Stability of a 17-Gene Genomic Prostate Score in Serial Testing of Men on Active Surveillance for Early Stage Prostate Cancer.

ABSTRACT

PURPOSE: Genomic testing may improve risk stratification in men with prostate cancer managed by active surveillance. We aimed to characterize the stability and usefulness of serial genomic test scores in men undergoing serial biopsies during active surveillance. MATERIALS AND METHODS: We compiled clinical and disease characteristics of men on active surveillance using an institutional Urologic Outcomes Database. We included patients initially diagnosed with Gleason 3 + 3 prostate cancer who elected active surveillance and received 2, 17-gene GPS (Genomic Prostate Score) results. We examined the association of GPS results and Gleason grade reclassification (Gleason 3 + 4 or greater) with definitive treatment using multivariable Cox proportional hazards regression models. RESULTS: We identified 111 men who underwent serial genomic testing. There were 49 grade reclassification events (44%) at a median followup of 64 months. The mean ± SD GPS change between the first and second biopsies was 2.1 ± 10.3. The GPS at first biopsy (per 5 units HR 1.04, 95% CI 1.00-1.07, p=0.03) was associated with an upgrade at second biopsy, although the second GPS was not (HR 1.02, 95% CI 0.99-1.05, p=0.13). The first and second GPSs (HR 1.09, 95% CI 1.04-1.14 and HR 1.09, 95% CI 1.04-1.14, each p <0.01) were associated with active treatment. CONCLUSIONS: The GPS undergoes small changes with time. Absolute GPS results at the first and second biopsies were associated with Gleason upgrading and transition from active surveillance to active treatment.