Scientists identify blood sugar molecule to predict Alzheimer’s onset ten years in advance

A 17-year-long study

The team conducted this study on 233 participants at the Swedish National Study on Aging and Care in Kungsholmen to evaluate different biomarkers for dementia. The participants' samples were collected between 2001 and 2004. In addition, all participants were checked on a regular basis for signs of memory loss and dementia. The authors also performed follow-ups every three to six years for a total of 17 years.

The researchers were able to demonstrate the role of glycans in the blood, which may allow the disease to be detected early, through this study.

In this disease, brain neurons begin to die as a result of an accumulation of the proteins amyloid beta and tau. The authors previously demonstrated a link between tau protein and glycan levels in Alzheimer's disease patients.

“We also show that a simple statistical model that takes into account blood glycan and tau levels, the risk gene APOE4 and a memory test, can be used to predict Alzheimer’s disease to a reliability of 80 percent almost a decade before symptoms such as memory loss appear,” said Sophia Schedin Weiss, corresponding author. 

This knowledge could pave the way for the development of simple screening procedures for the better treatment of the individuals impacted by the disease. It is critically important as previous Alzheimer's drug clinical trials have highlighted that treatment should begin early to protect neurons from dying before it is too late.

The study is published in the journal Alzheimer's & Dementia.

Study abstract:

There is an urgent need for novel blood biomarkers for the detection of Alzheimer's disease (AD). We previously showed that levels of the bisecting N-acetylglucosamine glycan epitope was elevated in cerebrospinal fluid in AD. However, its diagnostic value in blood is unknown. We analyzed blood levels of bisecting N-acetylglucosamine and total tau in a retrospective cohort of 233 individuals. Progression to AD was compared between the groups using Cox regression. The predictive value of the biomarkers was determined by logistic regression.