Correlations between serum trypsinogen-2 and pancreatic cancer.

Abstract

Background/aims: To investigate associations be- tween serum trypsinogen-2, pancreatitis and pancreatic cancer (PC) and determine cutoff values for PC diagnosis.

Methodology: We recruited 88 patients from Internal Medicine/Surgical Departments of General Military Hospital of Beijing PLA between 12/2009 and 6/2010. Serum samples were collected preoperatively from 23 PC patients, 30 pancreatitis patients and 35 healthy controls. Enzyme-linked immunosorbent assay was used to detect trypsinogen-2 semiquantitatively.

Results: Serum trypsinogen-2 levels of PC and pancreatitis patients were significantly higher than those of controls (51.2 ± 80.3, 107.7 ± 98.1 vs. 1.0 ± 0.5, p = 0.03, p < 0.001) and significantly higher in pancreatitis vs. PC patients (107.7 ± 98.1 vs. 51.2 ± 80.3, p = 0.01). Higher Balthazar CT grades correlated with higher trypsinogen-2 in pancreatitis group. ROC curves for trypsinogen-2 revealed optimal cutoff value 1.8 as lower PC detection limit with 95.7% sensitivity and 91.4% specificity, and optimal cutoff value 19.9 for upper PC detection limit with 87.0% sensitivity and 97.1% specificity. Trypsinogen-2 levels correlated with pancreatic injury level. An AUC of 0.73 (95% Cl: 0.59-0.84, p = 0.002) distinguished PC from pancreatitis.

Conclusion: Serum trypsinogen-2 is associated with PC and pancreatitis. Levels between 1.8 μg/L and 19.9 μg/L strongly suggest PC. Detection of serum trypsinogen-2 may provide simple, sensitive, specific non-invasive initial screening for early PC diagnosis.