4,068 posters, 219 videos, 101 topics, 69 sessions, 14,365 authors, 6,128 institutions

Alzheimer's Association International Conference 2026

July 12-15, 2026 | London, United Kingdom, and Online

13376

Subjective Cognitive Decline Improves Prediction Of Incident MCI Beyond AD Blood Biomarkers And Cognitive Measures

View Poster Contact Presenter

Poster Presenter

Thomas Claessen

Authors

Thomas Claessen, Jeremiah Aakre, Maria Vassilaki, Veronique Marcotte, Clifford Jack, Alicia Algeciras-Schimnich, Ronald Petersen, Argonde van Harten

Affiliations

Department of Neurology, Mayo Clinic, USA, Neurochemistry Laboratory, Department of Clinical Chemistry, Amsterdam UMC, the Netherlands, Alzheimer Center Amsterdam, Department of Neurology, Amsterdam UMC, The Netherlands, Department of Quantitative Health Sciences, Mayo Clinic, USA, Department of Radiology, Mayo Clinic, USA, Department of Laboratory Medicine and Pathology, Mayo Clinic, USA, Department of Neurology, Mayo Clinic, USA

Part of Topic

Biomarkers -> Biomarkers (non-neuroimaging) -> Prognostic utility

Background

•Alzheimer’s disease (AD) blood biomarkers and cognitive measures are associated with clinical disease progression, but do not capture subjective cognitive decline (SCD).

•Questionnaires that capture SCD are scalable and low-cost and may provide complementary prognostic information for incident mild cognitive impairment (MCI).

Objective

To determine whether SCD, cognitive screening, and AD blood biomarkers incrementally improve prediction of incident MCI among cognitively unimpaired older adults.

Methods

Inclusion: Cognitively unimpaired participants from the MCSA and the combined A4 placebo/LEARN cohorts (Table 1).

SCD: Participant- and informant-reported ECog-12 in MCSA and participant- and informant-reported CFI in A4/LEARN.

Cognitive screening: MMSE.

Biomarkers: Plasma p-tau217 and GFAP.

Analyses: Cox models evaluated associations with incident MCI. Incremental prediction was assessed using nested models (Demographics → SCD → Cognitive measures → Biomarkers).

Discrimination was quantified with Uno’s C-statistic.

Results

MCSA

•Participant-reported SCD (ECog-12 and concern) was associated with incident MCI (Figure 1).

•Informant-reported SCD (iECog) was also associated with incident MCI in the GFAP sample (Figure 1).

•Cognitive screening provided the largest incremental gain in discrimination, increasing C-statistics from 0.76–0.77 to 0.79–0.82.

•Adding plasma p-tau217 or GFAP only provided small additional gains (ΔC 0.00–0.01).

A4/LEARN

•Participant- (CFI) and informant-reported SCD (CFI-SP) were associated with incident MCI (Figure 2).

•Adding CFI improved model discrimination from 0.65 to 0.70.

•Plasma p-tau217 was strongly associated with incident MCI and improved discrimination: ΔC 0.06, p<0.0001.

Conclusions

•SCD measures were consistently associated with incident MCI and provided modest incremental predictive value.

•These findings support risk-stratification approaches that begin with scalable SCD measures and selectively add cognitive screening and biomarkers by setting.

Comments

Email is required but is not shown publicly.

Sort comments

Loading comments...