Replacement) study (NCT01561053) is currently in progress to confirm the observed trends. Methods: In the AMBAR study (Phase IIB/III multicenter, randomized, patient-and rater-blinded), patients with mild-moderate AD are distributed in 4 groups (1:1:1:1) of PE treatment according to different doses of albumin replacement (Albutein Ò , Grifols) plus intravenous immunoglobulin (Flebogamma Ò DIF), and a control group (sham PE). Treatment periods are intensive (one total PE [TPE] weekly, 2.5-3 L plasma removal/albumin replacement, 6 weeks) and maintenance (one low-volume PE [LVPE] monthly, 650-880 mL plasma removal/ albumin or immunoglobulin; 12 months). Outcomes are the changes from baseline of scores obtained with the ADAS-Cog and ADCS-ADL tests. Other endpoints are assessed through MMSE, NPB, NPI, CDR-Sb, ADCS-CGIC, CSDD, C-SSRS, QoL-AD, and RUD-Lite Ò . Interim results in a global sample (randomization codes not broken) are analyzed. Results: Results are available for 186 patients (as of June 2015) in 36 sites (Spain and US) out of the 364 patients sample size. Patient median age was 70 years old (range: 53-85), with 49.3% males. The total number of PE performed was 1367 (652 TPE, 715 LVPE). PE was well tolerated and safe (adverse events profile similar to PE in other conditions). After TPE (2 months), 45.2% of patients showed improvement in MMSE according to change from baseline score. After LVPE, this percentage ranged from 33.3% at 6 months to 25.8% at 14 months. In ADAS-Cog scores, percentages of patients showing improvement were 33.7%, 45.7% and 26.7%, respectively, while in ADCS-ADL scores the same percentages were 50.9%, 40.0% and 41.4%, respectively. Conclusions: The AMBAR study is designed to confirm previous findings on cognitive function improvement in AD patients treated with PE plus albumin replacement, to open new perspectives in the treatment of AD. With randomization codes unbroken, main outcome measures showed promising global improvement/worsening rates.