Comparative fall risk of patients treated with novel androgen receptor antagonists in prostate cancer: a systematic review and meta-analysis
摘要
Androgen receptor pathway inhibitors (ARPIs) are cornerstone treatments for advanced prostate cancer; however, their potential to increase fall risk remains a significant clinical concern. This meta-analysis aims to provide a rigorous, drug-specific evaluation of the association between novel ARPIs and the risk of falls.
MethodsWe systematically searched PubMed, Web of Science, Cochrane Library, and ClinicalTrials.gov for Phase 2 or 3 randomized controlled trials (RCTs) comparing ARPIs (enzalutamide, apalutamide, and darolutamide) with control groups (placebo or non-steroidal antiandrogens [NSAA]). The primary outcomes were risk ratios (RRs) for all-grade and grade ≥ 3 falls. To account for multiplicity across correlated outcomes and the limited number of studies, pooled RRs were estimated using random-effects models with the restricted maximum-likelihood (REML) method. All analyses were performed on a logarithmically transformed scale, with 97.5% confidence intervals (CIs). Prediction intervals (PIs) were calculated to assess the dispersion of effects. Subgroup analyses were stratified by specific ARPI agents, control types, and clinical stages.
ResultsEleven RCTs involving 12,239 patients were included. Overall, ARPIs were significantly associated with the increased risk of all-grade falls (RR 2.00, 97.5% CI 1.46–2.73, P < 0.0001, I2 = 77.6%; PI 0.67–5.99) and grade ≥ 3 falls (RR 2.15, 97.5% CI 1.32–3.52, P = 0.0008, I2 = 0%; PI 1.15–4.02). However, risk profiles varied substantially across individual agents. Enzalutamide was associated with the highest risk increase (RR 2.55 vs. placebo, 97.5% CI 1.62–4.01, I2 = 79.2%; RR 2.47 vs. NSAA, 97.5% CI 1.14–5.37, I2 = 71.8%), followed by apalutamide (RR 1.65, 97.5% CI 0.77–3.52, I2 = 87.2%). In contrast, darolutamide demonstrated a favorable safety profile with no statistically significant increase in the risk of all-grade falls (RR 1.25, 97.5% CI 0.87–1.79, I2 = 0%) or severe falls (RR 1.31, 97.5% CI 0.34–5.00).
ConclusionsCurrent evidence indicates that the increased risk of falls associated with ARPI therapy varies significantly among individual agents, rather than being a uniform class effect. While enzalutamide and apalutamide are statistically associated with elevated fall risk, darolutamide appears to maintain a more favorable safety profile. However, these drug-specific comparisons remain exploratory due to subgroup imbalances. Clinicians should consider proactive fall-risk assessments and individualized treatment selection, particularly for elderly or frail populations.