NEW YORK (360Dx) – A decades-old biomarker that detects about 40 percent of pancreatic cancers could play a more effective role in screening high-risk patients for the disease when it is combined with a glycan biomarker, sTRA, according to a research group developing the test.
Van Andel Research Institute is leading the development of the new approach that in a recent study detected nearly 70 percent of pancreatic cancers with a less than 5 percent false-positive rate.
The researchers published the results of their blinded study this month in Clinical Cancer Research. In it, they combined sTRA with the established CA19-9 biomarker to detect pancreatic ductal adenocarcinoma (PDAC), the most common malignancy of the pancreas. The researchers developed the combination in a training set of plasma samples and used the panels to determine results based on predetermined thresholds in a 50-sample validation set and a blinded 147-sample test set.
In addition to researchers at Van Andel, the group consists of scientists and clinicians at Michigan State University, Fred Hutchinson Cancer Research Center, University of Pittsburgh Medical Center, Medical University of South Carolina, and MD Anderson Cancer Center.
It's clear that subtypes of pancreatic cancer produce and secrete complimentary antigens, Brian Haab, a professor in the Center for Cancer and Cell Biology at Van Andel and the study's senior author, said in an interview. "This new marker detects a different complementary subgroup of pancreatic cancers and by using the biomarkers together, doctors can get better information that they can use to detect a broader range of pancreatic cancers with high specificity to make a more accurate diagnosis," he said.
Pancreatic cancer is devastating because of its tendency to spread before detection, and doctors have trouble diagnosing and treating it before the disease spreads. Only 8.5 percent of people with the condition survive past five years. By the time clinicians detect the disease, it has usually advanced, complicating treatment options and resulting in poorer outcomes.
Cesar Castro, director of the cancer program within the Massachusetts General Hospital Center for Systems Biology, called the CCR findings "welcome news. Historically, CA 19-9 has underperformed as a diagnostic marker and is more often used as a crude gauge of treatment response."
Despite recent inroads in cancer therapies, "PDAC continues to stymie oncologists’ best efforts and remains a highly lethal disease," Castro said. The most effective chemotherapy regimen creates significant toxicities and prolongs median survival from six months to 11 months after diagnosis.
Castro, who is not involved in the Van Andel-led research, is developing a multiplexed nanoplasmonic assay to fit into clinical workflows and enable high-throughput detection of PDAC.
Biomarker diagnostics
CA 19-9 has been available for decades, but it is not used for screening among high-risk patients, Haab said. The new approach to pancreatic cancer screening among high-risk patients uses a simple blood test and a choice of two or three assays. These assays detect and measure levels of sugars produced by pancreatic cancer cells that subsequently escape into the bloodstream. Sandwich immunoassays employ separate capture and detection antibodies combined with fluorescent or chemiluminescent detection.
The developers anticipate continuing with its validation as a research-use-only test in clinical studies lasting one or two years and, all going according to plan, launching it as a laboratory-developed test in a CLIA-certified CAP-accredited laboratory. After further clinical testing and validation of the assays, the researchers anticipate licensing the test to a diagnostics company that would seek US Food and Drug Administration clearance. The timeline for an FDA-cleared test is uncertain, Haab said.
For diagnosing high-risk patients, Castro said, accuracy is especially important. "In the absence of a pancreatic mass, the risk/benefit calculation for pursuing invasive biopsies tilts toward risk," Castro said. "If the biomarker combination performs well during prospective testing, integrating them into existing or emerging diagnostic approaches could improve uptake."
For example, a simplified fingerstick assay connected to wearable or smartphone devices could enable serial analyses of the signature, he noted. Any predetermined rate of increase could trigger notifications to one’s provider for further testing. Such dynamic, rather than static, evaluations of biomarkers would better reflect true biological trends rather than stochastic changes.  Â
The new test, if successful, could be used in a couple of different clinical diagnostic applications, Haab said. First, physicians could use it when they are having difficulty differentiating a benign from a cancerous state. "The new approach does a far better job of differentiating cancer from non-cancer," he said.
Second, the test could be useful to screen for the potential of pancreatic cancer in patients that are at high risk but are not experiencing symptoms. That includes people who have a family history of pancreatic cancer, who have had pancreatic cysts or chronic pancreatitis, or who were diagnosed with type 2 diabetes later in life, Haab said.
"The performance of the test is approaching a point where it justifies application to that group of people," he said. In this circumstance, clinicians will need to confirm a diagnosis by following up on a positive test with imaging of the pancreas, he said.
Further, the group has done follow-up research yet to be published which shows that patient groups that express sTRA as opposed to CA 19-9 are biologically different, and therefore the combined biomarkers could, in the long term, be used for cancer subtyping and guidance related to suitable treatments, Haab said.
Diagnostics R&D
Other research groups and companies are exploring use of new technology to address the need for early- and later-stage pancreatic cancer diagnosis.
On Wednesday, Lustgarten Foundation announced it was opening a fourth dedicated pancreatic cancer research laboratory at Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins in Baltimore, Maryland. The laboratory at Johns Hopkins University will join three other Lustgarten Foundation dedicated pancreatic cancer research laboratories, which include Cold Spring Harbor Laboratory, Dana-Farber Cancer Institute, and Massachusetts Institute of Technology.
According to Lustgarten, its laboratory at Johns Hopkins will leverage its expertise in early detection to intercept pancreatic cancer at an earlier stage when patients may be surgical candidates, and it will aim to develop novel therapeutic approaches to treat pancreatic cancer based on genetic alterations.
Immunovia said that a study using a 29-biomarker signature and its IMMray PanCan-d assay indicates the test could potentially be used to detect early-stage pancreatic cancer.
Adaptive Biotechnologies recently reported that it plans to launch various immune sequencing-based diagnostic tests, including for pancreatic cancer.
Meanwhile, BioFluidica has begun three different clinical studies to detect exosomes, cell-free DNA, and rare cells in patients' bloodstreams to identify acute lymphoblastic leukemia and other cancers, including pancreatic cancer.
As a next step, the Van Andel-led group will continue to offer the test as a research-use-only product to its clinician partners. It is working with an undisclosed diagnostic laboratory to obtain the necessary approvals to offer it as a laboratory-developed test, Haab said. When the laboratory is up and running, the researchers' clinical partners will start sending samples so that data can be quickly collected and analyzed, and the test further validated.
As part of its commercial plan, the group anticipates publishing results based on prospective trials, which could take at least a couple of years, he said. If the results are positive, the developers could start the process to get payor reimbursement for the test. The group has performed early prospective studies with regional hospitals and achieved promising results, Haab noted.