Blood-based Biomarkers As Predictors Of Longitudinal Cognitive Decline And Hippocampal Atrophy -the Vallecas Study:

Author Block: Michelle Estefanía Guamán Chulunchana¹, Linda Zhang³, Francisco Javier López-González⁴, Sonia Wagner Reguero³, Minerva Martinez Castillo³, Ana Belén Pastor¹, Nekane Moreno¹, Andrea Ruiz-Calvo¹, Cristina Sánchez-Martín², Eva Alfayate², Marta Molero-Cartón², Elizabeth L. Valeriano-Lorenzo⁵, Belén Frades-Payo⁵, Meritxell Valentí⁵, María Ascensión Zea-Sevilla, PhD⁵, Alexis Moscoso Rial⁶, Alfredo Ramirez⁷, Pamela Martino-Adami⁷, Steffi G. Riedel-Heller⁸, Martin Scherer⁹, Andreas Jeromin¹⁰, Miguel Calero¹¹, Michel J. Grothe¹², Anja Schneider¹³, Teodoro Del Ser⁵, Aitana Sogorb-Esteve¹, Jesús Silva-Rodríguez², Pascual Sánchez-Juan¹⁴.
¹Department of Biomarkers, Biochemistry and Molecular Genetics, CIEN Foundation, Queen Sofia Foundation Alzheimer Center, ISCIII, Madrid, Spain, ²Neuroimaging department, CIEN Foundation, Queen Sofia Foundation Alzheimer Center, ISCIII, Madrid, Spain, ³CIEN Foundation, Queen Sofia Foundation Alzheimer Center, ISCIII, Madrid, Spain, ⁴Bioinformatics department, CIEN Foundation, Queen Sofia Foundation Alzheimer Center, ISCIII, Madrid, Spain, ⁵Clinical evaluation department, CIEN Foundation, Queen Sofia Foundation Alzheimer Center, ISCIII, Madrid, Spain, ⁶Wallenberg Centre for Molecular and Translational Medicine, University of Gothenburg, Gothenburg, Sweden, ⁷University Hospital Cologne, Cologne, Germany, ⁸Institute of Social Medicine, Occupational Health and Public Health, University of Leipzig, Leipzig, Germany, ⁹Department of Primary Medical Care, Center for Psychosocial Medicine, University Medical Center Hamburg-Eppendorf, Hamburg, Germany, ¹⁰Tau Biosciences, Inc, Carlsbad, CA, USA, ¹¹Chronic Disease Programme (UFIEC), Instituto de Salud Carlos III, Madrid, Spain, ¹²Neuroimaging department, CIEN Foundation, Queen Sofia Foundation Alzheimer Center, ISCIII,, Madrid, Spain, ¹³DZNE, Germany, ¹⁴Scientific Director, CIEN Foundation, Queen Sofia Foundation Alzheimer Center, ISCIII, Madrid, Spain.

Abstract:
Background: Early identification of cognitively unimpaired individuals at high risk of progressing to Alzheimer’s Disease is a priority for prevention. While plasma p-tau217 is a leading biomarker, longitudinal comparisons between mass spectrometry metrics for predicting long-term clinical and biological outcomes remain scarce.
Methods: We analyzed Cognitively unimpaired older adults from the Vallecas Cohort with 10 years of follow-up were analyzed. Baseline plasma was analyzed using the PrecivityAD blood test (C2N) for %p-tau217, p-tau217, and Aβ42/40 ratio. We evaluated biomarker efficiency, particularly the APS2 score, predicting: i) cognitive decline and hippocampal atrophy, ii) clinical conversion to AD, and iii) accuracy using time-dependent ROC (tAUC) and predictive values (PPV/NPV) at 5 and 10 years.
Results: Linear mixed models revealed that individuals in the top decile (D1) of %p-tau217 and APS2 exhibited significantly faster mPACC decline and accelerated hippocampal atrophy vs. Q1-Q2 (p<0.001).

In Cox models, p-tau217 MS (HR=1.88, p<0.001) and APS2 (HR=1.95, p<0.001) were the strongest conversion predictors. While Aβ42/40 ratio showed a modest association (HR=1.25, p=0.0216), in fully adjusted models (including MRI and NfL), %p-tau217 had the strongest association (HR=1.69, p<0.001), followed by APS2 (HR=1.74, p<0.001); Aβ42/40 added no significance (p=0.08). Time-dependent ROC analysis showed %p-tau217 (MS) and APS2 achieved the highest discriminatory power at 5 years (tAUC=0.79 and 0.75), outperforming clinical-only (tAUC=0.69) and Aβ42/40 (tAUC=0.66). At 5 years, both %p-tau217 and APS2 demonstrated similar NPV (94.3% and 94%) for ruling out progression, whereas Aβ42/40 had 91% NPV. Being in the P90 for %p-tau217 yielded 66% PPV at 10 years for clinical conversion, while Aβ42/40 exhibited the lowest PPV (24.4%).
Conclusion: Plasma %p-tau217 and APS2 score are the most robust predictors of hippocampal atrophy and clinical progression to AD. The high NPV supports their use as reliable rule-out criteria, while the top decile identifies high-risk individuals with a >50% conversion rate within 10 years.