no significant financial conflicts.


An investigational serum biomarker panel accurately identified potentially resectable cases of pancreatic cancer, opening up the possibility that a blood test could identify the cancer early and improve its dismal prognosis.

The panel measured four metabolites and had an overall sensitivity of 71% and a specificity of 78% for pancreatic cancer; the specificity for resectable tumors also approached 78%, reported Dr. Masaru Yoshida and his colleagues (Cancer Epidemiol. Biomarkers Prev. 2013 March 29 [doi: 10.1158/1055-9965.EPI-12-1033]).

The panel was more accurate than were any of the currently used biomarkers, including CA19-9 and CEA, said Dr. Yoshida, professor and chief of metabolomics research at Kobe University, Japan, and his colleagues.

"This novel diagnostic approach, which is safe and easy to apply as a screening method, is expected to improve the prognosis of patients with pancreatic cancer by detecting their cancers early, when still in a resectable and curable state," Dr. Yoshida said in a press statement.

The researchers created the panel based on a study of a cohort of 85 subjects: 43 with pancreatic cancer and 42 healthy controls. First, they pared down 113 potentially useful metabolites into a workable testing panel. Of the 45 candidate metabolites, four of them – xylitol, 1,5-anhydro-D-glucitol, histidine, and inositol – showed the highest variation between the healthy controls and the patients. In the training cohort, the panel of these four markers had a sensitivity of 86% and specificity of 88% for pancreatic cancer.

Next, the researchers evaluated a validation cohort of 42 cancer patients, 41 healthy controls, and 23 patients with chronic pancreatitis. In this cohort, with an area under the curve of 0.76, the panel’s sensitivity for pancreatic cancer was 71% and specificity, 78%. In contrast, the sensitivity of CA19-9 was 69% and specificity 86%; for CEA, those numbers were 36% and 80%, respectively.

In the subset of patients with resectable pancreatic cancer, the panel’s sensitivity was 78%, compared with 56% for CA19-9 and 44% for CEA.

The metabolic panel was also able to identify chronic pancreatitis, with a 17% rate of false positives, compared with a false-positive rate of 30% for CA19-9 and 44% for CEA.

A blood panel could address three key problems in the field of pancreatic cancer, Dr. Yoshida and his colleagues said. "The first is the difficulty of detecting pancreatic cancer early in resectable stages. A sensitivity and specificity of about 80% for resectable disease ... should be acceptable for clinical use, because most patients with resectable cancer have no symptoms, and blood examinations are useful tools for initial screening examinations."

The second problem is the difficulty in differentiating pancreatic cancer from chronic pancreatitis. "Many gastroenterologists follow up chronic pancreatitis with scheduled CT, magnetic resonance imaging (MRI), and endoscopic ultrasound scans (EUS); tumor marker tests; and endoscopic retrograde cholangiopancreatography (ERCP), but the initial malignant changes are frequently overlooked, and pancreatic tumors can rapidly become unresectable. In addition, unnecessary resections for benign inflammatory lesions are sometimes done because of false-positive results from CA19-9 and/or imaging examinations."

The third clinical problem is the risk of complications that result from pancreatic examinations. Serum samples are easy and safe to obtain, with a low risk of adverse events and, with this panel, a potentially good return of information. A blood test could also be a useful population screening tool, the researchers noted.

The metabolites probably reflect metabolic derangement that arises not only from focal tumorigenesis, but also from systemic reactions to pancreatic disease, Dr. Yoshida and his coinvestigators wrote. Pancreatic cancer causes significant decreases in some amino and fatty acids, probably because the tumor recruits these substances to aid its rapid cellular proliferation.

"Patients with pancreatic disease are also troubled by malnutrition because of pancreatic endocrine and exocrine insufficiency. Therefore, there is a possibility that the decreases in serum metabolite levels also reflect malnutrition."

Finally, they said, the decrease in 1,5-anhydro-D-glucitol indicates the presence of hyperglycemia and recent glycosuria. "These results suggest that glucose tolerance was impaired in these patients because of pancreatic insufficiency."

The authors reported no financial conflicts. The work was supported by grants administered by agencies of the Japanese government.

Next Article: