Fecal Microbiome Signatures May Predict Pancreatic Ductal Adenocarcinoma

pancreas of human
Human pancreas, abstract illustration.
A team of investigators examined the potential of fecal and salivary microbiota to serve as diagnostic biomarkers for pancreatic ductal adenocarcinoma.

Stool microbiota-based classifiers may predict pancreatic ductal adenocarcinoma (PDAC) with high accuracy and specificity, independent of disease stage, according to study findings published in Gut.

The case-control study prospectively recruited participants between 2016 and 2019 from 2 hospitals in Spain. The analysis included 57 newly diagnosed, treatment-naïve patients aged over 18 years with PDAC, 29 patients with chronic pancreatitis (CP), and 50 control individuals matched for age, sex, and hospital. The researchers obtained fecal shotgun metagenomes for all participants and salivary metagenomes for 45 patients with PDAC, 12 with CP, and 43 control individuals.

The fecal microbiome composition of patients with PDAC was different, albeit at small effect sizes, compared with the control group (R2=0.02, P ≤.0001) and patients with CP (R2=0.02, P =.003).

A fecal metagenomic classifier identified PDAC with an accuracy of 0.84 area under the receiver operating characteristic curve (AUROC), based on 27 species. The most prominent positive marker species included Methanobrevibacter smithii, Alloscardovia omnicolens, Veillonella atypica, and Bacteroides finegoldii.

Accounting for carbohydrate antigen (CA) 19-9 increased the accuracy of the first model to AUROC 0.94, which was primarily driven by an increase in sensitivity. When the enrichment-constrained second model was amended with CA 19-9, a large increase was observed in accuracy from AUC 0.71 to 0.89, also driven by a significant improvement in sensitivity.

The PDAC-specific signatures were further validated against independent, external metagenomic datasets in various health conditions, in 5792 publicly available gut metagenomes from 25 studies in 18 countries. In addition, the 2 microbiome-based models had a high prediction accuracy in a German PDAC population in the validation phase.

The researchers also taxonomically profiled all fecal and salivary samples, along with biopsies of tumors and adjacent healthy pancreatic tissue of patients with PDAC with use of 16S rRNA amplicon sequencing. A rich and diverse pancreas microbiome was found, with at least 13 bacterial genera in at least 25% of samples. Lactobacillus spp, Akkermansia muciniphila, and Bacteroides spp were enriched in tumors relative to nontumor pancreatic tissue.

“The 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,” the researchers wrote. “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.”


Kartal E, Schmidt TSB, Molina-Montes E, et al. A faecal microbiota signature with high specificity for pancreatic cancer. Gut. Published online March 8, 2022. doi:10.1136/gutjnl-2021-324755