Profiling protease cleavage patterns in plasma for pancreatic cancer detection

Proteases are promising biomarkers for cancer early detection. Their enzymatic activity against peptide substrates allows for their straightforward detection using low-cost tests. However, the complexity of the human proteome makes it challenging to develop sensitive and selective tests against a specific protease biomarker. Here, we report a different approach by utilizing the total protease activity in plasma samples to detect pancreatic cancer. Instead of targeting a specific protease using a specific peptide substrate, we utilized an array of 360 FRET substrates to screen for cleavage patterns in plasma samples collected from screen negatives and pancreatitis or pancreatic ductal adenocarcinoma cancer (PDAC) patients. In this proof of concept study, we first screened all 360 substrates using a small cohort (n = 13) to identify the top 5 substrates that best separate different conditions. Then, we performed a validation study using a larger cohort (n = 86) and the selected substrates. There was a statistically significant increase in the total protease activity in PDAC samples compared to screen negative and pancreatitis samples. The selected substrates detected PDAC with an area under the curve (AUC) of 0.8. This work represents a novel strategy for identifying peptide substrates for the detection of PDAC and other cancers.