The potential role of frailty in modifying quality-of-life related outcomes in metastatic castration-resistant prostate cancer (mCRPC) patients receiving docetaxel, abiraterone, enzalutamine, and radium 223.

Abstract

63 Background: As the arsenal of therapeutic agents for mCRPC expands, frailty-informed care is emerging as a strategy for tailoring management decisions. Yet, data to guide this approach and to ascertain its impact on the mood, fatigue, pain, and quality of life (QoL) experienced by older men with mCRPC remain lacking. We examined patient-reported outcomes, stratified by a validated frailty index (FI), for mCRPC treatment with docetaxel chemotherapy (CHEMO), abiraterone (ABI), enzalutamide (ENZA), or radium 223 (RAD). Methods: Older (aged 65+) men starting one of four approved therapies for mCRPC were enrolled in a multicenter prospective cohort study. Assessing their mood, fatigue, pain, and QoL with PHQ-9, ESAS tiredness, ESAS pain, and FACT-G total as well as subscale scores, we used linear mixed effect models to determine change in each outcome over time (0, 3, 6 months). At end of treatment, we administered the Decisional Regret Scale. We then constructed a FI from 34 variables that span laboratory abnormalities, geriatric syndromes, instrumental activities of daily living, social support, as well as emotional, cognitive, and physical deficits. Following established cut-offs, we categorized patients as non-, pre-, and frail, then performed stratified linear mixed effects regression analyses to identify differences in outcomes by frailty status. Results: A total of 198 men (mean age 75.1) starting CHEMO (n = 70), ABI (n = 38), ENZA (n = 67), and RAD (n = 29) were included, of which 9.6%, 1.5%, 5.6%, and 2.5%, respectively, were determined frail. Frailty correlated only modestly with age (Pearson r = 0.27). Independent of frailty status, patients across treatment cohorts were similar in terms of baseline QoL-related measures, with the exceptions of mood (p = 0.033) and pain (p = 0.034). Over time, no significant change in QoL was reported, although all four therapies resulted in generally low levels of decisional regret and similar trends of improved pain but worsened mood (p = 0.006 and 0.02, respectively). At baseline, frailty status correlated with worse FACT-G total (p < 0.001) and functional well-being (p < 0.001), as well as worse depression scores (p < 0.001). According to FI-stratified analysis, frail patients experienced similar QoL-related outcomes to fit patients for all measures aside from mood (p < 0.001). Contrary to our hypotheses, frailty was not associated with significant worsening in emotional well-being or functional well-being in response to mCRPC treatment. Conclusions: Of the older men receiving care for mCRPC, frail patients may experience generally similar trends in QoL as fit patients. Interestingly, frailty status, rather than treatment modality, may play more of a contributory role to changes in QoL-related outcomes over time.