Biomarkers in prostate cancer : defining 'pussycat versus tiger' phenotype by proteomic modeling
Alghanem, Bandar (2010) Biomarkers in prostate cancer : defining 'pussycat versus tiger' phenotype by proteomic modeling. MRes thesis, University of Nottingham.
Prostate cancer is the one of the major causes of morbidity and mortality in the western world. It affects the prostate gland of males with a significant increase in the disease incidence every year. Current diagnostic and prognostic markers, such as prostate specific antigen (PSA), rectal examination and Gleason grades have their own limitations in a wider context of disease treatment and prediction. There is therefore a pressing need for novel and powerful biomarkers at protein or metabolite level. This study attempts to profile and identify candidate prostate cancer stage specific markers, within a defined population of samples. The samples were classified, based on the pathological information as “aggressive” (Gleason grade > 7) and “nonaggressive “(Gleason grade < 7). The proteomic protocols standardised at the John van Geest Cancer Research Centre, were used for the initial characterisation of the samples. The MS spectra obtained from the samples were used applied to an artificial neural network (ANN) based algorithm to generate predictive ions able to classify the samples. Three ions (m/z 1268.8, 998.6, 910.4) were able to predict and classify with high specificity and sensitivity. 24 samples were immunodepleted and subjected to nano-LC fractionation and MALDI-TOF analysis, generating 80-120 protein identities per sample. The three ions predicted previously by the ANN identified as Haemopexin, Gelsolin and Apolipoprotein B 100. Using ProfileAnalysis software, this study identified Apolipoprotein isoforms, including Apolipoprotein B 100, and Afamin as the proteins which showed differential expression in between the groups. This study identifies Apolipoprotein B 100 as a potential marker using two different modeling approaches suggesting this protein as the potential biomarker candidate. The utility of high throughput proteomic platforms such as Robotic liquid handling, MALDI-TOF and LC-MALDI for serum biomarker identification in PCa has been shown during this investigation.
Actions (Archive Staff Only)