A prototype test for Alzheimer’s disease biomarkers using dried blood spots is almost as accurate as a standard blood test for the same factors and accurately predicts Alzheimer’s disease-related changes 86% of the time.
“Blood biomarkers have emerged as accurate tools for detecting Alzheimer’s disease pathology, offering a minimally invasive alternative to traditional diagnostic methods such as imaging and cerebrospinal fluid analysis,” write lead author Nicholas Ashton, PhD, a researcher based at Banner Sun Health Research Institute in Arizona and also affiliated with the Sahlgrenska Academy at the University of Gothenburg, and colleagues in Nature Medicine.
“Yet, the logistics surrounding venipuncture for blood collection, although considerably simpler than the acquisition of imaging and cerebrospinal fluid, require precise processing and storage… guided by medical personnel. Consequently, limitations in their widescale use in research and broader clinical implementation exist.”
In this study, Ashton and team collected data from 337 individuals from various clinical sites. These included older adults who were cognitively unimpaired or who had some form of cognitive impairment or dementia like Alzheimer’s disease. A small Down syndrome group of 31 younger individuals, some showing signs of Alzheimer’s or other cognitive impairment and some not, was also included as people with the genetic condition are at high risk of early onset Alzheimer’s disease.
The main purpose of the study was to compare the accuracy of testing for phosphorylated tau‑217 (p‑tau217), which is now widely accepted as a highly accurate blood biomarker of Alzheimer’s‑type amyloid and tau pathology, using standard blood tests and dried blood spot testing.
Ashton and colleagues also tested for glial fibrillary acidic protein (GFAP) and neurofilament light (NfL) in some of the group. These are linked to Alzheimer’s but not as strongly as p-tau217 so the research team tested them more as supporting markers of brain inflammation rather than a specific Alzheimer’s disease diagnostic.
Overall, 252 of the 337 participants had both types of blood tests completed and 176 of the 337 participants had both types of blood test and also had cerebrospinal fluid testing carried out.
Of the 176 individuals who had all three types of test, 56% had signs of Alzheimer’s disease in their cerebrospinal fluid and 44% did not.
Both blood tests predicted the cerebrospinal‑fluid Alzheimer’s status well, but the standard blood test was more accurate than the finger‑prick blood spot test. More specifically, the standard test had a predictive accuracy of around 98% and the finger-prick test had an approximate accuracy of 86%. Both GFAP and NfL were also detected in dried blood spot tests at similar levels to those seen in the matching standard blood tests.
Although the finger prick test was a bit less accurate than the standard blood test overall, the test itself is much easier and less invasive to carry out. It can be done at home or in remote clinics and posted to a lab, which makes it a lot easier to reach large numbers of people and communities underserved by standard clinic testing.
While 86% accuracy for a medical test is still considered ‘good discrimination,’ “further methodological refinement and validation will be essential before clinical translation can be considered,” conclude the authors, who acknowledge the lower performance of the blood spot test versus the standard blood test makes it difficult to recommend wider rollout at the moment.
