Amyloid and tau pathology in the biomarker-detectable range occurred in 19.6% of decedentsStudy links specific brain protein patterns to Alzheimer disease progression

This cross-sectional clinicopathologic study utilized data from 2,711 decedents aged 18 years or older to examine the distribution of amyloid and tau pathology in relation to clinical dementia. The study was conducted using samples from the Biobank for Aging Studies of the University of Sao Paulo. The research focused on the prevalence of pathology within the biomarker-detectable range, specifically defined as Thal phase 3 to 5 and Braak stage IV to VI, and how these markers correlated with clinical outcomes and coexisting neuropathologies.

The investigation evaluated several key pathological markers: Thal amyloid phase, Braak neurofibrillary tangle stage, CERAD neuritic plaque score, and various coexisting neuropathologies. The primary objective was to determine the frequency of combined amyloid-tau pathology in the biomarker-detectable range and its distribution across different age groups and sex. The study also looked at the prevalence of clinical impairment (CDR > 0) across different pathological stages and cross-classified categories.

Regarding the primary outcomes, the study found that the frequency of combined amyloid-tau pathology (Thal 3-5 and Braak IV-VI) was 19.6% (531/2711). When analyzed by age, the frequency of this combined pathology showed a clear increase with age: 0% (23-50 years), 5.5% (61-70), 14.5% (71-80), 36.5% (81-90), and 45.9% (91-111). Furthermore, the prevalence of CDR > 0 showed a monotonic increase as the Braak stage progressed, rising from 16.3% at stage 0 to 86.6% at stage VI. When cross-classified, the prevalence of CDR > 0 was 15.3% for those with neither pathology, 58.5% for those with coexisting pathology only, 63.7% for those with combined amyloid-tau only, and 85.9% for those with both.

Secondary outcomes included multivariable models to assess the contribution of specific factors to the Clinical Dementia Rating Sum of Boxes (CDR-SB). The multivariable model for the contribution of coexisting pathology to CDR-SB yielded a partial R2 of 0.088 (95% CI, 0.064-0.114). The model for Braak stage variance showed a Lindeman, Merenda, and Gold contribution of 0.115. These metrics helped quantify the impact of specific pathological features on the progression of cognitive impairment.

Safety and tolerability data were not reported, as the study utilized post-mortem tissue samples rather than live clinical interventions. The study results indicate that amyloid and tau pathology in the biomarker-detectable range was uncommon before age 70. Additionally, the study noted that these pathologies were discordant in 17.6% of decedents. A significant finding was that coexisting brain disease accompanied cognitive impairment as often as advanced Alzheimer disease pathology.

These results contribute to the understanding of the clinicopathologic landscape of Alzheimer disease. While the study provides high-certainty data from tissue-based autopsy samples, it is important to note that the findings are observational and do not establish causality. The primary limitation is that the data are derived from deceased individuals, which may not perfectly mirror the progression of the disease in living patients. Clinical implications suggest that while amyloid and tau are central to the disease, coexisting neuropathologies play a significant role in the clinical presentation of cognitive impairment. Questions remain regarding how these specific pathology distributions translate to early intervention windows in living patients with varying degrees of coexisting pathology.