Diabetes And Cardiovascular Mortality In Men With Locally Advanced Prostate Cancer: Updated Analysis From Rtog 92-02

When combined with radiation therapy (RT), androgen deprivation therapy (ADT) with a gonadotropin-releasing hormone (GnRH) agonist leads to improved survival in locally advanced prostate cancer, but are associated with incident diabetes (DM) and cardiovascular (CV) disease. We provide longer-term follow-up on mortality outcomes from a large randomized trial assessing ADT duration in locally advanced prostate cancer. From 1992-1995, 1,554 men with prostate cancer (T2c-4, PSA <150 ng/mL) received RT and either 4 or 28 months of ADT with goserelin on RTOG 92-02. Proportional hazard models were used to analyze the relationship between treatment arm and CV mortality, and between prevalent DM and mortality outcomes. Covariates included treatment arm, age, race, stage, Gleason score, PSA, prevalent CV disease, hypertension, prevalent DM, and weight. After median follow-up of 11.3 years for surviving patients, there were 826 deaths, 218 (26%) due to prostate cancer and 197 (24%) due to CV disease. There were 153 deaths due to CV disease when censoring at time of salvage ADT. On multivariate analysis (MVA), longer-term ADT was associated with decreased prostate cancer mortality (PCM) versus short-term ADT (adjusted hazard ratio [HR] = 0.67; 95% confidence interval [CI] 0.50-0.89; p = 0.006). In contrast, prevalent DM was associated with greater all-cause mortality and non-prostate cancer mortality (NPCM), but not PCM. While weight, a risk factor for DM, was associated with greater PCM on MVA in a previous analysis (HR = 1.77; 95% CI 1.22-2.55; p = 0.002), this association was not statistically significant with longer follow-up (HR = 1.46; 95% CI 0.95-2.50; p = 0.08). CV mortality for men who received longer-term versus short-term ADT was 12.3% versus 9.7% at 10 years (Gray's p = 0.06). On MVA with and without censoring, duration of ADT was not associated with increased CV mortality, while conventional cardiac risk factors including age, prevalent CV disease, and prevalent DM were associated with greater CV mortality (Table). Results were similar with alternative definitions of CV mortality. With additional follow-up, longer-term GnRH agonist therapy continues to decrease PCM in locally advanced prostate cancer without increasing CV mortality, alleviating concerns about the safety of ADT. Neither prevalent DM nor weight was associated with greater PCM in men receiving combined RT and ADT.Oral Scientific Abstract 231; TableMultivariate analyses of CV mortalityCovariateComparisonWithout Censoring (n = 197)With Censoring (n = 153)HR(95% CI)p valueHR(95% CI)p valueTreatment ArmShort-term vs Longer-term ADT-1.18(0.88-1.58)-0.28-1.19(0.85, 1.66)-0.31Age<70 vs ≥70-2.00(1.45-2.75)-<0.0001-1.94(1.34, 2.81)-0.0004RaceBlack vswhite/Other-1.60(0.94-2.74)-0.085-1.47(0.80, 2.68)-0.22Prevalent CV DiseaseNo vs Yes-2.64(1.96-3.56)-<0.0001-2.45(1.75, 3.43)-<0.0001Prevalent HypertensionNo vs Yes-1.10(0.81-1.50)-0.54-1.15(0.81, 1.62)-0.44Prevalent DMNo vs Yes-1.87(1.29-2.70)-0.0008-1.94(1.30, 2.90)-0.0012Gleason Score2-6 vs 7-10-1.05(0.78-1.41)-0.75-0.94(0.67, 1.31)-0.71Clinical StageT2 T3 T4-1.031.04(0.77-1.39)(0.42-2.59)-0.850.93-1.170.88(0.84, 1.63)(0.27, 2.82)-0.350.83PSAContinuous1.00(1.00-1.01)0.631.00(1.00, 1.01)0.57 Open table in a new tab