Background: PF-04447943 is a potent, selective phosphodiesterase 9A (PDE9A) inhibitor that elevates guanoscine 3’,5’ – cyclic monophosphate (cGMP) in brain and cerebrospinal fluid. PDE9A inhibition enhances synaptic plasticity and improves memory in preclinical cognition models, and prevents decreases in dendritic spine density in transgenic mice that overexpress amyloid precursor protein (APP) leading to high levels of amyloid beta (Aβ) production (Tg2576). Objective: This Phase 2 multicenter study was designed to assess the efficacy, safety and pharmacokinetics of PF-04447943 compared with placebo in mild to moderate probable Alzheimer’s disease (AD). Methods: Subjects in overall good health with Mini Mental State Examination (MMSE) scores of 14-26 were randomized to 12 weeks treatment with PF-04447943 25 mg q12h (n=91) or placebo (n=100). Concomitant acetylcholinesterase inhibitor or memantine use was excluded. The primary outcome was the Alzheimer’s Disease Assessment Scale – cognitive subscale (ADAS-cog). The Neuropsychiatric Inventory (NPI), Clinical Global Impression-Improvement scale (CGI-I) and standard safety measures were secondary outcomes. Results: Completion rates were similar, 87% PF-04447943 vs 92% placebo. At week 12 the mean (SE) baseline adjusted decrease from baseline in ADAS cog for PF-04447943-treated patients was -1.91 (0.54). Placebo treated patients had a change of -1.60 (0.50). The difference between treatments was -0.31 (90% CI of -1.52, 0.90). Corresponding values for the NPI were -2.86 (0.72) vs -2.70 (0.67) with a treatment difference of -0.16 (90% CI of -1.78, 1.48). Neither these changes nor the distribution of CGI-I scores were statistically significantly different between groups. The incidence of serious adverse events (AEs) was similar between groups with 2 deaths in the placebo group. The PF-04447943 group reported more gastrointestinal AEs including diarrhea (5.5% vs 3%) and nausea (5.5% vs 1%) and had a higher rate of discontinuation due to AEs (6.6% vs 2%). Conclusions: Although generally safe and well-tolerated, 12 weeks PF-04447943 treatment did not improve cognition, behavior, and global change compared with placebo.