Reproducible gene expression measurement among multiple laboratories obtained in a blinded study using standardized RT (StaRT)-PCR

• Published in: Molecular diagnosis 2001-12, Vol.6 (4), p.217-225
• Main Authors: Crawford, Erin L., Peters, Godfridus J., Noordhuis, Paul, Rots, Marianne G., Vondracek, Martin, Grafstr[ouml ]m, Roland C., Lieuallen, Kimberly, Lennon, Gregory, Zahorchak, Robert J., Georgeson, Melanie J., Wali, Anil, Lechner, John F., Fan, Pan Sheng, Kahaleh, M.Bashar, Khuder, Sadik A., Warner, Kristy A., Weaver, David A., Willey, James C.
• Format: Article
• Language: English
• Subjects: Binding, Competitive - genetics; Cell Line; Databases, Genetic; DNA, Complementary - genetics; Double-Blind Method; Gene Expression; Gene Expression Profiling - classification; Gene Expression Profiling - standards; Gene Expression Profiling - statistics & numerical data; Humans; Lung - chemistry; Lung - cytology; Lung - metabolism; Reverse Transcriptase Polymerase Chain Reaction - standards; Reverse Transcriptase Polymerase Chain Reaction - statistics & numerical data; RNA, Messenger - genetics; Templates, Genetic; Terminology as Topic
• ISSN: 1084-8592; 1532-8619
• DOI: 10.1054/modi.2001.29789

Background: A method that provides standardized data and is relatively inexpensive and capable of high throughput is a prerequisite to the development of a meaningful gene expression database suitable for conducting multi-institutional clinical studies based on expression measurement. Standardized RT (StaRT)-PCR has all these characteristics. In addition, the method must be reproducible. StaRT-PCR has high intralaboratory reproducibility. The purpose of this study is to determine whether StaRT-PCR provides similar interlaboratory reproducibility. Methods and Results: In a blinded interlaboratory study, expression of ten genes was measured by StaRT-PCR in a complementary DNA sample provided to each of four laboratories. The average coefficient of variation for interlaboratory comparison of the nine quantifiable genes was 0.48. In all laboratories, expression of one of the genes was too low to be measured. Conclusion: Because StaRT-PCR data are standardized and numerical and the method is reproducible among multiple laboratories, it will allow development of a meaningful gene expression database.