Luikku, Antti J.;
Rauramaa, Tuomas;
Nerg, Ossi;
Koivisto, Anne M.;
Junkkari, Antti;
Martiskainen, Henna;
Soininen, Hilkka;
Hiltunen, Mikko;
Leinonen, Ville
Quantified analysis of Aβ plaques and microglia from in vivo cohort of patients with idiopathic normal pressure hydrocephalus
Beschreibung:
<jats:title>Abstract</jats:title><jats:sec><jats:title>Background</jats:title><jats:p>Alzheimer’s pathologic change is frequently observed among patients with idiopathic normal pressure hydrocephalus (iNPH). Amyloid‐beta (Aβ) plaques and pathologic tau deposits <jats:italic>in vivo</jats:italic> are the hallmarks of neuropathologically verified AD, but correlation of Aβ plaque amount to cognitive decline remains controversial. The microglia reaction related with the brain Aβ pathology is under intense research and may modulate the toxicity of Aβ accumulation. Our aim is to characterize Aβ pathology and load in detail and evaluate microglia reaction related with the plaques and their correlation with clinical outcome.</jats:p></jats:sec><jats:sec><jats:title>Method</jats:title><jats:p>51 patients were shunted for iNPH between 9/2010 and end of 2017 with frontal cortical biopsy positive for Aβ. Aiforia create‐machine learning (ML) software was used on whole slide images (WSI) of ionized calcium binding adaptor molecule 1 (Iba1)/Aβ stained biopsies. ML was used to recognize Aβ area and Iba1 stained microglia somas. Clinical Dementia Rating global score (CDR‐GS) and Alzheimer’s clinical syndrome (ACS) were evaluated retrospectively at the end of the median follow‐up of 7.6 years (range 4.1‐ 14) during which 40 patients died.</jats:p></jats:sec><jats:sec><jats:title>Result</jats:title><jats:p>During the follow‐up 19 patients were diagnosed with Alzheimer’s clinical syndrome. Trained image analysis yielded recognition precision for tissue, Aβ and microglia somas with 0.99, 0.98, and 0.96, respectively. Percentage of Aβ area, microglia soma density and reactive microglia within 25 µm of Aβ plaques were acquired. In univariate analysis, preoperative MMSE was significant predictor of CDR‐GS (p < .001) and ACS (p < .01). In multivariate analysis, high preoperative MMSE predicted lower CDR‐GS on follow‐up (OR 0.74 [0.63‐0.85]) and lesser development of ACS (OR 0.86 [0.75‐0.99]). Among demented subjects, there was a trend of higher Aβ area density (median 2.4 % for CDR‐GS 0.5 and 10.1 % for CDR‐GS 3), and reactive microglia (median 84 for CDR‐GS 0.5 and 178 for CDR‐GS 3).</jats:p></jats:sec><jats:sec><jats:title>Conclusion</jats:title><jats:p>Higher Aβ load and reactive microglia might predict worse cognitive outcome and dementia. As additional patients will be analyzed during spring 2021, we will expect this trend to turn significant. Good baseline cognitive performance seems to protect from later development of Alzheimer’s clinical syndrome and dementia among iNPH patients.</jats:p></jats:sec>