NEW YORK – Researchers have identified a biomarker, pentraxin 3, which could help diagnose pancreatic cancer and differentiate it from noncancerous conditions, such as intra-ductal papillary mucinous neoplasm or chronic pancreatitis.
In a study published in NPJ Precision Oncology this week, researchers from Queen Mary University of London described their efforts to analyze PTX3 levels in serum blood samples from 267 patients, 140 of whom had been previously diagnosed with pancreatic cancer.
Pancreatic cancer patients had significantly higher serum PTX3 levels compared to patients with nonmalignant conditions of the pancreas, gallstone disease, and normal healthy volunteers. The PTX3 levels were also higher in pancreatic cancer patients than those with intra-ductal papillary mucinous neoplasm or chronic pancreatitis, two conditions that can often be confused with pancreatic cancer when screened using CT scans, the authors wrote.
At a PTX3 serum level cutoff of 4.35 ng/mL, the sensitivity for detecting the biomarker was 86 percent, the specificity was also 86 percent, and the positive predictive value was 97.5 percent when using western blot analysis. In comparison, another biomarker that can be used to diagnose pancreatic cancer, CA19-9, has a medium-range sensitivity of between 41 percent and 86 percent and specificity of between 33 percent and 100 percent. Additionally, the sensitivity of CEA biomarker, often used in clinical practice to diagnose pancreatic cancer, was 39 percent in previous research.
"Thus, amongst patients with similar clinical presentation where pancreatic cancer is being considered as differential diagnosis, serum PTX3 is better than clinically used tumor-marker CEA and is similar in performance to CA19-9, making a case for a prospective cohort study with matched data-points for formal comparison," the authors wrote.
The researchers conducted several tests on the serum samples, using ELISA, western blotting, immunofluorescence, immunohistochemistry, mRNA in situ hybridization, a mini-organotypic assay, a small interfering RNA test, and an MTS assay.
They further analyzed the origin of PTX3 in in vitro and ex vivo analyses. Those results suggested that PTX3 originates from stromal cells mainly in pancreatic stellate cells. Pancreatic cancer is often characterized by dense desmoplastic stroma that has been activated by the same pancreatic stellate cells. They also found that deactivating pancreatic stellate cells led to downregulation of PTX3 transcripts.
Based on that analysis, the researchers concluded that PTX3 could determine if a patient has desmoplasia, or the growth of fibrous tissue associated with cancer, in their pancreas, which in turn could help with diagnosis.
The researchers further analyzed data from a 2013 clinical trial and determined that PTX3 levels could not predict whether a person would live longer or were likely to progress on just chemotherapy.
However, they also analyzed PTX3 data from another study, called STARPAC, in which patients received chemo with a derivative of vitamin A, called all-trans retinoic acid, targeted to the stroma surrounding the pancreatic tumor. In that study, a significant increase in PTX3 serum levels immediately after treatment was predictive of patients who would see a partial response or stable disease. Patients who did not have the PTX3 increase after treatment were more likely to eventually have progressive disease.
Pancreatic cancer is often characterized by very few cancerous cells but a large number of stromal cells that have been activated by the cancer cells. Because PTX3 occurs in these activated stromal cells, the researchers believe it can be used to determine if a tumor is malignant and measure the efficacy of treatments that target stroma. However, they acknowledged that this conclusion is based on a small sample size and would need to be validated in a larger study.
Nevertheless, the study's suggestion that PTX3 can be useful for pancreatic cancer diagnosis is notable because of the limitations of CT scans, the most commonly used diagnostic tool for differentiating benign and malignant masses, Hemant Kocher, professor of liver and pancreas surgery at Queen Mary University of London and lead author of the study, said in a statement. CT scans can only detect the presence of a mass but cannot determine if the mass is cancerous unless physicians use positron emission tomography, or PET-CT, scanning.
"This poses frequent diagnostic dilemmas in clinical practice, and there are currently no clinically applicable biomarkers for the early detection of [pancreatic cancer]," Kocher said. "The findings from our study suggest that PTX3 could be used as a biomarker to improve [pancreatic cancer] diagnosis, and warrants further testing to determine whether it could aid early detection of [pancreatic cancer] in the clinic."