# Genomic prostate score and treatment selection in favourable intermediate‐risk prostate cancer

## TL;DR

This study shows that a genomic test called GPS helps decide treatment for prostate cancer patients, with lower GPS scores linked to more patients choosing active surveillance.

## Contribution

The study demonstrates that GPS results and percent positive cores influence treatment selection in favorable intermediate-risk prostate cancer patients.

## Key findings

- 31% of patients selected active surveillance, with higher rates for lower GPS scores (58% for GPS 0–19 vs. 6% for GPS 41–100).
- GPS results and percent positive cores remained significant predictors of active surveillance selection in multivariable analysis.
- AS persistence was 91% at 12 months among selected patients.

## Abstract

To assess the factors associated with the use of active surveillance (AS) in NCCN favourable intermediate‐risk (FIR) prostate cancer (PCa) patients who received the 17‐gene Genomic Prostate Score (GPS) assay.

Contemporary data were collected from academic and large community group practices across the United States. Eligible patients had localized PCa classified as FIR per NCCN guidelines and received a GPS report between May 2017 and April 2019. Higher GPS results (scale: 0–100) were associated with a higher risk of adverse outcomes. The proportion of patients selecting AS was calculated with 95% confidence intervals. Uni‐and multivariable logistic regression analyses were performed to determine the association between AS selection and relevant covariates.

There were 324 eligible patients (Gleason Score 3 + 4, 79%; PSA 10–20 ng/ml, 19%; clinical stage T2b‐T2c, 2%; median percent positive cores, 16.7%; median GPS result, 26). The distribution of GPS results was 0–19 (23%), 20–40 (60%), and 41–100 (16%). Overall, 31% (95% CI 26%, 36%) selected AS: 58% (46%, 69%) with GPS 0–19, 27% (21%, 33%) with GPS 20–40, and 6% (1%, 16%) with GPS 41–100. In univariable models, the Gleason score, percent positive cores, PSA, and GPS results were significantly associated with AS selection. In a multivariable model, the percent positive cores and the GPS result remained significantly associated with AS selection. AS persistence was 91% (82%, 95%) at 12 months.

The GPS result and percent positive cores appear associated with AS use after controlling for relevant clinical variables in NCCN FIR prostate cancer patients.