Pancreatic cancer RNA sequencing

For optimal pancreatic cancer treatment, early and accurate diagnosis is vital. Blood-derived biomarkers and genetic predispositions can contribute to early diagnosis, but they often have limited accuracy or applicability. Here, we seek to exploit the synergy between both approaches by combining the standard clinical blood biomarker CA19-9 with novel genetic variants. Concretely, we aim to use deep sequencing and deep learning to improve pancreatic cancer diagnosis, to differentiate between resectable pancreatic ductal adenocarcinoma (rPDAC) from chronic pancreatitis (CP), and to estimate survival.

We obtained samples of nucleated cells found in peripheral blood from over 300 patients suffering from resectable pancreatic ductal adenocarcinoma, non-resectable pancreatic cancer (nrPC), chronic pancreatitis or none of these. We sequenced RNA with high coverage and reduced millions of raw to hundreds of high-quality, significant genetic variants. Together with CA19-9 levels, these served as input to deep learning to separate cancer from non-cancer, resectable PDAC from chronic pancreatitis, and resectable from non-resectable cancer.

All deep learning models achieved area under the curve (AUC) scores of over 90%. In particular, differentiating resectable PDAC from pancreatitis can be solved with an AUC of 98%. Moreover, we identified genetic variants to estimate survival in rPDAC patients.

Overall, we show that the blood transcriptome harbours genetic variants, which substantially improve non-invasive clinical diagnosis and patient stratification in pancreatic cancer.EGA study EGAS00001004706