Simple blood test can accurately spot dementia years before diagnosis

Before the symptoms of cognitive loss are severe enough to be diagnosed, something in the blood seems to be changing.

That is what a new study out of the University of East Anglia supports. Scientists looked at blood and stool samples taken from older adults. It was found that there are associations between changes occurring in the blood and those that could indicate the presence of early cognitive decline. In particular, the research indicates that chemicals associated with gut bacteria may have an impact on the brain much sooner than formal diagnoses of dementia occur.

In this study, published in the journal Gut Microbes, the researchers worked with 150 adults aged 50 and older. All of the participants were grouped into three comparable (matched) groups of 50 participants each: cognitively healthy adults, individuals with subjective cognitive impairment (SCI), and individuals with mild cognitive impairment (MCI).

Subjective cognitive impairment (also known as self-reported cognitive impairment) means that an individual believes they are becoming forgetful, but their abilities to perform well on standard neuropsychological tests remain intact. Mild cognitive impairment (MCI) is often viewed as an increased risk factor for developing dementia.

"The growing prevalence of dementia presents an incredible public health challenge," said Dr. David Vauzour, the primary researcher on the study and a member of the UEA's Norwich Medical School. "Dementia affects approximately 1 million people in the UK, while there are an estimated 55 million people worldwide who have dementia."

The researchers identified a total of 33 specific molecules present in fasting blood samples from the participants. They found several of the molecules associated with tryptophan metabolism, bile acids, TMAO metabolites, and cresol metabolites. They also obtained stool samples from each of the participants to compare the blood chemistry of the participants to the type of bacteria that reside in their gut. Rather than a compound-based signal, the most reliable indicator was a combination of multiple signals.

Six blood "metabolites" (5-hydroxyindole acetic acid, indole 3-propionic acid, choline, indoxyl sulfate, kynurenic acid, and kynurenine) were used as a basis for building predictive models using machine learning. For predicting if a person belonged to one of the three populations listed above (healthy adult / MCI / SCI), the area under the ROC curve (AUC) was 0.79. In healthy adults versus MCI, predictions increased to a maximum of 0.84 based upon the RF model.

In addition to being measured via AUC, the classification was characterized by 79% accuracy, which distinguishes healthy adults from MCI subjects, as noted by an additional finding by the University of Liverpool.

"I found this to be incredibly interesting," said Vauzour. "We saw evidence of very early changes in gut bacteria and the metabolites they produce in the bloodstream, even with very early signs of cognitive impairment."

There was no evidence of widespread loss of diversity among participants' gut bacteria. Alpha diversity measures did not show significant differences between groups. There were, however, substantial differences in beta diversity between healthy adults and those reporting some level of cognitive impairment. In a nutshell, the earliest signs of cognitive decline may be seen when individuals say that something is not right, rather than when a doctor can detect a problem through traditional clinical testing.

Research has confirmed a link between the blood of participants and specific gut bacteria. The Spearman correlation analysis conducted in this study linked the 33 blood metabolites to the 69 identified species of gut microorganisms. This was also supported through Procrustes analysis, which showed similar movement between these two sets of data.

As Dr. Vauzour stated, "The chemical changes in the blood of the participants were strongly correlated to differences in the makeup of the gut bacteria."

Most importantly, of the metabolites assessed in this study, five were found to have statistically relevant influences on how people performed on cognitive decline tests after accounting for factors such as age, BMI, sex, albumin, kidney function, liver function, and diet. These included choline, 5-hydroxyindole acetic acid, indole-3-propionic acid, indoxyl sulfate, and kynurenic acid.

For those metabolites where significant association was established with participants' scores on cognitive decline tests, three (choline, 5-hydroxyindole acetic acid, and indole-3-propionic acid) declined with perceived or mild cognitive decline. The concentrations of indoxyl sulfate and kynurenic acid increased during the progression of decline.

While the results do not indicate that the development of a blood test for dementia is imminent, Dr. Simon McArthur of Queen Mary University of London stated that the findings indicate that, potentially, in conjunction with dietary and microbial factors, it may be possible to identify the presence of dementia at a much earlier stage. This could occur even prior to any substantial damage to the brain.

Dr. Saber Sami, the Alzheimer’s Research UK UEA lead and a co-author of the study, stated that this research represents an important accomplishment. It combines advanced statistical analyses with a practical objective: the identification of risk of developing dementia as early as possible, using methods that reduce the stigma associated with seeing a doctor.

While this study does have limitations, including that this was an exploratory study of only 150 subjects, the authors could not definitively establish a cause-and-effect relationship. Additionally, as noted by the authors, some of the metabolites that changed in concentration may have been associated with systemic inflammation rather than cognitive decline specifically.

While the regression analysis conducted in this study did not adjust for multiple comparisons, the study did include the potential risks related to false positives. The dietary assessments completed using food frequency questionnaires are also subject to recall bias and error in the self-reported data collected. Finally, validation of the machine-learning analyses will require replication in larger and more independent populations.

This research suggests the possibility of detecting early indicators of cognitive decline from an individual's blood before they receive a diagnosis in a doctor's office. If the findings from larger studies continue to support the authors’ conclusions, doctors may one day use a simple blood test to refer patients for additional monitoring or earlier implementation of lifestyle interventions.

This research also enhances the momentum behind studies into the gut microbiome as a target for strategies aimed at preventing cognitive decline.

These strategies may include dietary interventions, probiotic supplementation, and the development of microbiome-based therapies or more individualized nutritional approaches.

Research findings are available online in the journal Gut Microbes.

The original story "Simple blood test can accurately spot dementia years before diagnosis" is published in The Brighter Side of News.

Like these kind of feel good stories? Get The Brighter Side of News' newsletter.