Cortical microstructural MRI for detection of early target engagement in AD drug trials: post‐hoc analysis of exploratory outcomes from a Phase 1b safety study for XPro1595 in AD

Abstract

AbstractBackgroundEvidence in support of the amyloid cascade hypothesis for Alzheimer’s disease (AD) suggests that pathological amyloid‐beta (Aβ) is initially associated with synaptic hyperactivity and accumulates in an activity dependent fashion. These findings are corroborated by longitudinal data from cohort studies with Aβ‐PET imaging, wherein Aβ accumulates first in gray matter (GM) cortices associated with the default‐mode network (DMN) before spreading in a semi‐hierarchical pattern to different neocortical regions as the disease progresses (Mattsson et al. 2019). Mapping brain regions most frequently affected by Aβ pathology at different stages of disease may provide a framework for investigation of new drug entities targeting other components of Aβ toxicity using less invasive imaging modalities. The hypothesis here is that indicators of disease specific, tissue‐level treatment effects should be first detectable in brain regions most frequently affected by active pathology.MethodXPro1595 is a selective, brain penetrant neutralizer of soluble TNF with potent anti‐neuroinflammatory effects recently assessed for safety and pharmacological activity in a small, open‐label phase‐1b dose finding study (NCT03943264) in patients with AD. Clinical trial results for patients (n = 9) with analyzable dMRIs who completed 12‐weeks of treatment with XPro1595 [0.3 mg/kg; n = 3) or (1.0 mg/kg; n = 6)] met criteria for post‐hoc analysis. Exploratory outcomes included the novel cortical diffusivity measurement PerpPD+. PerpPD+ represents the diffusion components perpendicular to cortical GM minicolumns, indicates microscopic disruption when increased, and has proven more sensitive than standard volumetrics to changes in indicators of synaptic function, neuroinflammation and neurodegeneration across the AD continuum.ResultMean MMSE scores at baseline were 23.7 (SD = 2.5) and 15.8 (SD = 7.0) in the low‐dose and high‐dose groups, respectively, indicating varying degrees of severity. Week‐12 results showed decreases in bilateral PerPD+ at the whole brain level and in multiple GM cortices documented in the literature with the highest frequency of early (precuneus, isthmus cingulate, posterior cingulate, insula, and lateral and medial orbitofrontal) and late (lingual, pericalcarine, paracentral, precentral, and postcentral) Aβ accumulation (Figures1‐2).ConclusionCortical diffusivity measurements related to minicolumnar microstructure may provide an opportunity for noninvasive detection of early, disease specific, tissue‐level target engagement in clinical trials for investigational new drugs in AD.