The misfolding of the amyloid beta protein is the strongest blood-based predictor for Alzheimer’s disease in the asymptomatic stage, indicating the risk of the disease years before it is clinically diagnosed and much earlier than the widely used plasma biomarker P-tau 217. This is the conclusion reached by a study led by Professor Klaus Gerwert (PRODI, Ruhr University Bochum, Germany). The research team analyzed blood samples from 17 years of 779 subjects in the population-based ESTHER cohort of the German Cancer Research Center (DKFZ) to compare the effectiveness of leading blood biomarkers throughout the entire course of Alzheimer’s disease. The study was published in the journal EMBO Molecular Medicine on July 16, 2026.

The study compares how well common blood-based biomarkers can predict eventual Alzheimer’s dementia in persons without clinical symptoms. To do so, the researchers were able to use blood samples from the ESTHER study, a major long-term study on the health of elderly persons at the DKFZ.

The analysis shows that different biomarkers predominate at different stages of the disease. “P-tau 217 demonstrated outstanding diagnostic effectiveness once the Alzheimer’s disease reached the clinical stage,” says Gerwert. Its accuracy was comparable to that of established biomarkers like the Aβ42/40 ratio in cerebrospinal fluid (CSF) and amyloid PET imaging. However, the biomarker’s predictive power was only moderate in the preclinical, asymptomatic stage (AUC = 0.67). For comparison: Values of about 0.8 or higher are considered highly significant.

In contrast, the biomarker for amyloid beta protein misfolding was predictive even in cognitively normal individuals and achieved an AUC of 0.79. In combination with demographic, genetic, and other blood biomarkers, the resulting biomarker panel based on protein misfolding achieved a very high AUC of 0.87, thereby facilitating highly accurate prediction of an eventual Alzheimer’s diagnosis long before clinical symptoms appeared. “Prevention and anti-amyloid therapies could noticeably slow the development of symptoms or perhaps even largely prevent them, especially in this early stage,” says Gerwert.

“Our study shows that the protein misfolding is the earliest measurable, blood-based marker of Alzheimer’s disease,” says Gerwert. “P-tau 217 is an outstanding biomarker once the disease has clinically manifested. When it comes to identifying persons with an increased risk of disease many years before the initial symptoms occur, however, the biomarker for protein misfolding provides the most significant predictive information.”

The results come at a pivotal time following the approval of disease-modifying anti-amyloid therapies. These treatments can slow the disease’s progression; however, in accordance with approval criteria, they should be initiated at an early stage – specifically, during mild cognitive impairment (MCI) or mild Alzheimer’s dementia. Furthermore, they can lead to serious side effects, including amyloid-related imaging abnormalities (ARIA), which involve swelling and microbleeds in the brain. Because of the necessary diagnostic requirements and possible risks, the treatment is currently an option for only a subset of patients.

“The possibility of treating Alzheimer’s in its earliest molecular stage is thus becoming one of the most important challenges in preventing the disease,” says Gerwert. The authors conclude that protein misfolding should form the foundation for blood-based screening strategies in the future. This would make it possible to identify persons at risk much earlier, more specifically select patients for preventive treatments and clinical studies, and provide more cost-effective and population-wide screening programs.