DEPEND – early identification of pancreatic cancer using pancreatic exocrine insufficiency and faecal elastase as screening tools


Session type:

Declan McDonnell1, Paul Afolabi1, Christopher Byrne1, Sam Wilding2, Victoria Goss2, Zaed Hamady1
1University Hospital Southampton NHS Foundation Trust, 2University of Southampton



Pancreatic cancer is among the deadliest cancers, with five-year survival rates between 5 and 10 percent. Persons with pancreatic cancer are frequently diagnosed with late-stage disease, due to the lack of specific symptoms presenting in the early stages of disease progression. There is evidence to suggest persons with pancreatic cancer frequently develop pancreatic exocrine dysfunction due to tumour in-growth and obstruction of the pancreatic duct. In this study, we will investigate whether a screening tool combining a 13C-mixed-triglyceride breath and faecal elastase tests can be used to distinguish between persons with pancreatic cancer and healthy controls.


25 participants with known pancreatic cancer will be age and sex matched with 25 healthy controls and 10 participants with known chronic pancreatitis. After a 12 hour fast, participants provide baseline breath samples before being given 250mg of a 13C isotope-containing triglyceride with a solid test meal. Additional breath samples are taken every 30 minutes for 4 hours. The collected breath samples are analysed for the ratio of 13CO2 to 12CO2 using a Continuous Flow Isotope Ratio Mass Spectrometer. Faecal elastase concentrations are measured using a standard hospital ELISA kit. The primary endpoint is determining pancreatic cancer based upon participants with pancreatic cancer having a significantly different test profile to healthy volunteers. A sample size of 25 per group will provide >90% power to detect a difference of 17.8% in the 13C-MTG breath test.


Results will be available at the conference. Comparisons will be made between the raw and cumulative percentage dose recovered (PDR) of 13CO2 and faecal elastase values between the groups at various time points and if the test demonstrates a notable difference at any particular time point which could be applied to clinical practice. Risk cut-offs will be considered to calculate the specificity, sensitivity, positive and negative predictive values of the breath and faecal tests separately, and in combination.


This study will assess the utility of breath and faecal tests in distinguishing between participants with and without known pancreatic cancer.

Impact statement

If results are positive, then breath and faecal tests may be used as screening tools for pancreatic cancer.