JAMA: Florbetapir-PET may quantify amyloid, detect Alzheimers
|Florbetapir-PET scan of high beta-amyloid burden (score 4 of 4), correlated with Alzheimer’s. |
Image Source: JAMA 2011;305: 275-283.
With 10 to 20 percent of clinically diagnosed Alzheimer's disease (AD) patients lacking AD pathology at autopsy, both the diagnosis and determination of risk for AD remain poor, despite up to 25 percent of persons eventually acquiring the disease. Extensive research is underway to discover ligands that can visualize beta-amyloid burden in the brain, a signature of AD, to diagnose and effectively determine risk for the disease.
While several studies have found that 11C-labeled Pittsburgh compound B (11C-PiB) may provide a robust visualization of beta-amyloid burden, the compound's 20-minute half-life limits its availability to specialized research centers.
With the ligand florbetapir F 18 having shown preliminary affinity and specificity to beta-amyloid, the researchers of the present study set out to "determine if florbetapir F 18 PET imaging performed during life accurately predicts the presence of beta-amyloid in the brain at autopsy," wrote Christopher M. Clark, MD, from Avid Pharmaceuticals and the University of Pennsylvania School of Medicine, both in Philadelphia, and co-authors.
Thirty-five elderly individuals underwent florbetapir-PET scans within 12 months of death. "Florbetapir-PET images were assessed visually using a semiquantitative visual score ranging from 0 (no amyloid) to 4 (high levels of cortical amyloid) by three board-certified nuclear medicine physicians who were not involved in any other aspects of the study," according to Clark and colleagues. These findings were correlated to cortical beta-amyloid pathology at autopsy (measured using immunohistochemistry or by silver strain), with the pathologists blinded to imaging results.
Additionally, florbetapir-PET was used to evaluate 47 young (mean age 26 years) and cognitively healthy individuals, with the hypothesis that the imaging would reveal fewer than 10 percent of these healthy participants presenting with beta-amyloid burden.
Good correlation was observed between whole brain florbetapir-PET scores and the postmortem amyloid pathology measured by both immunohistochemistry and silver stain neuritic plaque score, with values of .78 and .71, respectively. Of the 29 autopsied patients considered for primary analysis, 15 met pathological criteria for AD and 14 had low levels of beta-amyloid aggregations, therefore not fulfilling the pathological criteria for AD their late-in-life diagnoses.
Within these findings, 14 of 15 individuals were positively identified with AD using florbetapir-PET (sensitivity 93 percent) and 14 of the 14 patients with pathological confirmation as AD-negative were evaluated as such using PET (specificity 100 percent). Including the additional six patients assessed in the primary analysis only strengthened these results.
"In total, the blinded read results for the florbetapir-PET images agreed with the final autopsy with respect to the presence or absence of neuropathological criteria of AD in 28 of 29 cases," explained Clark and colleagues. They argued that this finding "suggests that a florbetapir- PET image provides an accurate and reliable assessment of beta-amyloid burden."
As hypothesized, all 74 of the younger individuals imaged presented with florbetapir-PET images that were negative for beta-amyloid burden. The agreement among the three nuclear medicine readers across the whole ranged from 91 to 99 percent.
Also of interest on an exploratory basis, one false-negative and two false-positive diagnoses of AD or dementia late in life were found via autopsy, whereas florbetapir-PET agreed with the autopsy results (considered the gold standard) in all three cases.
"[T]he authors convincingly showed that the antemortem imaging signals related to postmortem amyloid pathology," wrote Monique M. B. Breteler, MD, PhD, of University Medical Center Rotterdam, in the Netherlands, in an accompanying editorial JAMA comment. "However," continued Breteler, "the present study falls short in evaluating the sensitivity and specificity of florbetapir imaging."
Breteler called the 97 percent agreement between florbetapir-PET and the postmortem pathology "likely inflated," due to the nonrandom selection of persons who were more likely to have beta-amyloid burdens. Similarly, the findings that none of the younger cohort presented under florbetapir-PET with beta-amyloid burden could equally overestimate the specificity of florbetapir-PET in detecting beta-amyloid burden.
"[J]ust as the observation of a flock of white swans does little to support the notion that no black swans exist, so the observation of amyloid-negative scans in young persons does little to establish the specificity of this imaging mode in persons aged in their 70s and 80s, who may or may not have varying amounts of coexisting brain atrophy and other pathologic changes," Breteler argued.
Still, Clark and colleagues affirmed the importance of their findings in adding to a body of evidence indicating that biomarkers might signal beta-amyloid burden and help diagnose, predict and even treat AD.