Abstract:Background: Cancer cell lines are important research models for studying tumor biology in vivo. The accuracy of such studies is highly dependent on the phenotypic and genetic similarity of cell lines to patient tumors, but this is not always the case, particularly for pancreatic cancer.Methods: We compared the gene expression profiles of various pancreatic cancer cell lines and primary human pancreatic tumor tissues to determine which pancreatic cancer cell line best models human primary tumor. Profiles of messenger RNA (mRNA) expression of 33 pancreatic cancer cell lines and 892 patient samples of pancreatic adenocarcinoma (PAAD) were obtained from the Gene Expression Omnibus (GEO) database. Microarray data were normalized using the robust multichip average (RMA) algorithm and batch effect removal was performed using ComBat. The pooled data of each PAAD cell line were compared to patient tumors based on the top 2,000 genes with largest interquartile range (IQR), 134 gene-collections of cancer-related pathways, and 504 gene-collections of cancer-related functions using pairwise Pearson's correlation analysis.Results: PAAD cell lines were poorly correlated with patient tumor tissues based on the top 2,000 genes. Up to 50% of cancer-related pathways were not strongly recommended in PAAD cell lines, and a small proportion of cancer-related functions (12-17%) were poorly correlated with PAAD cell lines. In pan-pathway analysis, the cell lines showing the highest genetic correlation to patient tumors were Panc 03.27 for PAAD cell lines from a primary lesion site and CFPAC-1 for PAAD cell lines from a metastatic lesion site. In pan-function analysis, the cell lines showing the highest genetic correlation to patient tumors were Panc 03.27 for PAAD cell lines from a primary lesion site and Capan-1 for PAAD cell lines from a metastatic lesion site.Conclusions: The gene expression profiles of PAAD cell lines correlate weakly with those of primary pancreatic tumors. Through comparison of the genetic similarity between PAAD cell lines and human tumor tissue, we have provided a strategy for choosing the appropriate PAAD cell line.

Systematic variation in gene expression patterns in human cancer cell lines

Systematic transcriptional analysis of human cell lines for gene expression landscape and tumor representation

A gene expression database for the molecular pharmacology of cancer

A comprehensive transcriptional portrait of human cancer cell lines

Comparative Analysis And Integrative Classification Of Nc160 Cell Lines And Primary Tumors Using Gene Expression Profiling Data

Comparative analysis and integrative classification of NCI60 cell lines and primary tumors using gene expression profiling data

Gene Expression Patterns Within Cell Lines Are Predictive of Chemosensitivity

A DNA microarray survey of gene expression in normal human tissues

A single-cell analysis of breast cancer cell lines to study tumour heterogeneity and drug response

Common Human Cancer Genes Discovered by Integrated Gene-Expression Analysis

Human cancers over express genes that are specific to a variety of normal human tissues

Pan-cancer single-cell RNA-seq identifies recurring programs of cellular heterogeneity

Characterization of Expression-Based Gene Clusters Gives Insights into Variation in Patient Response to Cancer Therapies

Comprehensive analysis of the gene expression profiles in human gastric cancer cell lines

Single Cell Multi-Omics Reveal Intra-Cell-line Heterogeneity Across Human Cancer Cell Lines

Correlation of gene expression profiles to identify pancreatic cancer cell lines that best model primary human tumors

Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications

Diverse Gene Expression Patterns in Response to Anticancer Drugs Between Human and Mouse Cell Lines Revealed by a Comparative Transcriptomic Analysis

Gene Expression Patterns in Human Liver Cancers

Assessing the Gene Regulatory Landscape in 1,188 Human Tumors

Next-generation characterization of the Cancer Cell Line Encyclopedia