乌鸦传媒

A series of fecal microbiota-based markers predicted pancreatic ductal adenocarcinoma (PDAC) "with high accuracy and specificity," potentially making diagnosis at an earlier disease stage more feasible, an analysis found.

Based on a set of 27 microbial species, fecal metagenomic classifiers identified patients with PDAC with an accuracy of 0.84 area under the receiver operating characteristic curve (AUROC), performing much better than saliva-based classifiers, reported Peer Bork, PhD, of the European Molecular Biology Laboratory in Heidelberg, Germany, and colleagues.

When this panel of classifiers was combined with serum levels of CA 19-9 -- the only FDA-approved biomarker for PDAC -- it "significantly enhanced" the accuracy of PDAC detection, with performance improving to an AUROC of 0.94, they noted in .

"Fecal microbiome-based detection of PDAC may provide a non-invasive, cost-effective and robust approach to early PDAC diagnosis," Bork and team suggested.

Furthermore, these models showed comparable performance across all PDAC disease stages, indicating that "characteristic microbiome signatures emerge early during progression of the disease and that the fecal microbiome can serve for the early detection of PDAC," they wrote.

In a , Rachel Newsome, PhD, and Christian Jobin, PhD, of the University of Florida in Gainesville, noted that "there is an urgent need to identify novel and preferentially non-invasive early markers of pancreatic cancer," adding that this study represents an important contribution to the generation of predictive markers for PDAC, while highlighting the key role of microbiota in cancer surveillance.

However, "although promising, these findings have limited clinical value due to the cross-sectional nature of the study, and thus the predictive markers will need to be tested using a prospective cohort before reaching a conclusion on their clinical impact," they cautioned.

"[T]he described fecal microbiome signatures enabled robust metagenomic classifiers for PDAC detection at high disease specificity, complementary to existing markers, and with potential towards cost-effective PDAC screening and monitoring," Bork and colleagues concluded. "Furthermore, in view of previous reports on microbe-mediated pancreatic carcinogenesis in murine models and humans, we believe that the presented panel of PDAC-associated bacterial species may be relevant beyond their use for diagnosis, providing promising future entry points for disease prevention and therapeutic intervention."

Newsome and Jobin pointed out that a potential problem with these predictive models will be their ability to differentiate between cancer types.

"While [the investigators] can distinguish PDAC from breast or colorectal cancer microbial signature, their models need to also be tested against other cancer types to ensure that markers selected for PDAC are not shared," they wrote. But, overall, this study "brings clarity to the connection between microbiome signature and PDAC and represents significant progress for non-invasive cancer detection."

For this analysis, Bork and colleagues sampled saliva and feces, as well as normal and tumor tissue from the pancreas, from 136 patients at two Spanish hospitals -- 57 newly diagnosed with PDAC, 29 with chronic pancreatitis (a risk factor for pancreatic cancer), and 50 same-hospital inpatient controls.

Controlling for clinical and demographic variables such as age, obesity, diabetes, chronic pancreatitis, smoking, high alcohol consumption, blood type, and family history of cancer, Bork and colleagues were able to construct a molecular signature of 27 stool-derived microbes that accurately discriminated between cases of pancreatic cancer, chronic pancreatitis, and controls.

In order to test whether their results were generalizable beyond the Spanish study population, they evaluated the microbiome-based models in two validation scenarios.

In the first, the prediction accuracy of the models was confirmed in an independent German PDAC population, including 44 patients with PDAC and 32 controls. It was further validated against existing data from 25 studies across 18 countries that included patients with type 1 or type 2 diabetes, colorectal cancer, breast cancer, liver diseases, Crohn's disease, and ulcerative colitis, as well as healthy controls.

Comment