2026 ECNP Congress

PS02-2046

Stringent cerebrospinal fluid amyloid stratification reveals heterogeneous cerebrospinal fluid and limited plasma phosphorylated tau separation in preclinical Alzheimer’s disease

Daniel Benjamin Dumas — S. Prins, A. Gorbenko, T. Karikari, M. de Kam, P.H.C. Kremer, C.E. Teunissen, G.J. Groeneveld · Centre for Human Drug Research, Neuroscience, Leiden, Netherlands

2026 ECNP Congress

Background

Early identification of Alzheimer’s disease (AD) pathology in cognitively unimpaired (CU) individuals is critical for the development of disease-modifying therapies [1]. Cerebrospinal fluid (CSF) amyloid-β1–42 (Aβ1–42) is widely used to define amyloid pathology but does not fully capture downstream tau-related processes. Phosphorylated tau (p-tau) isoforms have emerged as sensitive biomarkers of AD, with p-tau217, p-tau231, and p-tau181 reflecting complementary aspects of early disease biology [2]. However, most studies have evaluated these markers across the AD continuum rather than in strictly CU populations. In addition, dichotomous CSF Aβ1–42 cut-offs are commonly used for classification and, in some early-phase interventional studies, for trial enrichment, although amyloid accumulation itself is inherently continuous. It remains unclear how applying stringent cut-offs that exclude intermediate values influences observed biomarker differences in preclinical populations.

Objectives

To assess whether p-tau isoforms in CSF and plasma differ between cognitively unimpaired older adults with and without CSF amyloid‑β (Aβ) pathology defined using stringent cut-offs.

Methods

In this cross-sectional study, 100 CU adults aged ≥65 years were classified as amyloid-positive (Aβ+; n = 50; CSF Aβ1–42 ≤1000 pg/mL) or amyloid-negative (Aβ−; n = 50; CSF Aβ1–42 ≥1700 pg/mL). These participants were originally recruited as healthy volunteers for a phase 0 biomarker study. Concentrations of CSF p-tau181, p-tau217, and p-tau231, and plasma p-tau181 and p-tau231 were measured using in-house Simoa HD-X assays (University of Gothenburg). Cognitive performance was assessed with a standardized test battery and the Mini-Mental State Examination. Group differences were evaluated with a mixed-effects model on log-transformed data adjusted for age, sex, and ApoE ε4 status. Effect sizes (Cliff’s δ on the raw scale and Cohen’s d on log-transformed data) and Spearman correlations were calculated. Exploratory analyses included distributional visualization and trichotomization of the amyloid-positive group based on its median CSF Aβ1–42 concentration to assess gradients within the amyloid continuum.

Results

No significant mean differences were observed for any p-tau isoform (CSF or plasma) between Aβ+ and Aβ− groups. Estimated group differences ranged from 1–27% across biomarkers, with wide confidence intervals. However, CSF p‑tau isoforms showed broader and right-shifted distributions in the Aβ+ group, with small standardized group separation (Cliff’s δ range 0.26–0.40), whereas plasma p-tau measures showed minimal separation. In exploratory analyses, trichotomization of the amyloid continuum suggested gradual distributional shifts in CSF p-tau isoforms, most prominently for p-tau217 and p-tau231, although substantial overlap remained. Across the cohort, p-tau isoforms correlated positively with age (ρ up to 0.26) and strongly within matrices but showed no consistent associations with CSF Aβ1–42 or cognitive outcomes (all |ρ| ≤ 0.19).

Conclusions

In cognitively unimpaired adults ≥ 65 years selected using stringent CSF Aβ1–42 cut-offs chosen to enhance contrast between Aβ+ and Aβ−groups, adjusted mean differences in CSF and plasma p-tau isoforms were not statistically significant despite distributional differences in CSF. These findings suggest that stringent amyloid stratification may attenuate detectable cross-sectional group differences, while p-tau variability reflects heterogeneous early disease processes and age-related effects. These results support interpreting p-tau biomarkers within a continuous and multimarker framework, particularly in strictly defined preclinical populations.

References

  1. Sperling, R.A., Aisen, P.S., Beckett, L.A., Bennett, D.A., Craft, S., Fagan, A.M., Iwatsubo, T., Jack, C.R., Kaye, J., Montine, T.J., Park, D.C., Reiman, E.M., Rowe, C.C., Siemers, E., Stern, Y., Yaffe, K., Carrillo, M.C., Thies, B., Morrison-Bogorad, M., Wagster, M.V., Phelps, C.H., 2011. Toward defining the preclinical stages of Alzheimer’s disease: Recommendations from the National Institute on Aging–Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement. 7 (3), 280–292.
  2. Karikari, T.K., Pascoal, T.A., Ashton, N.J., Janelidze, S., Benedet, A.L., Rodriguez, J.L., Chamoun, M., Savard, M., Kang, M.S., Therriault, J., Schöll, M., Massarweh, G., Soucy, J.P., Höglund, K., Brinkmalm, G., Gobom, J., Zetterberg, H., Blennow, K., Rosa-Neto, P., 2020. Blood phosphorylated tau 181 as a biomarker for Alzheimer’s disease: a diagnostic performance and prediction modelling study using data from four prospective cohorts. Lancet Neurol. 19 (5), 422–433.