Risk of conversion to mild cognitive impairment or dementia among subjects with amyloid and tau pathology: a systematic review and meta-analysis

This systematic review and meta-analysis aimed to quantify the risk of progression to mild cognitive impairment (MCI) or dementia for individuals who are not yet demented, specifically based on their biomarker status for beta-amyloid (A) and phosphorylated tau (T).

The researchers performed a comprehensive search across MEDLINE, EMBASE, and CENTRAL databases for studies using the NIA-AA 2018 framework (identifying pathology via amyloid PET or CSF analysis). Data from 36 cohorts (n=7,793) were pooled using a random-effects model to calculate odds ratios (OR) and hazard ratios (HR) for progression. They specifically compared groups defined by the presence or absence of amyloid (A) and tau (T) proteins, with a focus on cognitively unimpaired (CU) and MCI populations.

The study confirms that amyloid-beta (A+) positivity is a strong predictor of progression, with an OR for conversion of 5.18 in MCI patients and 5.79 in CU subjects. The risk is further refined by tau status: individuals who are both amyloid and tau positive (A+T+) show dramatically higher odds of conversion (OR 11.60 in MCI, 13.46 in CU) compared to those with amyloid only (A+T-). Notably, meta-regression indicated that the predictive strength of amyloid-beta positivity regarding cognitive conversion significantly diminishes as age increases. The researchers found no statistically significant difference in prognostic capability between CSF and PET-based measurements.

These findings reinforce the utility of a biological definition of Alzheimer's disease (the A/T/N framework) for identifying high-risk individuals before clinical symptoms manifest. By distinguishing between A+T+ and A+T- status, clinicians can better stratify risk. The study suggests that future preventive clinical trials should prioritize identifying patients based on these biomarkers, rather than relying solely on cognitive symptoms, to create more homogeneous study populations and improve the assessment of potential treatments.