Peroxisome proliferator-activated receptor gamma (PPAR-gamma) has been described to have a role in the modulation of various genes involved in Abeta homeostasis, inflammation, and energy metabolism, making it a candidate gene for risk of Alzheimer's disease (AD). A functional polymorphism in exon 2 of the PPAR-gamma gene has been related to AD, but the effects are inconsistent across studies. To determine the role of PPAR-gamma in genetic susceptibility to AD in a representative Chinese sample, we genotyped 362 AD patients and 370 healthy controls for PPAR-gamma Pro12Ala polymorphism by polymerase chain reaction-restriction fragment length polymorphism method. We also examined the potential impact of this polymorphism on plasma level of soluble receptor for advanced glycation end products (sRAGE), a decoy receptor whose reduction has been associated with a higher risk of AD. Our results suggest that PPAR-gamma Pro12Ala polymorphism was not associated with an increased risk of AD in the overall sample. Stratification analysis revealed that the PPAR-gamma Pro/Ala genotype may be associated with the development of early-onset AD in the individuals without APOE epsilon4 allele (OR=3.76, 95% CI=1.10-12.84; p=0.03), but this association became insignificant after Bonferroni correction (p (corr)=0.10). Moreover, in the subgroup of APOE epsilon4 noncarriers, Kaplan-Meier survival analyses indicated that AD patients with the Pro/Ala genotype presented with disease onset 4.6 years earlier than carriers of Pro/Pro genotype. Further investigation revealed that AD patients carrying Pro/Ala genotype had significantly lower plasma sRAGE levels than patients with Pro/Pro genotype. These findings suggest that the functional PPAR-gamma Pro12Ala polymorphism may modify the age at onset of AD.