When the Tests Don’t Agree: Why Amyloid PET Sometimes Deserves a Second Look

In memory and brain-health care, no single test should be interpreted in isolation.

A patient’s history, cognitive testing, blood biomarkers, MRI, and amyloid PET should tell a reasonably coherent story. When they do not, I do not assume that one result ends the discussion.

A newly published analysis from the IDEAS study found that real-world amyloid PET interpretations are generally reliable—but disagreement between local and expert readers still occurs, particularly when reader confidence is lower. Large-scale 2026 data also support the value of standardized Centiloid quantification as an additional way to interpret amyloid PET burden.

That recently mattered in my own practice. A patient had an amyloid PET that was initially reported as negative. Later blood biomarkers strongly suggested amyloid pathology. Rather than dismissing the discrepancy, I requested additional review, including expert interpretation and quantitative Centiloid analysis.

The result changed our understanding of the case.

For me, the lesson is simple: good care sometimes means going back, asking another question, requesting expert review, and making sure the pieces truly fit together.

At Cardiel Precision Brain Health, my goal is not simply to order more tests. It is to look carefully at the entire clinical picture—and pursue clarification when the data do not make sense, especially when the answer may affect an important diagnosis or treatment decision.

Have cognitive concerns or conflicting test results? A comprehensive neurologic review may help clarify what the findings actually mean and what should come next.

Featured paper: Chiotis K, et al. Comparison of Expert Reads Versus Local Clinical Reads of Amyloid PET Scans in the IDEAS Study. Journal of Nuclear Medicine. Published September 10, 2026.