Interlaboratory comparison of quantitative RT-PCR based detection for minimal residual disease in leukemias: a...

Real-time quantitative polymerase chain reaction (RQ-PCR) has been accepted as integral part of the management of patients with hematologic malignancies. Whereas standardization efforts of RQ-PCR, initiated by Europe Against Cancer (EAC) group, have been gradually widespread in the world, Japanese laboratories use their individual protocol for RQ-PCR analysis. Therefore, we assessed the variability of quantitative results obtained from 4 different laboratories in Japan, including 3 companies and Tohoku University Hospital, using identical peripheral blood or bone marrow samples of patients in chronic myeloid leukemia (CML; n = 11) and acute myeloid leukemia (AML; n = 2). RQ-PCR was designed to quantify the copy numbers of disease-specific fusion chimeras; BCR-ABL (CML) and AML1-ETO (AML). In 5 out of 13 samples, the quantitative results from 4 laboratories varied more than 10 times (up to 712 times). Thus, we next sought to determine factors affecting the variability of RQ-PCR results across laboratories, by sending back RNA and cDNA samples from each company to Tohoku University, and they were further proceed to yield quantitative data. The main difference between companies and Tohoku University was probably due to the difference of blood separation method (Blood lysis or Ficoll-Hypaque). On the other hand, the variability among 4 laboratories was the most noticeable in the PCR step, mainly attributable to the difference of primer/probe sequence among laboratories. In conclusion, our analyses indicate the importance to limit both preanalytical (sample processing) and analytical (RQ-PCR) interlaboratory variability for RQ-PCR protocol, and the need of further efforts on standardization program in Japan.