A simple and cost-effective blood test that targets two complementary biomarkers has demonstrated its potential to screen patients at high risk for pancreatic ductal adenocarcinoma (PDAC), the most common and lethal form of pancreatic cancer.
In a meta-analysis that included 729 patient samples from four different clinical centers, the new blood test showed significant improvement over CA19-9, the current biomarker used to assess disease progression in pancreatic cancer.
"If further validated using clinical assays, [the new assay] may be good enough for surveillance among patients with increased risk for PDAC, in combination with follow-up modalities," said Ying Liu, PhD, a research associate in the laboratory of Brian Haab, PhD, at the Van Andel Institute in Grand Rapids, Mich., who presented the results of the study to a virtual meeting on pancreatic cancer held September 29-30 by the American Association for Cancer Research.
As outlined in Ying's talk, the test may one day provide a way to detect PDAC early before it has a chance to spread to other parts of the body. Because the disease doesn't display any obvious symptoms in its early stages, it's often not detected until it already has spread to other parts of the body. Treatment options are currently limited, resulting in a low 5-year survival rate of about 9 percent.
In particular, the test may be used to screen at-risk individuals, including those with a family history of pancreatic cancer, who've experienced pancreatic cysts or chronic pancreatitis, and those who have been diagnosed with type 2 diabetes later in life, a growing problem.
"The goal of our study is trying to identify the patient with cancer in its early stage," Ying said in response to a question following her presentation. "That means we can identify more cancer patients with stage I and II disease," she said.
Added Anirban Maitra, MBBS, Scientific Director of the Pancreatic Cancer Research Center and Co-Leader of the Pancreatic Cancer Moon Shot at the University of Texas MD Anderson Cancer Center: "I think the first thing that we really need to do is to establish the validation of these findings. I mean these are fantastic findings but ultimately how do we get these to the lab and how do we apply these to our patients. That is the real intent." Maitra served on a panel asked to comment on this and other studies presented at the AACR meeting.
Years of Research
The new study supports previous work that demonstrated the improved performance of the combination assay over the FDA-approved serologic test for pancreatic cancer using CA19-9 (carbohydrate antigen 19-9) as a biomarker. A molecular feature shared by most PDAC patients is increased levels of the CA19-9 antigen.
The CA19-9 assay, developed more than 4 decades ago, detects only about 40 percent of pancreatic cancers, and displays about a 25 percent false-positive rate among patients with benign conditions of the pancreas. Given its limitations, the test is not reliable enough to detect or diagnose disease and is used primarily in the clinic to track disease progression.
Despite its shortcomings, no biomarker has been able to replace this standard test since it first entered practice, although many labs have been working to develop alternatives, some showing promise.
The new Van Andel assay, described in the AACR talk, not only measures CA19-9, but also another sugar called sTRA (sialylated tumor-related antigen), which is produced by a different subset of pancreatic cancers than CA19-9.
Studies in the Van Andel lab previously demonstrated that these two biomarkers are complementary, with some pancreatic tumors detected through CA19-9 only, and others through sTRA. When the two are combined in the same assay, the team hypothesized that the test would detect subtypes of pancreatic cancer that may have been missed by using one of the two tests on its own.
In an article published last January 2019 in the journal Clinical Cancer Research, the Van Andel team showed that, in a double-blind study, their new assay-which combined both CA19-9 and sTRA biomarkers-detected nearly 70 percent of pancreatic cancers with a less than 5 percent false-positive rate (2019: doi: 10.1158/1078-0432.CCR-18-3310).
"This is the first biomarker, to our knowledge, to statistically significantly improve upon CA19-9 in a double-blinded test with preset thresholds and classification rules," the researchers wrote in their article.
To validate these results, the researchers partnered with other clinical laboratories, using a meta-analysis that included six sets of 729 blood samples from the University of Pittsburgh, Mercy Health Saint Mary's in Grand Rapids, and the Medical University of South Carolina in Charleston.
Here, the researchers pitted the CA19-9 test against a two-marker panel composed of CA19-9 and sTRA and a three-marker panel composed of CA19-9 and two versions of the sTRA immunoassay. The goal was to detect pancreatic cancer among samples that included PDAC patients, healthy controls and patients with benign conditions.
The results showed that at 95 percent specificity, the sensitivity was improved from 36-55 percent across all datasets for CA19-9 alone to 61-67 percent across datasets for the three-marker panel.
In a blinded multi-lab study with the same specificity threshold, the researchers tested their assay in sera collected from 340 patients. The results showed 95 percent specificity and 61 percent sensitivity with the three-panel assay.
"Though we are still blinded in terms [that] we don't know which patient is a control, or which patient is a case, we do know by keeping the same specificity the panel achieved better sensitivity than CA19-9 alone," Yin said.
"So, both the previous study and the meta-analysis show that the new marker sTRA developed in our lab is complementary to CA19-9, and it has been validated to have a better performance than CA19-9 using the blinded study."
Warren Froelich is a contributing writer