Abstract B44: Reproducibility and short-term intraindividual variability of telomere length measurement using a monochrome multiplexing quantitative PCR

Abstract

Abstract Background: Telomeres are repetitive DNA structure at the end of chromosomes that are required for genomic integrity and stability. Numerous epidemiologic studies have examined telomere length in peripheral blood in relation to cancer and other aging-related diseases. In many of these studies, a measurement of telomere length from a single blood draw is quantified using a real-time PCR technique, and assumed to reflect a replicative history of hematopoietic stem cells in individuals. We examined the reproducibility of telomere length measurement and the relative magnitudes of sources of variation using sequential samples collected over a 9-month period. Methods: Relative telomere length in peripheral blood was estimated using a single tube monochrome multiplex quantitative PCR assay in blood DNA samples from 27 non-pregnant adult women (aged 35 to 74 years) collected in 7 visits over a 9-month period. Each specimen was assayed in triplicate on each of two plates. Reproducibility of the technical triplicates within a plate was assessed using the coefficient of variation (%CV). A linear mixed model was used to estimate the components of variance for telomere length measurements attributed to variation among women and variation between time points within women. Results: Within-plate technical replicate variability ranged from 8.5 to 15% (mean %CV=12.9). Plates had a significant systematic influence on telomere length measurements, explaining 16% of the variance; however, telomere length measurements between different plates were highly correlated (Pearson's correlation coefficient r=0.93, p<0.01). After controlling for plate effects, 64% of the remaining variance was estimated to be accounted for by variance due to subject. Variance explained by date of collection within a subject was minor, contributing 5% of the remaining variance. Compared to classification into quartiles based on all 7 time point measures and all six technical replicates for each time point, using the six technical replicates from a single time point resulted in the same quartile classification for 73%. Using a single time point and only 3 technical replicates decreased that agreement to 64%. These differences in quartile assignment were almost always only across neighboring quartile categories. Conclusion: Our data demonstrate good reproducibility of telomere length measurement using blood from a single draw. However, the existence of technical variability, particularly plate effects, reinforces the need for technical replicates and balancing of case and control samples across plates. Citation Information: Cancer Prev Res 2010;3(12 Suppl):B44.