A combination of two plasma-based biomarkers, tau phosphorylated at threonine-181 (P-tau181) and neurofilament light (NfL), may predict future cognitive decline and progression to Alzheimer’s disease dementia in individuals with mild cognitive impairment (MCI), a study shows.

In a paper published in Nature Aging, scientists in Sweden showed that the two dementia biomarkers can provide patient-level prognostic information in MCI comparable to cerebrospinal fluid biomarkers.

The findings could aid the development of routine blood tests to track Alzheimer’s disease progression in at-risk populations, said researchers from the Clinical Memory Research Unit, Lund University, Malmö.

In their paper, the researchers said Alzheimer’s disease was characterised by the accumulation of these proteins in the brain that are thought to cause neuronal death, eventually leading to dementia.

The proteins are also found in blood and they sought to investigate whether tests based on their plasma concentrations could be used to diagnose Alzheimer’s disease or distinguish it from other common forms of dementia.

A team of researchers from Sweden and the UK developed and validated models of individualised risk prediction of cognitive decline and transition to Alzheimer’s disease dementia using data from 573 patients with minor cognitive impairments from two independent cohorts.

The authors compared the accuracy of several models based on various combinations of blood biomarkers to predict cognitive decline and dementia over four years.

They found that  a model combining tau phosphorylated at P-tau181 and NfL, but not Aβ42/Aβ40, had the best prognosis performance of all models. The area under the curve (AUC) was 0.88 for four-year conversion to AD. This was stronger than a basic model of age, sex, education and baseline cognition, and performed similarly to CSF biomarkers.

The authors concluded that their findings demonstrated the value of using specific combinations of blood-based biomarkers to make individualised predictions about the progression of Alzheimer’s disease.

They also provided a publicly available online individualised risk prediction tool at predictprogression.com for people with MCI, based on their combined plasma biomarker models.

However, standardised assays with universal cut-offs and replication of the findings in large cohorts were still needed, they cautioned.