A multisite analysis of the concordance between visual image interpretation and quantitative analysis of [18F]flutemetamol PET images: Neuroimaging / Optimal neuroimaging measures for early detection. (7th December 2020)

• Record Type: Journal Article
• Authors: Farrar, Gill; Nordberg, Agneta K; Sur, Cyrille; Scott, David; Lhommel, Renaud; Collij, Lyduine; Molinuevo, Jose Luis; Hansson, Oskar; Gunn, Roger N; Rowe, Christopher C; Dore, Vincent; Buckley, Chris J
• Is Part Of: Alzheimer's & dementia. Volume 16(2020)Supplement 5
• Page Start: n/a
• Page End: n/a
• Publication Date: 2020-12-07
• Subjects: Alzheimer's disease -- Periodicals; Alzheimer Disease -- Periodicals; Dementia -- Periodicals; Démence; Maladie d'Alzheimer; Périodique électronique (Descripteur de forme); Ressource Internet (Descripteur de forme); 616.83
• DOI: 10.1002/alz.043157
• Languages: English
• ISSNs: 1552-5260

Abstract

Abstract: Background: [ 18 F]flutemetamol PET scanning provides information on brain amyloid load and has been approved for routine clinical use based upon visual interpretation of scans as either negative (equating to none or sparse amyloid plaques in pathology SoT), or amyloid positive (equating to moderate or frequent plaques). Quantitation is however fundamental to the practice of Nuclear Medicine and hence can be used to supplement amyloid reading methodology especially in unclear cases. Data was collected from multiple sources to assess comparability of the two methods of image interpretation. Method: 9 separate studies where visual reading of [ 18 F]flutemetamol and quantitative analysis of images was available were combined to form a single cohort to measure the percentage agreement between the two methods of image interpretation. Scans covered a wide range of cases (eg from cognitively unimpaired subjects to patients attending the memory clinics see table). Methods of quantifying amyloid ranged from using CE marked/510K cleared software (eg Cortex ID), to other research tools (eg PMOD, CapAIBL) as well as newer methods such as Amyloid IQ . Concordance between visual and quantiative imaging used thresholds robustly established using pathology as the standard of truth Result: A total of 2, 566 [ 18 F] flutemetamol images were collected from 9 clinical studies, visually interpreted by readers using the approved reading methodology and compared to quantitative measures using thresholds which had a binary delineation of amyloid. The mean concordance between visual read and quantitation was 95.01% (SD 3.00%), the median was 95.0%, and the range of the 9 studies was from 89.7% to 99.0%. A total of 379 images were compared using CE marked/510k cleared amyloid quantitation software and the percent agreement in these studies was 98.8% to 99.0%. Conclusion: Given the high concordance of visual and quantitation from over 2500 scans, users are able apply quantitative amyloid PET software methods to supplement visual inspection. The decision to use these methods as an adjunct to visual reading is left to the choice of the physician as both methods of image interpretation are comparable. The European Medicines Agency (EMA) have agreed to update the prescribing instructions for [ 18 F]flutemetamol (Vizamyl) based upon this analysis.