Elsevier, European Journal of Cancer, 1(45), p. 74-81, 2009
DOI: 10.1016/j.ejca.2008.09.030
Full text: Download
AIM: Assessment of intra- and inter-laboratory variation in multi-centre real-time reverse-transcribed PCR (qRT-PCR)-based mRNA quantification of a prognostic marker in breast cancer using external quality assurance (EQA). METHODS: A questionnaire on the methodologies used and EQA calibrators were sent to 5 participating laboratories from 4 European countries, which measured mRNA levels of PITX2 splice variants and reference genes by qRT-PCR. RESULTS: Differences in the methodology included PCR quantification methodology and equipment, RNA extraction and cDNA synthesis procedures. The intra-laboratory coefficient of variation (CV) ranged from 5 to 23%, and the inter-laboratory CV ranged from 17 to 30%. The inter-laboratory CV was reduced to 13% by using prediluted calibrators and by harmonising the data in the central QA laboratory. Additional normalisation using reference genes did not decrease the variation further. CONCLUSIONS: Both externally provided calibrators and centralised harmonisation are required to reduce the intra-laboratory variation in multi-centre qRT-PCR results to an acceptable level.