췌장암에서 다중 비번역 RNA 기반 위험모델의 예후적 가치

Prognostic Value of Multi-non-coding RNA-based Risk Models in Pancreatic Cancer

The combinational multi-gene biomarkers can reflect the heterogeneity and clinico-pathological status of tumors, thereby offering improved prognosis classification and risk assessment compared to single-gene expression biomarker. Long non-coding RNAs (lncRNAs) and microRNAs are non-coding RNA molecules that regulate gene expression at multiple levels and affect cancer progression. LncRNAs and microRNAs can be used as diagnostic and prognostic biomarkers and can predict treatment response. Due to the difficulty in obtaining biopsy samples and high mortality rate, developing wellvalidated biomarkers with clinical utility is challenging in pancreatic cancer. This article reviews recent attempts to use lncRNA and microRNAs expression to construct prognostic risk models in pancreatic cancer. We evaluated the data sources, signatures of non-coding RNAs, analytical methods, and performance of 13 kinds of risk models. Additionally, the relevance of immune status and drug sensitivity with these risk models was summarized. This analysis will help to identify and validate multi-gene-based prognostic signatures with clinical utility in pancreatic cancer.