Effectiveness of the combined evaluation of KLK3 genetics and f/tPSA ratio for prostate cancer diagnosis

Purpose. Forty per cent of serum PSA variability depends on inherited factors. We ascertained whether the knowledge of KLK3 genetics enhance PSA diagnostic performance in patients with clinical suspect of prostate cancer (PCa). Materials and Methods. We studied 1058 men who consecutively underwent prostate biopsy for clinical suspect of PCa. At histology PCa was present in 401 and absent in 657 cases. Serum tPSA and f/tPSA were determined. Four polymorphisms of KLK3 gene (rs2569733, rs2739448, rs925013 and rs2735839) and one polymorphism of SRD5A2 gene (rs523349) were studied. The influence of genetics on PSA variability was evaluated by means of multivariate linear regression analysis. The performances of tPSA and f/tPSA alone or combined with a genetic-based patients’ classification were defined by receiver operating characteristic curve analyses. Results. For PCa diagnosis f/tPSA index alone (cut-off = 11%) was superior to tPSA (cut-off = 4ng/ml) or f/tPSA reflex testing (61%, 43% and 54% positive predictive values respectively). PSA correlated with KLK3 genetics (p=0.001 for rs2735839 polymorphism and p=0.003 for rs2569733, rs2739448 and rs925013 haplotype combination). In patients with different KLK3 genetics, two optimal f/tPSA cut-off (11% and 14.5%) were found. For f/tPSA values between 11%-14.5% PCa probability ranged from 30.0% to 47.4% according to patient’s genetics. Conclusions. f/tPSA is superior to tPSA for PCa diagnosis, independently from tPSA results. f/tPSA findings below 11% are positively associated with PCa, those above 14.5% are negatively associated with PCa, while the interpretation of those between 11% and 14.5% is improved by patients’ KLK3 genetics analysis.