Gene expression profiling of circulating tumor cells in breast cancer

Background: Enumeration of circulating tumor cells (CTC) in metastatic breast cancer predicts relapse-free survival and treatment failure while scanty data are currently available on their molecular features. Since CTCs might represent a surrogate tissue, their transcriptional characterization could...

Full description

Saved in:
Bibliographic Details
Published in:European journal of cancer (1990) 2011-10, Vol.47, p.S12-S13
Main Authors: Cappelletti, V, Fina, E, Miodini, P, Callari, M, Musella, V, Agresti, R, Moliterni, A, Daidone, M G
Format: Article
Language:English
Subjects:
Blood
Breast cancer
CD3 antigen
CD8 antigen
Cell survival
Data processing
Epidermal growth factor receptors
ErbB-2 protein
Gene expression
Leukocytes
Lymphocytes B
Lymphocytes T
Metastases
Monocytes
Prognosis
RNA
Transcription
Tumor cells
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Background: Enumeration of circulating tumor cells (CTC) in metastatic breast cancer predicts relapse-free survival and treatment failure while scanty data are currently available on their molecular features. Since CTCs might represent a surrogate tissue, their transcriptional characterization could likely allow to identify pathways involved in metastatic dissemination and to obtain clinically relevant information for monitoring prognosis and treatment response. Materials and Methods: Predefined numbers (200, 100, 50) of MCF7 and MDA-MB-468 cells were spiked into blood from healthy donors captured using the AdnaTest EMT-1/Stem CellSelect kit (AdnaGen) and profiled (Illumina, DASL) in parallel with controls without cells and with RNA (100, 10, 1, 0.5 ng) from un-spiked cells. Controls were washed with PBS or with the AdnaWash buffer designed to improve leukocyte removal from captured cells. Results: Gene expression detection rates for captured cells were around 60%. As expected detection rates dropped to lower values in control samples either washed with AdnaWash (30%) or with the standard washing buffer (45%). Samples derived from different numbers of spiked cells and their cognate RNA (100, 10 ng) clustered together while controls and low RNA-samples (1, 0.5 ng) clustered separately. Gene expression variability was modest among replicates of samples with low RNA levels, but high among control samples. Gene Ontology analysis of genes exclusively expressed by captured spiked cells revealed an enrichment in those associated with ectodermic derivation and glandular function. AdnaWash treatment of controls reduces the expression of leukocytes genes. Comparison of genes expressed in spiked cells with those expressed in their cognate RNA revealed a consistent overlap (90%), whith samples derived from captured spiked cells (5% of all genes) enriched in genes associated to T-cells, B-cells and monocytes. Expression of key breast cancer genes (ER, EGFR, ERBB2) increased in spiked cells and their RNA compared to controls washed with standard or AdnaWash buffer while typically immune genes (CD3, CD8) were expressed at high levels only in samples not processed with the AdnaWash. Conclusion: We have developed a protocol allowing to obtain reliable gene expression profiles from as low as 50 CTCs.
ISSN:0959-8049