Blood-based biomarkers for Alzheimer disease: mapping the road to the clinic

Abstract

Biomarker discovery and development for clinical research, diagnostics and therapy monitoring in clinical trials have advanced rapidly in key areas of medicine - most notably, oncology and cardiovascular diseases - allowing rapid early detection and supporting the evolution of biomarker-guided, precision-medicine-based targeted therapies. In Alzheimer disease (AD), breakthroughs in biomarker identification and validation include cerebrospinal fluid and PET markers of amyloid-β and tau proteins, which are highly accurate in detecting the presence of AD-associated pathophysiological and neuropathological changes. However, the high cost, insufficient accessibility and/or invasiveness of these assays limit their use as viable first-line tools for detecting patterns of pathophysiology. Therefore, a multistage, tiered approach is needed, prioritizing development of an initial screen to exclude from these tests the high numbers of people with cognitive deficits who do not demonstrate evidence of underlying AD pathophysiology. This Review summarizes the efforts of an international working group that aimed to survey the current landscape of blood-based AD biomarkers and outlines operational steps for an effective academic-industry co-development pathway from identification and assay development to validation for clinical use.

Figures

Figure 1. Challenges in developing a blood-based biomarker of CNS disease.

There are several considerations when finding a biomarker for a CNS disease, particularly if the biomarker is to be blood-based. Blood–brain barrier: biomarker must be able to cross the blood–brain barrier to allow detection (A). Diurnal differences in CSF and blood: diurnal differences in protein concentrations exist in both CSF and blood. Biomarker levels may not peak at the same time in CSF as in blood, and a biomarker assay will either require sampling at the peak concentration, or sufficient sensitivity to detect the biomarker throughout the day (B). Active/passive transport: biomarkers may originate in either CSF or blood. Those originating in CSF may enter the blood via active or passive transport, and understanding the exact nature of the derivation from CSF will be essential to develop an assay (C). Concentration differences: biomarker levels are not always similar in CSF and blood. For example, Aβ concentrations are 10-fold lower in plasma than in CSF (35 pg/ml vs. 350 pg/ml) (D). Abbreviations: Aβ=amyloid beta; CNS=central nervous system; CSF=cerebrospinal fluid.

Figure 2. Candidate blood-based biomarkers identified in the landscape analysis.

Technology (A) and platform (B) categories for the 196 high-quality studies identified in the landscape analysis.

Figure 3. Schematic representation of promising blood-based biomarker candidates

Four promising blood-based biomarker candidates are represented in this schematic: Aβ: Aβ peptides, in particular Aβ1–42, are implicated in AD pathogenesis; however, it is the ratio of Aβ1–42/Aβ1–40 that appears to be the most promising Aβ-related biomarker in the blood. BACE1: The first step in the generation of Aβ peptides is the cleavage of APP by the β-secretase, BACE1. Measurement of BACE1 activity in the blood may be useful for predicting progression from MCI to AD dementia. Tau: Phosphorylated tau protein is a major component of intraneuronal neurofibrillary tangles, which are often present in AD. The abnormal phosphorylation of tau is thought to be driven by Aβ peptides. Tau levels in the blood may be useful as a predictor of future cognitive decline. NF-L: NF-L is an axonal protein, released into the brain interstitial fluid following neuronal/axonal injury. NF-L levels in the blood are elevated in AD and other neurodegenerative diseases, therefore, blood-based NF-L could be useful as a biomarker of neurodegeneration. Abbreviations: Aβ=amyloid beta; AD=Alzheimer’s disease; APP=amyloid precursor protein; BACE1=β-site amyloid precursor protein cleaving enzyme 1; MCI=mild cognitive impairment; NF-L=neurofilament light.

Figure 4. Idealized validation process for blood-based biomarkers.

Abbreviations: AD=Alzheimer’s disease; CSF=cerebrospinal fluid; MCI=mild cognitive impairment.

Figure 5. Validation status and patient numbers of the 19 prioritized biomarkers.

Few of the current blood-based biomarker assays in development have undergone validation in an external cohort. Of the 26% of assays that had some degree of external validation, only 2 (10·5%) were in large cohorts such as ADNI or AIBL (A). Studies are often underpowered, with small patient numbers (B).

Figure 6. Potential collaboration points between academia and industry.

Academic and industrial approaches to biomarker development are inherently different, but could be extremely useful if combined. Close collaboration between industry and academia would allow for sharing of expertise in product testing, access to cohorts and clinical data, as well as allow for sharing ideas and theories with regards to clinical endpoints and context. By merging the two approaches, a method where the context of use is the primary focus throughout the process can be established. This would allow for synergistic development of a new biomarker between academics and industrial partners, sharing a wealth of experience.