Importance: The use of second-generation antiandrogens (AAs) in the treatment of prostate cancer is increasing. Retrospective evidence suggests an association between second-generation AAs and adverse cognitive and functional outcomes, but further data from prospective trials are needed.
Objective: To examine whether evidence from randomized clinical trials (RCTs) in prostate cancer supports an association between second-generation AAs and cognitive or functional toxic effects.
Data sources: PubMed, EMBASE, and Scopus (inception to September 12, 2022).
Study selection: Randomized clinical trials of second-generation AAs (abiraterone, apalutamide, darolutamide, or enzalutamide) among individuals with prostate cancer that reported cognitive toxic effects, asthenic toxic effects (eg, fatigue, weakness), or falls were evaluated.
Data extraction and synthesis: Study screening, data abstraction, and bias assessment were completed independently by 2 reviewers following the Preferred Reporting Items for Systematic Reviews and Meta-analyses and Enhancing the Quality and Transparency of Health Research reporting guidelines. Tabular counts for all-grade toxic effects were determined to test the hypothesis formulated before data collection.
Main outcomes and measures: Risk ratios (RRs) and SEs were calculated for cognitive toxic effects, asthenic toxic effects, and falls. Because fatigue was the asthenic toxic effect extracted from all studies, data on fatigue are specified in the results. Meta-analysis and meta-regression were used to generate summary statistics.
Results: The systematic review included 12 studies comprising 13 524 participants. Included studies had a low risk of bias. An increased risk of cognitive toxic effects (RR, 2.10; 95% CI, 1.30-3.38; P = .002) and fatigue (RR, 1.34; 95% CI, 1.16-1.54; P < .001) was noted among individuals treated with second-generation AAs vs those in the control arms. The findings were consistent in studies that included traditional hormone therapy in both treatment arms for cognitive toxic effects (RR, 1.77; 95% CI, 1.12-2.79; P = .01) and fatigue (RR, 1.32; 95% CI, 1.10-1.58; P = .003). Meta-regression supported that, across studies, increased age was associated with a greater risk of fatigue with second-generation AAs (coefficient, 0.75; 95% CI, 0.04-0.12; P < .001). In addition, the use of second-generation AAs was associated with an increased risk of falls (RR, 1.87; 95% CI, 1.27-2.75; P = .001).
Conclusions and relevance: The findings of this systematic review and meta-analysis suggest that second-generation AAs carry an increased risk of cognitive and functional toxic effects, including when added to traditional forms of hormone therapy.