POSSIBLE BIOMARKERS FOR PROSTATE CANCER: DEVELOPMENT AND VALIDATION OF A FULLY AUTOMATED, RAPID, HIGH THROUGH OUT SPME-FAST GC-MS METHOD USING A NEW IONIC LIQUID COLUMN FOR THE DETERMINATION OF SARCOSINE AND N-ETHYLGLYCINE IN URINE AND URINARY SEDIMENTS

Aim of the study There is an increasing interest on seeking for new biomarkers able to screen between indolent and aggressive prostate cancer (PCa), in order to improve diagnostic accuracy. Among them, sarcosine has been suggested to be a promising tool in PCa diagnostics. The goal of our study was to construct a sequential process to move these preliminary biological indications into the clinical setting. Sarcosine (S) and N-ethylglycine (N-EG) were evaluated as potential biomarkers for early prostate cancer detection and for prediction of its aggressiveness. Materials and methods A fully automated SPME-Fast GC-MS method for the determination of S and N-EG in biological fluids was used for high-throughput analyses. Fifty-six samples of urine and urinary sediments were collected from 33 patients with clinically localized prostate cancer (PCa), 10 with histological confirmation of benign prostatic hyperplasia (BPH) and 13 healthy patients (HP). Urinary creatinine levels were measured by a standard DCA Microalbumin/ Creatinine assay. All the measurements were performed in a blinded manner. The study was approved by the hospital ethical board. Results The developed method allowed to detect S and N-EG within 20 min., thus assessing its feasibility for point of control testing. As for urinary sediments, S and N-EG did not fulfill the conditions of useful markers: N-EG was absent in all the analyzed samples, whereas S was detected only in 15/56. As for S in urine, the Kruskal-Wallis test showed a significant difference (p<0.05) among HP, BPH and PCa patients. A further evaluation using the multiple comparisons test showed the presence of significant differences between PCa and HP and between PCa and BPH patients. Swas also related to GS (<7 and ≥7) but no statistically significant differences were observed (p>0.05). Similar results (p=0.08) were obtained also when we stratified patients in organ confined disease and extracapsular extension. ROC analysis showed that the highest values of sensitivity (79%) and specificity (87%) were obtained in correspondence of a cut-off value of 179 μg sarcosine/g creatinine, whereas on the same groups of patients, data related to the preop. PSA cut-off value of 4 ng/ml allowed to obtain a sensitivity and selectivity of 85% and 80%, respectively. The Area under the ROC curve was AUC=0.821±0.0695 for S, whereas the AUC of preop. total PSA was slightly larger than that of S (0.933±0.0362), thus presenting comparable results and without reaching a statistically significant difference (p=0,1831). Conclusions The fully automated SPME-Fast GC-MS method allows high throughput analyses at low cost, being sensitive, specific and non-invasive. Whereas S and N-EG cannot be considered reliable biomarkers for PCa in urinary sediments, our results suggest that S could be effective as a urinary biomarker: the analysis of a greater number of samples taken both from HP and from people with PCa will be helpful to definitively clarify its role.