More evidence needed to support PET imaging in AD

Current literature on amyloid-beta PET imaging for Alzheimer’s disease (AD) is insufficient to provide conclusive evidence that the test leads to better clinical outcomes, according to a review summarized online Sept. 30 in JAMA Internal Medicine.

“The available evidence on [amyloid-beta] PET has many limitations beyond the sparse number of studies,” wrote Steven D. Pearson, MD, and colleagues from the Institute for Clinical and Economic Review in Boston. “Nearly all the studies have evaluated diagnostic performance, not whether test results improve clinical management or patient outcomes. Many potential benefits and harms of the imaging have not been evaluated.”

The review of evidence was conducted by the Institute for Clinical and Economic Review in 2012 and was funded by private foundations, insurers and a variety of other companies. A total of 558 articles had data on potential diagnostic tests for AD, 15 of which focused on amyloid-beta PET.

One study from Clark et al published in 2012 in the Lancet Neurology was described as “pivotal” by Pearson and colleagues. This study found that in its patient population, 66 percent of whom had AD diagnosed by post-mortem biopsy, the median sensitivity of amyloid-beta PET was 92 percent and the specificity was 95 percent. However, the authors pointed out that the ranges for sensitivity and specificity were 69 percent to 95 percent and 90 percent to 100 percent, respectively.

Lacking in the literature is evidence of the impact of an amyloid-beta PET scan on clinical management, particularly for patients with non-AD dementia, according to Pearson and colleagues. Effect on family anxiety and life planning also has not been assessed, and the current studies may not feature physicians or patient populations that are representative nationally.

“At present, the medical literature provides extremely limited data with which to evaluate the clinical utility of [amyloid-beta] PET,” wrote the authors. “There are reasonable data showing that, when read by well-trained interpreters, [amyloid-beta] PET is highly accurate in determining whether there is amyloid in the brain. However, the clinical utility of a positive scan result remains uncertain.”

Last week, the Centers for Medicare & Medicaid Services released a coverage decision denying coverage for amyloid-beta PET scans in routine clinical care, limiting coverage to one test per patient through coverage with evidence development.