MP82-15 DOES EARLY ANDROGEN DEPRIVATION THERAPY AFTER BIOCHEMICAL RECURRENCE FOLLOWING RADICAL PROSTATECTOMY INCREASE OVERALL SURVIVAL? RESULTS FROM SEARCH

Abstract

INTRODUCTION AND OBJECTIVES: Androgen deprivation therapy (ADT) isstandard therapy formenwithmetastaticdisease (mets)e whether they present with mets or mets develop after failed curative treatment. In the PSA era, a rising PSA is the first sign of recurrence after treatment andADT ismost often today begunprior tomets.Whether earlier initiation of ADT has a survival benefit remains hotly debated. We examined the relationship between early ADT and overall survival in men who had a biochemical recurrence (BCR) after radical prostatectomy (RP). METHODS: We retrospectively analyzed data on 1,053 patients from 6 VA hospitals in the SEARCH database who had a BCR after RP between 1988 and 2013. Early ADT was defined as receiving ADT when PSA 5 ng/mL. Thus, the reference group contained patients who never received ADT or received late ADT (PSA >5 ng/mL). Cox models were used to test the association between early ADT and time from BCR to overall survival.Modelswereadjusted for age, race, surgical center, yearof surgery,PSA,pathologicalGleasonscore, time fromsurgery to recurrence, pathological features, and radiation therapy. In sensitivity analyses, we redefined early ADT as pre-ADT PSA 10 ng/mL or 20 ng/mL and also compared results for patients who did not receive ADT until time of mets. RESULTS: Median follow-up after recurrence was 4.8 years. 219 patients (21%) hadearlyADTand834did not.Onunivariable analysis, early ADTwas associatedwith decreased risk of death (HR0.74, p1⁄40.010). After adjusting for imbalances between the two groups, early ADT remained associated with decreased risk of death (HR 0.72, p1⁄40.021). When early ADTwas redefined as pre-ADT PSA 10 ng/mL, this association remained on multivariable analysis (HR 0.79, p1⁄40.045). However, when early ADT was defined as pre-ADT PSA 20 ng/mL, the association between early ADTandoverall survival was no longer significant (HR0.83, p1⁄40.107). ADT did not show a statistically significant survival benefit among those who received ADT at time of mets (HR 0.74, p1⁄40.152). CONCLUSIONS: We found an overall survival benefit for men with BCR after RP if ADT was initiated before PSA reaches 10 ng/mL, and this benefit was stronger when ADT was initiated before PSA reaches 5 ng/mL. While the survival benefits of ADT must be weighed against long-term side effects, our data support treating with ADT before the time of mets in order to increase overall survival. Further studies are warranted to support these findings.