Future perspective and clinical applicability of the combined use of plasma phosphorylated tau 181 and neurofilament light chain in Subjective Cognitive Decline and Mild Cognitive Impairment

Abstract

We aimed to assess diagnostic accuracy of plasma p-tau181 and NfL separately and in combination in discriminating Subjective Cognitive Decline (SCD) and Mild Cognitive Impairment (MCI) patients carrying Alzheimer's Disease (AD) pathology from non-carriers; to propose a flowchart for the interpretation of the results of plasma p-tau181 and NfL. We included 43 SCD, 41 MCI and 21 AD-demented (AD-d) patients, who underwent plasma p-tau181 and NfL analysis. Twenty-eight SCD, 41 MCI and 21 AD-d patients underwent CSF biomarkers analysis (Aβ1-42, Aβ1-42/1-40, p-tau, t-tau) and were classified as carriers of AD pathology (AP+) it they were A+/T+ , or non-carriers (AP-) when they were A-, A+/T-/N-, or A+/T-/N+ according to the A/T(N) system. Plasma p-tau181 and NfL separately showed a good accuracy (AUC = 0.88), while the combined model (NfL + p-tau181) showed an excellent accuracy (AUC = 0.92) in discriminating AP+ from AP- patients. Plasma p-tau181 and NfL results were moderately concordant (Coehn's k = 0.50, p < 0.001). Based on a logistic regression model, we estimated the risk of AD pathology considering the two biomarkers: 10.91% if both p-tau181 and NfL were negative; 41.10 and 76.49% if only one biomarker was positive (respectively p-tau18 and NfL); 94.88% if both p-tau181 and NfL were positive. Considering the moderate concordance and the risk of presenting an underlying AD pathology according to the positivity of plasma p-tau181 and NfL, we proposed a flow chart to guide the combined use of plasma p-tau181 and NfL and the interpretation of biomarker results to detect AD pathology.

Keywords: Alzheimer’s Disease; Mild Cognitive Impairment; NfL; Plasma biomarkers; Subjective Cognitive Decline; p-tau181.

Figure 1

Plasma biomarkers levels across diagnosis/ATN groups. (a) Log p-tau181 levels across diagnosis/ATN groups. Values quoted in the y-axis indicate Log p-tau181 levels. Horizontal bars indicate significant differences between groups. SCD AP− vs SCD AP + p = 0.002, Cohen’s d = 1.629; SCD AP− vs MCI AP + p < 0.001, Cohen’s d = 2.332; SCD AP− vs AD-d p < 0.001, Cohen’s d = 2.395; SCD AP + vs MCI AP− p = 0.048, Cohen’s d = 1.144; MCI AP− vs MCI AP + p < 0.001, Cohen’s d = 1.703; MCI AP− vs AD-d p < 0.001, Cohen’s d = 1.772. (b) Log NfL levels across ATN biomarkers’ profile groups. Values quoted in the y-axis indicate Log NfL levels. Horizontal bars indicate significant differences between groups. SCD AP− vs SCD AP + p = 0.003, Cohen’s d = 1.700; SCD AP− vs MCI AP + p < 0.001, Cohen’s d = 1.516; SCD AP− vs AD-d p = 0.001, Cohen’s d = 1.167; SCD AP + vs MCI AP− p = 0.014, Cohen’s d = 1.257; MCI AP− vs MCI AP + p = 0.001, Cohen’s d = 1.131; MCI AP− vs AD-d p = 0.004, Cohen’s d = 1.223. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 2

ROC curves for accuracy of plasma p-tau181, NfL and the combined model (NfL + p-tau181) in distinguishing A+ from A−, T+ from T−, N+ from N−, and AP+ from AP− patients in SCD and MCI, considering both separately and together.

Figure 3

Flow chart for the potential use and interpretation of plasma biomarkers in clinical setting for the early detection of Alzheimer’s Disease.