Diagnostic value of tissue mRNA and microRNA expression profiles for differentiating pancreatic cancer from chronic pancreatitis: A pilot study

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Background: Differential diagnosis of pancreatic ductal adenocarcinoma (PDAC) in patients with chronic pancreatitis (CP) remains challenging because of overlapping morphological and imaging features. Transcripts of tumor-associated genes and microRNAs (miRNAs) are considered promising biomarkers of malignant transformation. Our group previously investigated the expression of 17 tumor-associated messenger RNAs (mRNAs) in pancreatic neoplasms; however, their expression in PDAC compared with CP and their regulatory interactions with tissue miRNAs have not been comprehensively evaluated. Aim: To analyze the differential expression of 17 tumor-associated mRNAs genes and miRNA profiles in tissue samples of PDAC and CP, and to perform an integrative assessment of regulatory patterns to determine the potential diagnostic value of the identified markers. Methods: A single-center, prospective, cross-sectional, exploratory diagnostic biomarker study was conducted. Pancreatic tissue samples obtained by ultrasound-guided fine-needle aspiration biopsy between October 2020 and July 2023 were examined. The relative expression of 17 target mRNAs (CCNB1, PYGL, KIF22, UBE2C, CDK1, PKM, ELOVL6, NAPEPLD, MYC, CLDN18, GPC1, MUC1, MUC5AC, MUC4, MUC16, PLAU, and ITGA2) was assessed by real-time polymerase chain reaction. Subsequently, small RNA next generation sequencing and bioinformatic analysis were performed to assess differential miRNA expression. Integrative in silico analysis included evaluation of miRNA-mRNA regulatory interactions and functional annotation of the identified molecules using miRNet 2.0, KEGG, and Gene Ontology databases. Results: The study included 18 patients, of whom 13 were diagnosed with PDAC and 5 with CP by histological examination. Comparative analysis of mRNA expression profiles revealed significantly elevated levels of ITGA2 (p = 0.028), MUC1 (p = 0.019), and PKM (p = 0.0007) in PDAC samples compared with CP. The highest AUC value was obtained for PKM (AUC = 1.00; 95% confidence interval 1.000–1.000; p = 0.0007). Small RNA sequencing identified 1319 miRNAs, of which 21 showed differential expression between PDAC and CP according to the criterion |log2FC| ≥ 1.0 at p 0.05; however, none reached the threshold for statistical significance after multiple testing (padj 0.05) between the PDAC and CP groups. The most pronounced downregulation was observed for hsa-miR-205-5p (log2FC = -11.885; p = 0.003), while hsa-miR-146a-5p showed the greatest upregulation (log2FC = 2.993; p = 0.0001). Integrative analysis identified eight miRNAs (hsa-miR-335-5p, hsa-miR-155-5p, hsa-miR-335-3p, hsa-miR-342-3p, hsa-miR-31-5p, hsa-miR-205-5p, hsa-miR-22-3p, and hsa-miR-28-3p), that functionally connected to PKM, MUC1, and ITGA2. Hsa-miR-335-5p showed the highest network connectivity (degree of connectivity was 9,809,622). Conclusion: The elevated expression of PKM, MUC1, and ITGA2, together with altered expression of hsa-miR-205-5p and hsa-miR-335-5p, suggests that these molecules may serve as potential components of a diagnostic panel for differentiating PDAC from CP. These findings require further validation in independent cohorts. Integrative analysis of mRNA and miRNA profiles may represent a promising approach for selecting combinations of molecular markers for subsequent development and validation of comprehensive diagnostic models for pancreatic neoplasms.