Dextromethorphan/quinidine (AVP-923) efficacy and safety for treatment of agitation in persons with alzheimer's disease: Results from a phase 2 study (NCT01584440)

Introduction: Neuropsychiatric symptoms such as aggression/agitation are common in some patients with AD, adversely impact patient and caregiver quality of life and increase risk for institutionalization. No FDA-approved treatments exist for agitation in AD. AVP-923, a combination of dextromethorphan and low-dose quinidine (DM/Q), is being investigated for ADrelated agitation. The CNS active component, dextromethorphan, is an uncompetitive NMDA antagonist, sigma-1 agonist, and serotonin/norepinephrine reuptake inhibitor. Methods: This was a multicenter, randomized, double-blind, placebo-controlled, 2-stage study using a Sequential Parallel Comparison Design (SPCD) method. Eligible patients had probable AD, clinically meaningful agitation [characterized by a Clinical Global Impression-Severity (CGI-S) score ≥4 (moderately ill)], and Mini-Mental State Examination (MMSE) score of 8-28. In stage 1 (weeks 1-5), patients were randomized (4:3) to placebo or AVP-923 dosed as follows: week 1: 20/10 mg QD, weeks 2-3: 20/10 mg BID, weeks 4-5: 30/10 mg BID. In stage 2 (weeks 6-10), patients randomized to AVP-923 continued at the same dose; those initially on placebo were stratified according to response, then re-randomized 1:1 to placebo or AVP-923 (titrated as in stage 1). Placebo response was defined as CGI-S agitation score ≤3 [minimally ill] and ≥25% reduction in the Neuropsychiatric Inventory [NPI] agitation/aggression domain; participants not meeting these criteria were classified as placebo nonresponders. The primary endpoint was change from baseline on the NPI agitation/aggression domain based on standard SPCD methodology that includes data from stage 1 plus the placebo nonresponders that were re-randomized in stage 2. Secondary endpoints included change in total NPI, individual NPI domains, NPI-4D, NPI-4A, NPI-Caregiver Distress Score, Alzheimer's Disease Cooperative Study-Clinical Global Impression of Change (ADCS-CGIC), Patient Global Impression of Change (PGI-C), Quality of Life-Alzheimer's disease (QOL-AD), ADCS-Activities of Daily Living (ADL), Caregiver Strain Index (CSI), MMSE, and Cornell Scale for Depression in Dementia (CSDD). Standard safety and tolerability outcomes were assessed. Stage 1 and stage 2 results were also analyzed separately. Results: 220 patients were enrolled and 194 (88%) completed the study. In stage 1, 218 patients were included in the modified intent-to-treat (mITT) efficacy population [AVP-923 (n=93); placebo (n=125)]. In stage 2, 202 patients comprised the mITT: 83 patients from the stage 1 AVP-923 group continued on AVP-923; 89 placebo nonresponders were re-randomized [AVP-923 (n=44); placebo (n=45)], and 30 placebo responders were re-randomized [AVP-923 (n=15); placebo (n=15)]. The primary endpoint was significant for AVP-923 vs placebo (P≤0.001). The mean (SD) baseline NPI agitation/aggression domain scores were 7.1 (2.6) for AVP-923 vs. 7.0 (2.4) for placebo, and showed significant improvement for AVP-923 vs placebo both at week 5 (stage 1, mean [SD] change from baseline -3.3 [2.98] vs -1.7 [3.10]; (P≤0.001) and at week 10 (stage 2, mean [SD] change from baseline -2.0 [3.19] vs -0.8 [3.59]; P=0.02). ADCS-CGIC agitation improved significantly with AVP-923 vs placebo on the primary analysis (P≤0.001), as did PGI-C. Significant improvement for AVP-923 vs. placebo was also seen for secondary endpoints including: NPI total score, NPI-4D, NPI-4A, NPI Caregiver Distress Score, CSDD, and CSI. Treatment-emergent adverse events (TEAEs) occurred in 61.2% of patients taking AVP-923 and 43.3% of patients taking placebo and were considered to be at least possibly related to treatment in 16.4% and 10.2%, respectively. Adverse events led to discontinuation in 5.3% on AVP-923 and 3.1% on placebo. Serious AEs (SAEs) occurred in 7.9% and 4.7% of patients, respectively; no SAEs were considered treatment-related. No deaths occurred during the study. Falls were reported as an AE in 8.6% of patients on AVP-923 and 3.9% on placebo; however, history of falls was more common in the AVP-923 group at baseline (20.4% vs. 12.8%). Other AEs occurring in ≥5% of patients were diarrhea 5.9% vs. 3.1%, and urinary tract infection 5.3% vs. 3.9%. Conclusions: AVP-923 (dextromethorphan/quinidine) was associated with significant improvement in agitation in patients with AD as seen on the primary, and on the majority of secondary endpoints. These improvements were deemed clinically meaningful by patients/caregivers and physicians (as assessed by PGI-C and ADCS-CGIC ratings), and were associated with reduction in caregiver burden. AVP-923 was generally well tolerated over this 10 week trial.