Expanded blood test improves detection of pancreatic cancer

Researchers have reported that a blood-based panel combining four protein biomarkers has shown strong ability to detect early-stage pancreatic ductal adenocarcinoma, outperforming the commonly used CA19-9 test and offering a potential route to earlier diagnosis in high-risk patients

Pancreatic ductal adenocarcinoma, which arises from the exocrine cells of the pancreas, represents the most common form of pancreatic cancer and accounts for approximately 95 per cent of cases. Prognosis remains poor, largely because most patients receive a diagnosis at an advanced stage. When clinicians identify the disease while it remains localised, the five-year relative survival rate is about 44 per cent. Once metastasis has occurred, survival at five years falls to around 3 per cent. This stark contrast has driven sustained efforts to develop reliable tools for early detection.

“Pancreatic cancer usually doesn’t present with symptoms until it’s too late for surgery, when the cancer has already metastasised to other parts of the body,” said Zaret. “Our goal was to look for biomarkers in the blood that appear in early-stage pancreatic ductal adenocarcinoma patients, to catch the disease early,” he explained.

At present, the carbohydrate antigen CA19-9 serves as the most widely used blood biomarker for pancreatic cancer. Both malignant and normal pancreatic cells release this protein into the bloodstream, and elevated concentrations can indicate cancer or a range of benign conditions. Thrombospondin-2, known as THBS2, is expressed in pancreatic ductal adenocarcinoma tumour tissue and, when measured in blood alongside CA19-9, has previously shown improved discrimination between cancer and healthy individuals. Despite this, Dr. Kenneth S. Zaret, who led the study, and his colleagues have reported that these markers lack sufficient accuracy for routine early screening. Zaret works within the Department of Cell and Developmental Biology at the University of Pennsylvania, Philadelphia, Pennsylvania, USA.

“CA19-9 is widely used to monitor diagnosed pancreatic cancer but isn’t recommended as a standalone screening test – benign conditions can elevate it in some people, while others may have low levels, even if they have pancreatic cancer.

“THBS2 is investigational and can complement CA19-9, but its prediagnostic performance has been mixed,” Zaret said.

To identify additional biomarkers with greater sensitivity in early disease, the research team analysed plasma samples from two independent cohorts. These comprised 537 samples obtained through the Mayo Clinic and 135 samples from the Hospital of the University of Pennsylvania. Each cohort included patients with confirmed pancreatic cancer, individuals without cancer and patients with benign pancreatic disease. This design allowed the investigators to assess whether candidate markers could distinguish malignant disease from non-cancerous conditions that often resemble it clinically.

Comparative analysis of protein levels across these samples highlighted two proteins, alanyl aminopeptidase (ANPEP) and polymeric immunoglobulin receptor (PIGR) which appeared at elevated levels in patients with early-stage pancreatic ductal adenocarcinoma. On the basis of these findings, the researchers constructed a four-marker panel that combined ANPEP and PIGR with CA19-9 and THBS2.

Across both cohorts, the four-biomarker blood panel demonstrated strong diagnostic performance. When the researchers compared patients with stage one or stage two pancreatic ductal adenocarcinoma with healthy controls, the area under the curve reached 0.97 in the Mayo cohort and 0.96 in the Pennsylvania cohort. (The area under the curve represents a standard statistical measure of a test’s ability to discriminate between groups, where a value of 1.0 indicates perfect separation.)

The panel also distinguished cancer from benign pancreatic disease. In the Mayo cohort, the area under the curve reached 0.87 for early-stage disease and 0.91 when the analysis included all disease stages. Overall, the four-marker approach correctly identified 91.9 per cent of pancreatic cancers across all stages and 87.5 per cent of early-stage cases. By comparison, testing based on CA19-9 alone detected 82.7 per cent of cancers overall and 76.2 per cent of early-stage tumours. The improvement in detection across all stages reached statistical significance, while the increase in early-stage sensitivity did not, despite an absolute gain of 11.3 percentage points.

If future prospective studies confirm these results, the panel could strengthen clinical decision-making for individuals at elevated risk of pancreatic cancer. Zaret explained that a more accurate blood test could help clinicians decide which patients should proceed to imaging studies, thereby improving the chances of identifying tumours while they remain at a stage to be treatable.

“With the addition of ANPEP and PIGR, the panel helps to overcome known limitations associated with CA19-9 and THBS2 testing – such as patients who genetically under express CA19-9 or tumours that present as different molecular subtypes – and could therefore reduce the number of missed cancer cases while keeping false positives low,” he said.

The authors also highlighted important limitations, saying that the analysed cohorts did not include individuals with known high-risk features, such as a strong family history of pancreatic cancer, germline BRCA mutations or recent onset of diabetes. This absence may introduce bias into estimates of test performance. In addition, the retrospective design means that larger, prospective studies will be required to establish how the biomarker panel performs in real-world screening settings.

For further reading please visit: 10.1158/1078-0432.CCR-25-3297