LONGITUDINAL CHANGES IN PATIENT-REPORTED OUTCOMES (PROs) IN PATIENTS WITH HIGH-RISK BIOCHEMICALLY RECURRENT (BCR) NONMETASTATIC CASTRATION-SENSITIVE PROSTATE CANCER (NMCSPC) FROM THE EMBARK STUDY.

In the phase 3 EMBARK trial (NCT02319837), enzalutamide (ENZA)+leuprolide acetate (L) and ENZA alone significantly delayed metastasis-free survival (MFS) vs placebo (P)+L in high-risk BCR nmCSPC. The PRO analysis assessed pain progression using the Brief Pain Inventory Short-Form (BPI-SF) and health-related quality of life (HRQoL) using Functional Assessment of Cancer Therapy-Prostate (FACT-P), European QoL 5-Dimensions 5-Level (EQ-5D-5L), and European Organization for Research and Treatment of Cancer QoL Questionnaire-Prostate 25 (QLQ-PR25). The primary PRO analysis showed no significant differences in time to first and confirmed (at the next visit) clinically meaningful deterioration (TTFD/TTFCD) between either treatment arm and P+L for pain progression and HRQoL. The secondary objective of the EMBARK PRO study was to analyze the change from baseline in PRO scores for pain and HRQoL using a restricted maximum likelihood-based repeated measures approach (MMRM) in a longitudinal analysis. The findings of this secondary analysis are presented here. Patients with high-risk BCR nmCSPC (prostate-specific antigen [PSA] doubling time ≤9 months; screening PSA ≥2 ng/mL above nadir after radiotherapy or ≥1 ng/mL after radical prostatectomy) were randomized (1:1:1) to receive ENZA+L, ENZA alone, and P+L. At week 37, study treatment was suspended for patients whose PSA levels were undetectable (<0.2 ng/mL) at week 36. PROs were assessed at baseline and every 12 weeks until metastasis/death. Clinically meaningful deterioration was defined as a decrease of ≥10 points from baseline in the FACT-P total score and an increase of ≥2 points in the BPI-SF worst pain score. Analysis of observed data using the MMRM approach was limited to timepoints where 50% of the patients in each treatment arm had non-missing data. Separate models were considered for each PRO score. The primary focus of this analysis was to examine treatment differences up to week 205 in the intent-to-treat population. At baseline, 327–332 patients per group completed the PRO questionnaire; completion rates post-baseline up to week 205 were ≥85%. Patients were asymptomatic and experienced good HRQoL at baseline. No clinically meaningful deterioration in pain progression was observed up to week 205 with any treatment. For worst pain, overall differences between each treatment arm and P+L were not statistically significant (Figure 1A). For FACT-P total score, differences in change from baseline were statistically significant favoring P+L but did not meet the a priori threshold for clinically meaningful differences (Figure 1B). No significant differences in change from baseline were observed between arms for urinary symptoms and bowel symptoms/function. Statistically significant differences were observed in sexual activity favoring ENZA alone and hormone treatment-related symptoms favoring P+L (Figure 2). EQ-5D-5L visual analogue scale scores remained stable across all arms (Figure 1C). HRQoL scores showed trends of returning to baseline following week 37. Across multiple patient-reported surveys, patients in all arms reported high HRQoL scores at baseline, and no clinically meaningful differences in HRQoL were observed longitudinally up to week 205. When looking at specific subdomains, the ENZA-alone group experienced the smallest deterioration in the sexual activity domain, while the P+L group experienced the smallest deterioration in hormone-related symptoms. The EMBARK clinical and PRO analyses show that ENZA+L or ENZA alone improves MFS compared with P+L while maintaining HRQoL in patients with high-risk BCR nmCSPC. [ABSTRACT FROM AUTHOR]