CLRC—TR—08—02 Analysis of serial UKCTOCS-OC data: discriminating abilities of proteomics peaks

This study aims to prove empirically that the information contained in serum proteome mass spectra can improve the discriminating ability of CA125 in early stages of ovarian cancer (OC) development. Our serial data collected in the UKCTOCS project over the period of 7 years allow us to reject the null hypothesis at significance level 5% for detection up to 15 months in advance of the diagnosis which was confirmed by histology/cytology. Moreover, in addition to what has been previously reported we identified certain mass spectrometry (MS) peaks that, in combination with CA125 level, can provide reliable long term prediction of OC. The peaks with m/z-values 7772 Da and 9297 Da were the most informative in extending the period of significant discrimination, and in combination with CA125 they proved to be able to predict the disease up to 6 and 4 months earlier than CA125 alone, respectively.