We report on the attempt to develop a simple non-invasive screening protocol for prostate cancer (PCa). Absolute concentrations of 19 macro and trace elements (Ag, Al, B, Ba, Ca, Cd, Cu, Fe, K, Mg, Mn, Na, P, Pb, S, Si, Sr, Tl, Zn) were determined using inductively coupled plasma optical emission spectroscopy (ICP-OES) and atomic absorption spectroscopy (AAS) techniques in 34 urine samples from patients with biopsy-confirmed PCa and 32 urine samples from controls. All the possible concentration ratios were calculated as well. Various data processing methods, including Principal Component Analysis, Logistic Regression, and Decision Trees, were applied for data modeling to study if the elemental concentration profile may contribute to the development of new screening tools for prostate cancer. Several statistically significant differences between the two groups were observed both in the individual element concentrations and in their ratios. The mathematical classification models built for the prediction of the patient's status with respect to PCa based on elemental profile have shown the accuracy of up to 89%, thus exceeding the accuracy of the standard prostate-specific antigen testing.