The Decipher genomic classifier (GC) has demonstrated the ability to predict prostate cancer outcomes independently. Researchers sought to verify the GC in a randomized phase III trial of dose-escalated salvage radiotherapy (SRT) following radical prostatectomy.
In a phase III trial of 350 men with biochemical recurrence after radical prostatectomy who were randomly assigned to 64 Gy or 70 Gy without concurrent hormonal therapy or pelvic nodal RT, a clinical-grade whole-transcriptome assay was performed on radical prostatectomy samples obtained from patients enrolled in Swiss Group for Clinical Cancer Research (SAKK) 09/10. A predetermined statistical strategy was created to determine how the GC will affect clinical results. The main outcome was biochemical development; the secondary outcomes were clinical development and delay in hormone treatment. Age, T-category, Gleason score, post radical prostatectomy persistent prostate-specific antigen (PSA), PSA at randomization, and randomization arm were all adjusted in multivariable analyses to take competing hazards into account.
With a median follow-up of 6.3 years, the analytic cohort of 226 patients was typical of the whole experiment (interquartile range 6.1-7.2 years). The GC (high versus low-intermediate) was independently correlated with biochemical progress (subdistribution hazard ratio (sHR) 2.26, 95% CI1.42-3.60; P<0.001), clinical progress (HR 2.29, 95% CI 1.32-3.98; P=0.003), and hormone therapy use (sHR 2.99, 95% CI 1.55-5.76; P=0.001). Compared to GC low-intermediate patients, GC high patients had 5-year independence from biochemical advancement of 45% as opposed to 71%. Both the general cohort and individuals with lower vs. higher GC scores did not benefit from the dose increase.
The predictive value of the GC has been proven in this investigation, which is the first modern randomized controlled trial in patients treated with early SRT without concomitant hormone treatment or pelvic nodal RT. High-GC patients were more than twice as likely to develop biochemical and clinical progression and undergo salvage hormone treatment than lower-GC patients, independent of common clinicopathologic factors and RT dosage. These findings support the therapeutic utility of Decipher GC for individualized concurrent systemic treatment in the context of postoperative salvage.