An assessment of the baseline variability in the level of DNA damage in women as measured by the single cell gel electrophoresis assay

It is thought that a correlation between levels of DNA damage expressed in an individual and the potential for the future development of cancer exists, and that biomonitoring for such damage could eventually be used to detect individuals at risk before the onset of disease. The purpose of studies such as that described in this thesis is to develop and refine methods of biomonitoring for genetic damage. The objective of this thesis was to assess the background levels of DNA damage in women using the Single Cell. Gel Electrophoresis (SCGE) assay and interpret the meaning of the observed variability with respect to future study design and the sample size necessary to ensure statistical significance. Thirteen female subjects were recruited from the Vancouver Hospital-UBC site and the Occupational Hygiene programme at the University of British Columbia. Each subject was to provide six blood samples over a period of ten weeks, at varying or irregular intervals. A total of 73 blood samples were obtained. Peripheral blood lymphocytes were isolated from the samples, embedded in agarose, lysed to release the nuclear contents, and exposed to an electric current in order to allow the DNA to migrate from the nucleus. This procedure, known as the SCGE or "Comet" assay, enables the detection of single strand breaks and alkali-labile sites in the cellular DNA, as smaller fragments created by breakage will travel farther from the nucleus of the cell than larger fragments or unbroken DNA. Both technical and biological variability were observed in the sample data. A calculation of the coefficient of variation for several groups of data provided a crude estimate of variation for this study. The components of overall or total variability could not be determined, but both the inter- and intra- individual variability appeared to exceed the replicate-to-replicate technical variability. The internal standard used in this study did not provide the information desired with respect to day-to-day variability. It was concluded that the observed variation within individual subjects in the study necessitates the use of a longitudinal or multiple-samples-over-time study design, rather than cross sectional. The sample size necessary for statistical significance in a cross-sectional study comparing two groups with an alpha of 0.05 and a power of 0.80 is approximately 70 individuals per group if the detection of an increase of fifteen percent in image length is desired, and approximately 20 individuals per group if an increase of thirty percent in image length is to be detected.Medicine, Faculty ofPopulation and Public Health (SPPH), School ofGraduat