Quality of Life and Weight Variation during Androgen Deprivation Therapy in Patients with High-Risk Prostate Cancer: Long-Term Data from a Phase III Trial

We assessed the impact of weight (wt) variation on quality of life (QoL). Using data from a Phase III trial in high-risk prostate cancer (HRPC) and based on Patient Reported Outcomes (PRO), we compared long-term QoL between HRPC patients (pts) with wt gain or loss during initial androgen deprivation therapy (ADT). From 10/2000 to 01/2008, 630 pts were randomized to 36 (310 pts) vs. 18 months (m) (320 pts) of ADT. Wt was measured at baseline and regularly during ADT. We assessed QoL by the validated EORTC 30 items questionnaire regrouped into 9 scales. PRO was measured up to 6.5 and 7 years for pts receiving 18 and or 36 m of ADT, respectively. All items and scales scores were linearly transformed to a 0 to100 points scale. We estimated means and standard deviation of items and scales for each group at each time point. We analyzed all items and scales scores with general linear model with repeated measures to evaluate changes between pts who gain vs. lost wt over time, in both ADT groups. P- value <0.01 was considered statistically significant to account for multiple comparisons and a difference in mean scores of ≥10 points was considered clinically relevant. 311/630 pts were retained for the analysis. Reasons for exclusion: not exact dose of ADT (96), no wt at baseline (89) or at the end of ADT (130), no QoL at baseline or at 6 m (4). During ADT course, 209/311 (67.2%) pts gained and 102/311 (32.8%) lost wt. There was no significant difference in the mean ± SD wt gain at the end of ADT in pts receiving 36- or 18 m (2.57 ± 5.39 vs. 1.93 ± 4.22, p = 0.27). Wt gain over 2 kg was similar between cohorts (50% vs. 52.8%, p = 0.62), but wt gain over 4 kg was lower in the18 m cohort (27.1% vs. 37.4%, p = 0.056). Wt loss was not significant between cohorts, irrespective of amount lost. Wt loss impacted on overall survival in the 36 m cohort [HR (95% CI) = 1.93 (1.24-2.99), p = 0.004], although no significant difference was seen in the 18 m cohort [HR = 1.18 (0.81-1.72), p = 0.39]. Considering the unified 30 items, overall adherence to QoL questionnaires was 83.1% (2887/3309). It was 93.6% vs. 95.9% at baseline and 70.7% vs 87.0% at 5 years, for 18 vs. 36 m, respectively. Wt loss impacted on QoL. Significant differences (p<0.01) were observed for 8 items and 2 scales related to overall QoL, including emotional, cognitive, fatigue and appetite loss, and financial difficulties in the 18 m cohort. Similarly, significant differences (p<0.01) were observed for 8 items and 6 scales related to physical, role, emotional, cognitive, and dyspnea in the 36 m cohort. Likewise, several items reached clinical relevance in the 18 (overall QoL, memory) and in the 36 m cohort (role, dyspnea, limited physical activity, fatigue). In HRPC pts treated with long-term ADT, wt loss during ADT is associated with poorer outcomes in QoL in several domains and a worst survival in the 36 m cohort.