Recent findings from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) and the A4 studies reveal critical insights into the use of blood-based biomarkers, specifically plasma p-tau 181 and p-tau 217, as proxies for amyloid pathology in Alzheimer’s disease (AD). The research highlights that while these biomarkers offer advantages such as low cost and minimal invasiveness, they present significant challenges in terms of statistical power and bias when compared to traditional neuroimaging measures like amyloid positron emission tomography (amyloid-PET) Centiloids.

The analysis indicates that substituting amyloid-PET Centiloids with blood-based p-tau measures can lead to a substantial increase in required sample sizes, ranging from 1.5 to 6.5 times larger to achieve the same statistical power (80%). Furthermore, employing blood-based biomarkers as exposure variables introduces regression dilution bias, which can attenuate the strength of observed associations. This highlights the need for careful consideration of the methodological implications when designing studies that incorporate these biomarkers.

The findings underscore a pivotal shift in how researchers should approach the design and interpretation of studies in AD pathology. The increased sample size requirements and potential biases associated with blood-based biomarkers necessitate a reevaluation of current research paradigms, particularly in large population-based studies. As the field moves toward integrating these biomarkers into clinical practice, understanding their limitations will be crucial for ensuring robust and reliable outcomes in AD research and therapeutic development.

Source: academic.oup.com