Literature Review

Automated measurements of plasma predict amyloid status

View on the News

Findings represent significant advances

The investigation by Palmqvist et al. “makes several significant advancements in the field,” said Sid E. O’Bryant, PhD, professor of pharmacology and neuroscience at the University of North Texas Health Science Center in Fort Worth, in an accompanying editorial. The study’s protocol design clears the ground for a context of use of a blood screen for amyloid positivity. Also, the fully automated immunoassay “yields performance measurements that are superior to [those of] many earlier nonautomated procedures,” said Dr. O’Bryant. When Dr. Palmqvist and colleagues applied their discovery findings from a training cohort directly to a test cohort, it produced strong results. “This study suggests that the field is one step closer to the actual application of blood-based biomarkers with specific contexts of use in Alzheimer’s disease.”

The main concern about the plasma biomarkers, however, is the scalability of the methods used to measure them. “If primary care physicians are to use such a technology, the technology must have the capacity to conduct hundreds of millions of assays annually around the globe,” said Dr. O’Bryant. “A blood test for primary care must fit into the existing protocols and parameters in clinical laboratory settings. The blood collection and processing procedures are not applicable to standard clinical lab practice and will cause substantial barriers to clinical application.”

In addition, the study authors emphasize the utility of the immunoassay for primary care, but the study was designed to test for amyloid positivity, which is more appropriate for clinical trials. “No currently available drugs for patient use target amyloid,” said Dr. O’Bryant. “Therefore, this specific context of use is geared more toward clinical trial application than primary care physicians who currently need a test for the presence or absence of Alzheimer’s disease so currently available treatments and support can be put in place for patients and family members.”

Nevertheless, Dr. Palmqvist and associates have presented promising data, Dr. O’Bryant continued. The question in the field is ceasing to be whether blood biomarkers can be used in Alzheimer’s disease, and becoming how they can be used.


 

FROM JAMA NEUROLOGY

Measuring plasma amyloid-beta 42 and amyloid-beta 40 using a fully automated immunoassay predicts amyloid-beta status in all stages of Alzheimer’s disease, according to research published online ahead of print June 24 in JAMA Neurology. Analyzing APOE genotype in addition to these biomarkers increases the accuracy of the prediction. This blood test thus could allow neurologists to identify patients at risk of amyloid-beta positivity who should undergo further assessment, said the authors. It also could be used to enroll amyloid-beta–positive participants in clinical trials.

Dr. Sebastian Palmqvist, neurologist at Skåne University Hospital in Malmö, Sweden

Dr. Sebastian Palmqvist

In vivo PET imaging and analysis of cerebrospinal fluid (CSF) can detect amyloid-beta, but these procedures are expensive, and their availability is limited. Clinicians need readily available methods for detecting amyloid-beta, and research has indicated that blood-based biomarkers correlate with those in CSF. Fully automated immunoassays, such as the Elecsys test developed by Roche Diagnostics, have recently demonstrated high reliability and precision for CSF amyloid-beta. Using the Elecsys assay, Sebastian Palmqvist, MD, PhD, a neurologist at Skåne University Hospital in Malmö, Sweden, and colleagues sought to examine the accuracy of plasma amyloid-beta and tau, together with other blood-based biomarkers, at detecting cerebral amyloid-beta.

Testing the immunoassay in two cohorts

Dr. Palmqvist and colleagues examined participants in the prospective Swedish BioFINDER Study, which enrolled patients between July 6, 2009, and February 11, 2015. This cohort included 513 cognitively unimpaired (CU) participants, 265 participants with mild cognitive impairment (MCI), and 64 participants with Alzheimer’s disease dementia. Investigators collected blood and CSF samples at the same time from all participants. Participants’ amyloid-beta status was ascertained using the Elecsys CSF amyloid-beta 42/amyloid-beta 40 ratio. The researchers defined amyloid-beta positivity with an unbiased cutoff of less than 0.059.

Dr. Palmqvist and colleagues also examined a validation cohort that included 237 participants who had been enrolled between January 29, 2000, and October 11, 2006, in Ulm and Hannover, Germany. This group included 34 CU participants, 109 participants with MCI, and 94 participants with mild Alzheimer’s disease dementia. The investigators applied the same cutoff of CSF amyloid-beta 42/amyloid-beta 40 to define amyloid-beta positivity in this cohort as they applied to the BioFINDER cohort.

Automated immunoassay had high predictive accuracy

The mean age of the BioFINDER cohort was 72 years, and 52.5% of participants were female. Overall, 44% of this cohort was amyloid-beta positive, including 29% of CU participants, 60% of participants with MCI, and 100% of participants with Alzheimer’s dementia. The investigators found statistically significant positive correlations between all plasma and corresponding CSF biomarkers in this cohort.

Plasma amyloid-beta 42 and amyloid-beta 40 levels predicted amyloid-beta status with an area under the receiver operating characteristic curve (AUC) of 0.80. When the researchers added APOE to the model, the AUC increased significantly to 0.85. Accuracy improved slightly when the researchers added plasma tau (AUC, 0.86) or tau and neurofilament light (AUC, 0.87) to amyloid-beta 42, amyloid-beta 40, and APOE. The results were similar in CU and cognitively impaired participants, and in younger and older participants.

In the validation cohort, the mean age was 66 years, and 50.6% of participants were female. When Dr. Palmqvist and colleagues applied the plasma amyloid-beta 42 and amyloid-beta 40 model from the BioFINDER cohort to this population, they obtained a slightly higher AUC (0.86), but plasma tau did not increase predictive accuracy.

The investigators performed a cost-benefit analysis using a scenario in which 1,000 amyloid-positive participants are included in a trial and given a cost of $4,000 per participant for amyloid PET. Using plasma amyloid-beta 42, amyloid-beta 40, and APOE in this scenario reduced PET costs by as much as 30%-50%, depending on the cutoff.

Next Article: