Purpose:
The goal of the study was to detect the existence of an abnormal proteomic pattern in prostate cancerpatients by comparing them to control subjects from a matching age group.
Method:
The study was conducted on50 newly diagnosed prostate cancer (PCa) patients and 50 normal individuals who were proven clinically and bylaboratory investigations to be healthy. Plasma samples were processed using MB-WCX magnetic beads andsamples were tested using a mass spectrometer. Mass spectra was acquired using Bruker's FLEX-analysisprogram and later the peaks were analysed using the ClinProTools analysis. Three statistical models (Geneticalgorithm GA, Supervised Neural Network SNN, Quickclassifier QC) were generated to detect the peaks.
Results:
The results showed 26 peaks were found to be significantly expressed between the cases and thecontrols with PWKW <0.05. Out of these, six peaks were over expressed in cases, while 20 peaks were under-expressed. The GA model generated the best peak combination, showing five peaks with the m/z ratios 2485.97,1061.24, 3295.1, 4612.54 and 2817.28. This model achieved a sensitivity of 87.5% and a specificity of 92.9%during external validation.
Conclusion:
It can be concluded that proteomic profiling can be an effective method for the discovery of newblood-based tumour markers for PCa patients. Moreover, MALDI-TOF proteomic profiling represents a newfrontier for screening and early diagnosis of prostate cancer in Egypt. The analytical performance of multiplexedtumour profiles exceeds that of the single traditional tumour markers.