EVIDENCE THAT ORAL P38 MAPK ALPHA ANTAGONISM IMPROVES EPISODIC MEMORY IN PATIENTS WITH EARLY ALZHEIMER’S DISEASE (AD)

Abstract

Neflamapimod is an oral antagonist of p38α, a kinase implicated in hippocampal dysfunction in preclinical studies. Two phase 2a clinical trials with neflamapimod in early AD (reported at CTAD, Dec 2016) met their primary objectives by demonstrating disease-relevant pharmacologic activity; and secondarily demonstrated improvement in measures of episodic memory. We report herein further analyses of the episodic memory findings. Subjects: patients with MCI due to AD or mild AD (MMSE 20–28). Study 302: 15 patients received neflamapimod for 12 weeks. Memory function assessed with Wechsler Memory Scale (WMS)–IV, analyzed as Immediate-Recall composite [Logical-Memory (LM) I + Verbal-Paired-Associates (VPA) I + Visual-Reproduction (VR) I; range=0–136 points]; and Delayed-Recall composite [LM II + VPA II + VR II; range=0–92 points]. Study 303:8 patients received neflamapimod for 6 weeks. Memory function assessed with Hopkins Verbal Learning Test–Revised (HVLT-R), Total(Immediate)-Recall (range=0–36 points) and Delayed-Recall (range=0–12 points). In Study 302: the mean WMS Immediate-Recall score at baseline was 48.4(SD=14.6) and increased by 5.7(9.7) points at day 28 (p=0.03; Wilcoxon signed-rank test) and by 10.0(11.7) points at day 84 (p=0.005). The mean WMS Delayed-Recall at baseline was 13.2(10.6) and increased by 5.0(4.5) points at day 28 (p=0.001) and by 8.87(9.4) points at day 84 (p<0.001). All subcomponents of WMS composites showed statistically significant (p<0.05) increases at day 84, except VRII (p=0.053). In study 303: mean HVLT-R Total-Recall score was 19.1(4.3) at baseline and increased by 3.6(5.6) at day 40 (p=0.014; paired t-test). The mean HVLT-R Delayed-Recall was 5.4(1.7) at baseline and increased by 2.1(2.6) at day 40 (p=0.028). Within-subject effect sizes (Cohen's davg) were similar between the two studies and ranged between 0.59 and 0.86. Pharmacokinetics-Pharmacodynamics (PK-PD): linear regression model indicated individual subject plasma drug concentration profiles were significantly correlated to change in combined WMS-Immediate and Delayed-Recall (r2=.70, p=0.001). The low sample size precluded PK-PD analysis in Study 303. Consistency across the studies using different measures, robustness of statistics, effect size and drug concentration-response relationship preliminarily indicates a positive effect of neflamapimod on episodic memory. A 6-month placebo-controlled study is planned to confirm these findings.