This retrospective study evaluated the 17-gene Genomic Prostate Score (GPS) in 409 patients with clinically localised or locally advanced prostate cancer managed with active surveillance (AS), radical prostatectomy (RP), or radiotherapy (RT), with a median follow-up ≥6 years, to assess its potential for guiding management de-escalation.
Low GPS scores (present in 58% of the cohort) identified a 'discordant low-risk' phenotype—patients with NCCN ≥2 but low GPS—in 46% of NCCN≥2 AS, 57% of NCCN≥3 RP, 36% of NCCN≥4 RT, and 48% of NCCN5 locally advanced patients; 6-year event-free survival in these discordant low-risk groups was 88% (AS), 100% (RP and RT), and 95% (locally advanced), comparable to 100% 6-year treatment-free survival in concordant low-risk patients.
Retrospective design limits causal inference; no randomized de-escalation arm was tested; the study does not establish GPS-guided de-escalation as standard of care, only supports feasibility of future trials.
In patients with higher NCCN risk categories, a low GPS score may identify a substantial subset with favorable 6-year outcomes—consider GPS testing when results could influence the decision to de-escalate treatment intensity. Prospective trials are still needed before routine de-escalation based on GPS alone.
Explore related topics
