作者
Nicolai Franzmeier,Marc Suárez‐Calvet,Lukas Frontzkowski,Estrella Morenas‐Rodríguez,Gernot Kleinberger,Leslie M. Shaw,John Q. Trojanowski,Christian Haass,Michael Ewers
摘要
Abstract Background Loss of function of TREM2, a key receptor expressed on microglia, is associated with increased Alzheimer’s disease (AD) risk. In biomarker‐defined AD patients, we found previously that higher cerebrospinal‐fluid (CSF) levels of soluble TREM2 relative to the level of p‐tau (sTREM2/p‐tau ratio) are associated with attenuated clinical AD progression. This suggests that higher sTREM2/p‐tau, indicative of more activated microglia at a given level of pathology, may be beneficial in AD. Here, we asked the more general question whether higher sTREM2/p‐tau ratio levels attenuate the general risk to develop AD‐related cognitive decline and neurodegeneration in a large sample of cognitively normal to AD dementia subjects. To assess AD risk, we used a pre‐established polygenic hazard score (PHS) including 31 SNPs that are associated with age of AD onset (e.g. Tan et al., Brain, 2019). We determined whether higher sTREM2/p‐tau levels attenuate the effect of polygenic AD risk (i.e. PHS) on future cognitive decline and neurodegeneration. Methods We included 648 elderly cognitively normal to dementia subjects from ADNI with available CSF‐biomarker data (i.e. sTREM2, Aβ 1‐42 and p‐tau), longitudinal cognitive assessments and structural MRI (∼4yrs follow‐up). The PHS was determined on GWAS data. Using linear regression, we tested the association between PHS, CSF‐assessed AD biomarkers or sTREM2, controlling for age, gender, education and diagnosis. Second, we determined whether higher sTREM2/p‐tau levels moderated the effect of PHS on annual change rates in global cognition (i.e. ADAS13), memory (i.e. ADNI‐MEM) and MRI‐assessed hippocampal volume, controlling for age, gender, education and diagnosis and baseline cognition. Results Higher polygenic AD risk, i.e. PHS, was associated with lower CSF Aβ 1‐42 (β=‐0.354,p<0.001) and higher p‐tau levels (β=0.328,p<0.001), but not with sTREM2 (β=0.007,p=0.864; Figure 1). We found significant sTREM2/p‐tau x PHS interactions, such that individuals with higher sTREM2/p‐tau ratios had a lower effect of PHS on global cognitive (β=‐0.210,p=0.002) and memory decline (β =0.258,p=0.002) as well as hippocampal atrophy rates (β=0.262,p=0.002; Figure 2). Conclusion A higher sTREM2/p‐tau ratio is protective against the AD‐related polygenic risk for faster cognitive decline and hippocampus atrophy.