Researchers have now confirmed that PET imaging with the contrast agent Pittsburgh Compound B (PiB) can detect amyloid-ß in the brains of dementia patients. Early work with mouse models of Alzheimer’s disease showed that the imaging agent can detect amyloid-ß in the brains of mice, and later clinical studies with dementia patients suggested that PiB can label amyloid-ß in humans, but validating these results requires pathological analysis of the brain. Researchers from Massachusetts General Hospital in Boston and the University of Pittsburgh have performed the first postmortem study of a patient who participated in a PET imaging study using PiB.The 76-year-old patient was evaluated over several years and exhibited symptoms that corresponded to many different neurological disorders. A conventional PET scan was consistent with a diagnosis of Alzheimer’s disease. Eventually, the patient was diagnosed with dementia with Lewy bodies, a common condition. Then the patient participated in the PET imaging study using PiB.The imaging study showed that PiB labeled the entire cerebral cortex, the outer layer of the brain (see figure at right). Three months after the imaging study, the patient died from an unrelated cause, and an autopsy was performed. The autopsy confirmed the diagnosis of dementia with Lewy bodies. The distribution of amyloid-ß seen at autopsy matched the overall distribution seen with PiB imaging when the patient was alive, confirming that the imaging agent detects amyloid-ß in humans (see figure below). The researchers reported these findings in the March issue of Archives of Neurology.The autopsy showed also that amyloid-ß had accumulated mostly in the blood vessels, a common condition known as cerebral amyloid angiopathy. PiB could not differentiate between this condition and Alzheimer’s disease. Although the contrast agent cannot in itself provide a diagnosis, the researchers believe that it may be useful for tracking the effect of a drug that rids the brain of amyloid.Brian J. Bacskai, the primary author of the study, said that the specificity and sensitivity of PiB are not completely established and that it will be important to determine positive and negative correlations for labeling after many more autopsies.