Study: PET biomarker gets validation for Alzheimer's disease

Twitter icon
Facebook icon
LinkedIn icon
e-mail icon
Google icon
Representative 18F-flutemetamol–positive Alzheimer's disease case with PET scan.
Image source: Annals of Neurology

18F-flutemetamol performs similarly to the 11C-Pittsburgh compound B (11C-PIB) parent molecule within the same subjects and provides high test-retest replicability and potentially much wider accessibility for clinical and research use for Alzheimer’s disease, based on a study in the September issue of the Annals of Neurology.

The most widely studied PET ligand for in vivo beta-amyloid imaging is 11C-PIB, according to the study authors, who added that its availability is limited by the need for an onsite cyclotron. As a result, they suggested that the validation of the 18F-labeled PIB derivative 18F-flutemetamol could “significantly enhance access to this novel technology.”

For the study, Rik Vandenberghe, MD, PhD, from the neurology department at University Hospitals Leuven in Leuven, Belguim, and colleagues included 27 patients with early-stage clinically probable Alzheimer's disease, 20 with amnestic mild cognitive impairment, 15 cognitively intact healthy volunteers above and 10 healthy volunteers below the age of 55 years.

The primary endpoint was the efficacy of blinded visual assessments of 18F-flutemetamol scans in assigning subjects to a raised versus normal uptake category, with clinical diagnosis as the standard of truth, according to the authors. As secondary objectives, the researchers determined the correlation between the regional standardized uptake value (SUV) ratios for 18F-flutemetamol and its parent molecule 11C-PIB in 20 of the subjects with Alzheimer's disease and 20 of the patients with mild cognitive impairment. They also sought to determine test-retest variability of 18F-flutemetamol SUV ratios in five of the subjects with Alzheimer's disease.

The investigators found that blinded visual assessments of 18F-flutemetamol scans assigned 25 of 27 scans from subjects with Alzheimer's disease and one of 15 scans from the elderly healthy volunteers to the raised category, corresponding to a sensitivity of 93.1 percent and a specificity of 93.3 percent against the standard of truth.

Correlation coefficients between cortical 18F-flutemetamol SUVRs and 11C-PIB SUV ratios ranged from 0.89 to 0.92, according to the authors, who added that the test-retest variabilities of regional SUV ratios were 1 to 4 percent.

Overall, Vandenberghe and colleagues wrote that the trial met its primary objective, with visual assignment of 18F-flutemetamol scans by readers blinded to the clinical diagnosis led to correct categorization of scans according to the standard of truth in 25 of 27 clinically probable cases with Alzheimer’s disease and in 14 of 15 elderly healthy volunteers, corresponding to a specificity of 96 percent and a sensitivity of 93 percent.

However, the researchers noted that a larger healthy volunteer sample size is needed before generally applicable SUV ratio values can be defined.

Based on these phase II results, the authors concluded that establishing the metrics of 18F-flutemetamol PET is an “important step” toward validating the modality as an imaging biomarker for Alzheimer’s disease-related beta-amyloidosis, with potentially much wider availability than its 11C-labeled parent molecule. “This novel ligand could vastly augment the impact of beta-amyloid imaging on research and clinical practice in the future,” they wrote.

GE Healthcare sponsored the study.