IntroductionIstradefylline offers a novel mechanism (adenosine A2A receptor antagonism) to treat OFF episodes in Parkinson's disease (PD). It may potentially offer improved tolerability versus other adjuncts, but comparative safety data are lacking.MethodsA systematic review and Bayesian network meta-analysis (NMA) incorporating RCTs of PD adjuncts until January 10, 2024, was conducted to estimate relative safety. Inconsistency was assessed and heterogeneity evaluated by global I2-statistic and between-study heterogeneity. Incidences of safety outcomes were summarized from RWE identified according to the same criteria.Results100 RCTs and 55 RWE publications were identified; 76 RCTs were included in NMAs. Istradefylline demonstrated lower odds of serious AEs (odds ratio [OR] = 0.56; 95 % CrI: 0.32, 0.99), treatment-emergent AEs (0.43; 0.25, 0.73), treatment-related AEs (0.33; 0.19, 0.56), hallucinations (0.25; 0.06, 0.97), and withdrawal due to AEs (0.37; 0.19, 0.68) versus amantadine. Istradefylline showed lower odds of dyskinesia (0.63; 0.41, 0.99) and hypotension (0.19; 0.03, 0.82) versus catechol-O-methyl transferase inhibitors (COMTi), lower odds of nausea (0.58; 0.33, 0.99) versus dopamine agonists (DA), and lower odds of hypotension (0.09; 0.01, 0.52) versus monoamine oxidase-B inhibitors (MAO-Bi). Sensitivity analysis of RCTs published since 2000 found a reduction in odds of dyskinesia and hallucinations for istradefylline versus DA. RWE were heterogeneous but demonstrated lower incidence of certain AEs with istradefylline, specifically dyskinesia (versus MAO-Bi), somnolence (versus DA and COMTi), peripheral edema and hallucinations (versus amantadine), and nausea (versus all comparators).ConclusionIstradefylline exhibits a favorable safety profile versus other PD adjuncts, as demonstrated by RCTs and RWE.