Amyloid PET has become a pivotal imaging biomarker for Alzheimer disease (AD), enabling in vivo detection and quantification of β-amyloid deposition. However, variability in quantitative measurements across tracers, scanners, and analysis methods has considerably limited direct comparison of amyloid burden between studies and centers. To address this issue, the Centiloid Project has been developed to provide a standardized quantitative scale for amyloid PET, harmonizing results across tracers and institutions. In this literature review, we aim to summarize current evidence on the development, validation, and clinical application of the Centiloid scale in amyloid PET imaging, emphasizing its methodologic foundations, tracer-specific conversions, and diagnostic thresholds. Methods: A literature review was conducted using PubMed, Scopus, and Embase databases from January 2015 through October 2025. Search terms included "amyloid PET," "Centiloid," "SUVR," and "Alzheimer disease." We reviewed articles with quantitative amyloid PET analyses and Centiloid conversion and studies which compared multiple tracers using MIMneuro software Centiloid calibration. Results: The Centiloid method demonstrates strong intertracer and interscanner harmonization when standardized to the 11C-Pittsburgh compound B reference, with high correlation (r 2 > 0.9) across 18F-labeled tracers, including florbetapir, flutemetamol, and florbetaben. Most articles identified a Centiloid threshold between 20 and 25 Centiloids as indicative of significant amyloid pathology. Implementation of the Centiloid framework improved comparability across clinical trials and longitudinal studies, facilitating integration into the AT(N) research framework. Conclusion: The Centiloid scale represents a critical advancement in quantitative amyloid PET imaging, providing a universal reference that enhances data reproducibility, cross-trial comparisons, and clinical decision-making. Continued work is needed to expand standardization for novel tracers and hybrid imaging systems, ensuring full clinical translation of this metric in dementia imaging.
Doroudinia et al. (Tue,) studied this question.