O4–03–07: Efficacy of rosiglitazone (RSG) in patients with mild to moderate Alzheimer's disease (AD)

Abstract

RSG (PPARγ agonist) is approved for type–2 diabetes treatment. Risner et al reported cognitive endpoints from a 24–week, double–blind, placebo–controlled monotherapy trial (N=511) in AD patients (baseline MMSE 16–26). No significant effects were observed on ADAS–Cog in the intention–to–treat (ITT) population; however, a significant interaction between treatment and APOE ϵ4 allele status on ADAS–Cog (P=0.014) was observed at Week 24 Last Observation Carried Forward (LOCF). Exploratory analyses suggested that APOE ϵ4 non–carriers showed significant clinical improvement whereas APOE ϵ4 allele carriers showed no improvement and some decline was noted. RSG was generally well tolerated. No new safety concerns were identified compared to the well–established safety profile of RSG. To report results of RSG on cognition, behavioral symptoms, and functional ability in mild–to–moderateAD patients in both the ITT population and by APOE ϵ4 carriage status. Patients were randomized to receive placebo or an unmarketed form of RSG 2, 4, or 8mg/day for 24 weeks. Cognitive symptoms, behavioral symptoms, and function were assessed using ADAS–Cog, NPI, and CIBIC–Plus and DAD, respectively. An analysis of covariance model was used based on the ITT and pharmacogenetic populations using an LOCF approach. Observed case (OC) analyses were also performed. No significant treatment effects were observed at Week 24 in the ITT population for CIBIC–Plus, NPI, DAD or MMSE. Exploratory analyses (not adjusted for multiplicity) of ADAS–Cog at Week 24 demonstrated improvement in the “Word recognition” item in RSG 8 mg group (P=0.046 OC; P=0.14 LOCF). At Week 24, RSG 8 mg group also showed significant improvement in CIBIC–Plus (P=0.021 OC; P=0.11 LOCF) and in the irritability/lability domain of NPI (P=0.018 LOCF). Exploratory analyses suggested that the pattern observed on ADAS–Cog (improvement in APOE ϵ4 non–carriers and lack of improvement or decline for APOE ϵ4 allele carriers) was also observed at Week 24 for CIBIC–Plus, DAD, and NPI (Figure). These exploratory findings cannot be used to inform clinical management of patients with AD. Further investigation of rosiglitazone for symptomatic treatment of patients with mild–to–moderate AD is warranted. Risner et al. The Pharmacogenomics Journal advance online publication, 31 January 2006; doi:10.1038/sj.tpj.6500369.