Clinically relevant factors for blood‐based Alzheimer biomarker interpretation

AbstractBackgroundBlood‐based biomarkers for Alzheimer disease (AD)‐related processes are moving towards large‐scale clinical implementation at a very rapid pace. Maintaining diagnostic performance when moving from group‐level data in well‐curated clinical cohorts to individual diagnostics in real‐world clinical settings can be challenging. This presentation gives an updated account of clinically relevant factors that may influence AD blood biomarker concentrations in individual patients.MethodsBlood‐based AD biomarker data (amyloid β 42/40 ratio [Aβ42/Aβ40], phosphorylated tau forms [P‐tau], neurofilament light [NfL], and glial fibrillary acidic protein [GFAP]) from human experimental and observational cohort studies were summarized and examined in relation to demographic data, including age, sex, race, ethnicity, body mass index (BMI), as well as kidney and liver function tests, other co‐morbidities, drugs, and food intake.ResultsIn most cohorts, age, creatinine, and BMI were associated with NfL, GFAP, and to a lesser extent p‐tau, but their diagnostic performances were not severely influenced. In general, there were no or only minor effects of common co‐morbidities and medications on AD blood biomarker concentrations. However, individual drugs may have strong effects. For example, a drug affecting microglial activity (minocycline), increased plasma NfL concentration several‐fold, and treatment with a neprilysin inhibitor against heart failure decreased plasma Aβ42/Aβ40 (determined by immunoassay) by ∼30%, i.e., with an effect size greater than what is caused amyloid plaque pathology. Surprisingly, intake of a standardized meal reduced plasma NfL, GFAP, and P‐tau (181 and 231) concentrations by 10‐20% after two hours, which is an effect similar in size to what has been seen in anti‐Aβ antibody trials. Across cohorts and biomarkers, plasma Aβ42/Aβ40 appeared to be the most resistant to these types of confounding factors.ConclusionMoving from group level interpretation in research to individual diagnostics in clinical practice requires special attention to unusual biomarker results and improved knowledge of clinically important confounders. Some of this knowledge will be gained in clinical practice by thoughtful biomarker interpretation and knowledge exchange between clinical and laboratory specialists. Some confounding factors may be assay‐specific. Continued intense collaboration between clinics and laboratories, as well as method comparison and refinement studies, will be highly important.