The amyloid-β42/40 (Aβ42/40) ratio should be included as a cerebrospinal fluid (CSF) biomarker for Alzheimer disease (AD) and may be appropriate for use in the clinical diagnostic setting, according to a team of researchers affiliated with the University Medical Center Goettingen in Goettingen, Germany.1 CSF plays and important role in the differential diagnosis of AD, but lack of standardization compromises use in clinical laboratories, presenting risk of over- and underdiagnosis. The findings suggest that analysis of Aβ40 as well as Aβ42 concentrations may provide greater sensitivity and specificity.
The researchers explained that Aβ40 is the most abundant C-terminal Aβ-variant in CSF and correlates well with the total Aβ concentration; thus, it has the ability to serve as a surrogate-marker for total CSF Aβ. They pointed to studies showing that the Aβ42 to Aβ40 concentration ratio can to improve diagnostic accuracy in patients who appear to have either unusually low or high overall Aβ. To assess whether assessment of the Aβ42/40 ratio is feasible in a clinical diagnostic setting, the researchers compared parallel CSF biomarker measurement results of 2 different certified clinical laboratories. They also confirmed findings in a subset (n=10) of patients using amyloid positron emission tomography/computed tomography.
Parallel aliquots of CSF samples from 114 subjects were sent to 2 independent laboratories for analysis of t-Tau, p-Tau, Aβ42, and Aβ40 and concentrations and Aβ42/40 concentration ratios, with measurements done in accordance with each laboratory’s specific standard operating procedures and cut-off points. Mean age of study subjects was 66.5 years, and 52% of the cohort were men. Biomarker levels and corresponding diagnostic classifications were compared, and the degree of agreement between results of the different laboratories was assessed.
The Aβ42 and Aβ40 measurements performed by the 2 laboratories showed statistically significant correlations, with R2 = 0.5078 for Aβ42 and R2 = 0.4308 for Aβ42/40, respectively (P < .0001 for both parameters). Statistically significant correlations for p-Tau and t-Tau measurements (P < .0001 for both parameters) and CSF Aβ40 (R2 = 0.5739; P < .0001) also were seen. Not surprisingly, though, substantial discrepancies in regard to the Aβ42 concentrations and Aβ42/40 ratios were noted, likely owing to use of different assay kits (IBL vs Fujirebio).
Discrepancies were not seen for Aβ40 or t-Tau and p-Tau measures; however, diagnostic interpretations between laboratories were often discordant for Aβ42 as well as t-Tau and p-Tau, as cut-off points differed between laboratories. Of note, however, is that, when subjects were diagnostically classified via interpretation of Aβ42 levels (normal vs pathological range), a discordance was cited in 32% of cases whereas, when the classification was based on the Aβ42/40 ratio, discordance was reduced by nearly half (17% of cases).
According to the study, these findings lend support to the argument that Aβ42/40 ratio may represent a better neurochemical biomarker of pathologic Aβ deposition than Aβ42 concentration alone. They added that their findings illustrate the importance of improving and standardizing CSF marker interpretation in the clinical diagnostic laboratory setting to reduce false-negative and false-positive AD diagnoses.
1. Vogelgsang J, Wedekind D, Bouter C, et al. Reproducibility of Alzheimer's disease cerebrospinal fluid-biomarker measurements under clinical routine conditions. J Alzheimers Dis. 2018;62:203-212.