1296 LOGISTIC REGRESSION MODELING OF FACTORS AFFECTING RATES OF HIGH GRADE PROSTATE CANCER IN THE REDUCTION BY DUTASTERIDE OF PROSTATE CANCER EVENTS (REDUCE) STUDY

Abstract

You have accessJournal of UrologyProstate Cancer: Localized1 Apr 20111296 LOGISTIC REGRESSION MODELING OF FACTORS AFFECTING RATES OF HIGH GRADE PROSTATE CANCER IN THE REDUCTION BY DUTASTERIDE OF PROSTATE CANCER EVENTS (REDUCE) STUDY Gerald L. Andriole, Donald J. Tindall, Tuevo L.J. Tamella, Matthew C. Somerville, and Roger S. Rittmaster Gerald L. AndrioleGerald L. Andriole St Louis, MO More articles by this author , Donald J. TindallDonald J. Tindall Rochester, MN More articles by this author , Tuevo L.J. TamellaTuevo L.J. Tamella Tampere, Finland More articles by this author , Matthew C. SomervilleMatthew C. Somerville Research Triangle Park, NC More articles by this author , and Roger S. RittmasterRoger S. Rittmaster Research Triangle Park, NC More articles by this author View All Author Informationhttps://doi.org/10.1016/j.juro.2011.02.996AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookTwitterLinked InEmail INTRODUCTION AND OBJECTIVES In the REDUCE study, dutasteride (DUT) compared with placebo (PBO) reduced the risk of biopsy-detectable prostate cancer (PCa) by 23% in men at increased risk of PCa. However, there was no significant difference in the incidence of Gleason 7–10 PCa (6.7% DUT, 6.8% PBO). We developed a multivariate logistic model to evaluate the effect of baseline and post-baseline variables on the incidence of high grade tumors (HGTs) in REDUCE. METHODS REDUCE was an international randomized, double-blind, PBO-controlled study. Men received daily DUT (0.5 mg) or PBO for 4 yrs. Eligible subjects included men 50–75 yrs, with prostate specific antigen (PSA) level 2.5–10 ng/ml and a single negative prostate biopsy (6–12 cores) within 6 months of enrollment. Biopsies (10 cores) were performed at 2 and 4 yrs or when clinically indicated. An analysis using stepwise covariate selection and baseline parameters was done to determine factors independently associated with risk of PCa. An additional model included post-baseline prostate volume (PV) at biopsy. RESULTS Baseline predictors of PCa that were independently associated with an increased risk of PCa were increased age, low PV, low percentage of free PSA, low number of cores at entry biopsy and family history of PCa. With these covariates, the odds ratio (OR) estimates for biopsy-detected PCa in DUT-treated men (vs PBO) was 0.74 (95% confidence interval [CI], 0.66–0.82; p<0.0001) for all tumors and 0.92 (95% CI, 0.76–1.12; p=0.42) for Gleason 7–10 tumors. The additional model demonstrated that post-baseline PV at time of biopsy was highly significant (p<0.0001), leading to a decrease in OR to 0.62 (95% CI, 0.49–0.79; p=0.0001) for HGTs. This significant treatment effect is seen in the rates of HGTs by PV at biopsy (Table). Table. Rate of Gleason score 7–10 cancer by prostate volume at biopsy Prostate volume at biopsy (cc) Placebo (n=3424) Dutasteride (n=3305) <20 10/109(9.2%) 35/383(9.1%) 20 to<30 46/431(10.7%) 56/966(5.8%) 30 to<40 42/716(5.9%) 45/1061(4.2%) 40 to<50 38/857(4.4%) 22/790(2.8%) 50 to<60 17/777(2.2%) 11/465(2.4%) 60 to<70 16/593(2.7%) 4/266(1.5%) 70 to<80 11/393(2.8%) 0/128(0.0%) <>80 12/543(2.2%) 0/107(0.0%) Missing 41/648(6.3%) 47/566(8.3%) CONCLUSIONS In REDUCE there was no significant difference in the incidence of Gleason 7–10 tumors between the DUT and PBO groups. However, a logistic regression model that included baseline covariates and PV at biopsy found the calculated OR for Gleason 7–10 tumors to be 0.62, demonstrating low PV to be a significant risk factor for HGTs. When comparing rates of high grade PCa between treatments by PV at biopsy, the PBO-treated subjects demonstrated a higher rate of PCa (except the 50–<60 cc subgroup). © 2011 by American Urological Association Education and Research, Inc.FiguresReferencesRelatedDetails Volume 185Issue 4SApril 2011Page: e518-e519 Peer Review Report Advertisement Copyright & Permissions© 2011 by American Urological Association Education and Research, Inc.MetricsAuthor Information Gerald L. Andriole St Louis, MO More articles by this author Donald J. Tindall Rochester, MN More articles by this author Tuevo L.J. Tamella Tampere, Finland More articles by this author Matthew C. Somerville Research Triangle Park, NC More articles by this author Roger S. Rittmaster Research Triangle Park, NC More articles by this author Expand All Advertisement Advertisement PDF downloadLoading ...