Utility of peritoneal fluid for multiplexed mRNA profiling of peritoneal metastasis from pancreatic cancer.

BACKGROUND

Patients with peritoneal metastasis (PM) have poor survival rates, and few methods for prognostic stratification of patients with proven PM exist. Peritoneal fluid (PF) provides a direct reflection of the microenvironment of the peritoneal cavity, and transcriptomic profiling of PF represents a novel and potentially clinically valuable approach that warrants investigation. We aimed to evaluate the utility of formalin-fixed paraffin embedded (FFPE) PF sediments left over after conventional peritoneal cytology diagnosis for multiplexed mRNA expression profiling and to identify dysregulated mRNAs in PM vs. controls. Furthermore, we explored whether certain mRNAs upregulated in PM hold prognostic value.

METHODS

Samples included were FFPE sediments of malignant PFs from cases of PM from pancreatic cancer (PM-PC, n = 19) scheduled for pressurized intraperitoneal aerosol chemotherapy (PIPAC) and benign ascites (controls, n = 16). RNA was extracted from the FFPE sediments. Expression profiling of 760 cancer-related mRNAs was performed, followed by unsupervised clustering, differential gene expression analysis, ROC/AUC analysis, generation of KM plots, and Cox proportional hazards regression. Cellular expression of proteins encoded by selected dysregulated mRNAs was analyzed using immunocytochemistry.

RESULTS

In PF sediments with malignant cells, 56 upregulated and 130 downregulated mRNAs were identified. Selected up- and downregulated mRNAs were similarly dysregulated at the protein level. Upregulation of six mRNAs (ESRP2, EPCAM, SFN, ITGB6, ESRP1, and GRHL2) was associated with short-term survival (P < 0.0025).

CONCLUSION

Our data indicate that PF from PM-PC patients is suitable for transcriptomic analysis. Numerous upregulated mRNAs known to play a role in tumour progression were identified. Our data indicate that mRNA profiling may be a clinically useful tool for further prognostic stratification of patients with proven PM, but larger studies are needed to validate our findings.