381 Background: Prostate cancer (PCa) is diagnosed in approximately 1.5 million men worldwide each year. Many men have clinically low or intermediate risk PCa and can be managed with active surveillance (AS). However, many men are unnecessarily treated with surgery, radiation or hormone therapy which are associated with reduced quality of life. Molecular biomarkers can provide additional prognostic value to clinical information to help inform and improve patient management. A novel six-gene (FOXM1, MCM3, MTUS1, TTC21B, ALAS1, PPP2CA) signature has been developed and validated as prognostic for multiple outcomes that indicate aggressive PCa. Methods: N = 504 men with clinically low or intermediate risk PCa (Cancer of the Prostate Risk Assessment (CAPRA) score of 0 to 5 inclusive) were identified from two cohorts of patients treated with radical prostatectomy in Ireland or Sweden. Expression of the six-gene signature was measured by RT-qPCR and combined with CAPRA scores in a clinically validated algorithm to calculate a continuous Molecular Clinical Risk Score (MCRS). The prognostic value of MCRS was compared with benchmark clinical information; CAPRA score and European Association of Urology (EAU) risk categories. Comparisons for binary outcomes (adverse pathology, high primary Gleason) were based on univariable and bivariable logistic regression models. Comparisons for time-to-event outcomes (biochemical recurrence) were based on univariable and bivariable Cox regression models. Clinical utility of MCRS and benchmark clinical information was assessed using decision curve analysis (DCA). Results: MCRS was more prognostic than clinical benchmarks and added statistically significant independent prognostic value to clinical benchmarks (see table). DCA showed that the net benefit of using MCRS to guide treatment decisions was generally higher than EAU and CAPRA. Conclusions: Among patients who are clinically low or intermediate risk, the six-gene MCRS provides more prognostic value than clinical information, adds statistically significant independent prognostic value to clinical information, and has more clinical utility than clinical information for multiple PCa outcomes. The six-gene signature can inform and improve patient management and lead to better patient outcomes. Adverse pathology High primary Gleason Biochemical recurrence Univariable analysis AUC (95% CI) AUC (95% CI) C index (95% CI) MCRS 0.77(0.73 to 0.82) 0.79(0.75 to 0.84) 0.72(0.66 to 0.78) EAU 0.61(0.56 to 0.66) 0.69(0.65 to 0.74) 0.58(0.51 to 0.65) CAPRA 0.64(0.59 to 0.69) 0.73(0.68 to 0.78) 0.64(0.57 to 0.71) Bivariable analysis Increase in AUC (95% CI) Increase in AUC (95% CI) Increase in C index (95% CI) MCRS (added to EAU) +0.17(0.12 to 0.22) +0.11(0.08 to 0.16) +0.15(0.08 to 0.19) MCRS (added to CAPRA) +0.13(0.09 to 0.19) +0.07(0.04 to 0.11) +0.08(0.04 to 0.14)
Krzyzanowska et al. (Sun,) studied this question.