Development of a Serum-Based MicroRNA Signature for Early Detection of Pancreatic Cancer: A Multicenter Cohort Study

Jing Huang,Ge Gao,Yang Ge,Jianzhou Liu,Hongtu Cui,Ren Zheng,Jialin Wang,Si Wang,Vay Liang Go,Shen Hu,Yefu Liu,Minwei Yang,Yongwei Sun,Dong Shang,Yantao Tian,Zhigang Zhang,Zhongyuan Xiang,Hongyang Wang,Junchao Guo,Gary Guishan Xiao
DOI: https://doi.org/10.1007/s10620-024-08338-4
2024-03-09
Digestive Diseases and Sciences
Abstract:A grim prognosis of pancreatic cancer (PCa) was attributed to the difficulty in early diagnosis of the disease.
gastroenterology & hepatology
What problem does this paper attempt to address?
The paper aims to address the difficulty of early diagnosis of pancreatic cancer (PCa). Pancreatic cancer has a very poor prognosis, with a 5-year relative survival rate of only about 10%, mainly due to the difficulty of early diagnosis. Current diagnostic methods used in clinical practice, such as imaging techniques and blood biomarkers, lack sensitivity and specificity. Therefore, researchers urgently need to find new biomarkers to improve the overall survival rate of pancreatic cancer patients. The study was conducted through the following steps: 1. **Identification of candidate microRNAs (miRNAs)**: First, candidate miRNAs were identified using microarray analysis in formalin-fixed, paraffin-embedded (FFPE) tissues. 2. **Validation cohort study**: These candidate miRNAs were then validated for their clinical utility in serum samples. A total of 1273 participants from four centers were recruited for the study, divided into training and validation cohorts. The collected serum samples were analyzed using real-time polymerase chain reaction (RT-PCR). The results showed that 27 miRNAs were significantly differentially expressed in pancreatic cancer tissues compared to benign controls. Four miRNAs were selected from these as a diagnostic panel, which outperformed commonly used tumor markers such as CA19-9, CA125, CEA, and CA242. This miRNA panel demonstrated excellent performance in distinguishing early-stage pancreatic cancer patients from healthy controls or non-pancreatic cancer patients (including chronic pancreatitis and pancreatic cystic tumors), with areas under the curve (AUC) of 0.971 (95% CI 0.956–0.987) and 0.924 (95% CI 0.899–0.949), respectively. Additionally, the panel was able to exclude interference from other gastrointestinal tumors, with a specificity of 90.2%. The conclusion is that this four-miRNA panel shows significant discriminative ability in the screening of early pancreatic cancer and may develop into a novel, non-invasive early detection method.