XPro1595, a selective soluble TNF Neutralizer, in Early Alzheimer's Disease with inflammation (ADi): Results from the Phase 2 MINDFuL Trial.

Background: Alzheimer's disease (AD) pathophysiology involves chronic neuroinflammation driven by tumor necrosis factor (TNF). XPro1595 selectively neutralizes pathological soluble TNF while preserving neuroprotective transmembrane TNF signaling, offering a mechanistically distinct approach from current amyloid-targeting therapies. Methods: The MINDFuL trial was a Phase 2, multicenter, randomized, double-blind, placebo-controlled study conducted across 35 centers in eight countries. Participants aged 50-85 years with Early AD (mild cognitive impairment or mild dementia) and elevated inflammatory biomarkers received weekly subcutaneous injections of XPro1595 (1.0 mg/kg) or placebo for 24 weeks. The primary endpoint was the Early Mild Alzheimer's Cognitive Composite (EMACC). A prospectively defined biomarker-enriched population included amyloid-positive participants with a high inflammatory burden (ADi), defined by the presence of ≥2 inflammatory markers. Results: Of 206 randomized participants, 200 comprised the modified intent-to-treat (mITT) population and 100 comprised the ADi population. In the mITT analysis, no significant difference was observed in the EMACC (p=0.672). However, the ADi population showed promising signals across multiple clinical endpoints favoring XPro1595, including the EMACC (Cohen's d = 0.27), Neuropsychiatric Inventory (NPI, d = 0.23), particularly the subfactor agitation/hyperactivity (d = 0.37), and a patient-reported outcome: Goal Attainment Scale (GAS, d = 0.18). These results were consistent with biomarker changes that favored XPro1595, as plasma pTau217 (d = -0.18) and GFAP (d = -0.19) were attenuated in the ADi population. In addition, many effect sizes were larger when assessing participants with higher drug exposure. Notably, no amyloid-related imaging abnormalities (ARIA) were observed, even though ~70% of the participants were APOE ε4 carriers, and one-third had baseline microbleeds. Conclusions: While XPro1595 did not meet the primary endpoint in the overall population, participants with biologically-defined AD and high inflammatory burden demonstrated consistent trends toward cognitive benefit and neuropsychiatric improvement. The complete absence of ARIA distinguishes XPro1595 from amyloid-targeting therapies and supports its potential for use in combination strategies and potentially for those at a higher risk of ARIA. These findings support selective TNF inhibition as a promising therapeutic approach for persons with AD and elevated inflammatory burden.