Baseline Perineural Invasion is Associated with Shorter Time to Progression in Men with Prostate Cancer Undergoing Active Surveillance: Results from the REDEEM Study

Abstract

We evaluated the association of perineural invasion with disease progression in men with prostate cancer on active surveillance. We retrospectively analyzed the records of 302 men on active surveillance for low risk prostate cancer (T1c-T2a), Gleason 6 or less, 3 or fewer positive cores, 50% or less of any core involved and prostate specific antigen 11 ng/ml or less in the REduction by Dutasteride of clinical progression Events in Expectant Management (REDEEM) study. Patients underwent study mandated biopsies 18 and 36 months after enrollment. Disease progression was divided into pathological (4 or greater positive cores, 50% or greater core involvement, or Gleason greater than 6 on followup biopsy), therapeutic (any therapeutic prostate cancer intervention) or clinical (pathological or therapeutic progression). Time to disease progression was analyzed with Cox models adjusting for patient age, race, baseline prostate specific antigen, number of sampled and involved cores, tumor length and treatment. A total of 11 patients (4%) had perineural invasion on baseline biopsy. Perineural invasion was not associated with any baseline features (each p>0.05). During the study clinical progression developed in 125 patients (41%), including pathological progression in 95. One, 2 and 3-year clinical progression-free survival for men with vs without perineural invasion was 82%, 27% and 27% vs 93%, 67% and 58%, respectively (p <0.05). On multivariable analyses perineural invasion was associated with clinical (HR 2.39, 95% CI 1.16-4.94, p=0.019) and pathological progression (HR 2.21, 95% CI 0.92-5.33, p = 0.076). Among patients with prostate cancer on active surveillance perineural invasion was associated with an increased risk of clinical progression. The 2-year risk of clinical progression with perineural invasion was 73%. If these results are confirmed, patients with perineural invasion may not be good active surveillance candidates.