Florbetapir, an FDA-approved PET tracer that detects amyloid in the brain, predicts global and cognitive decline much like its predecessor, Pittsburgh compound B (PiB), according to a new study. Scientists led by Murali Doraiswamy, Duke University, Durham, North Carolina, and Michael Pontecorvo, Avid Radiopharmaceuticals, Philadelphia, report in the March 11 Molecular Psychiatry that amyloid pathology detected by florbetapir, trade name Amyvid®, forecast deterioration over three years in people who were cognitively normal, had mild cognitive impairment (MCI), or were diagnosed with Alzheimer’s disease (AD). “It’s encouraging to see that this tracer acts like PiB,” said Catherine Roe of Washington University in St. Louis. “This compound, and ones like it, have great promise for the clinic, so it’s important to study them longitudinally to see if they behave the same way.”  Roe was not involved in the study. 

Previously, several studies reported that a positive PiB-PET scan predicts subsequent cognitive decline (see Okello et al., 2009Wolk et al., 2009). Two years ago, the Food and Drug Administration approved PiB’s longer-lived, F-18-labeled cousin florbetapir for clinical use (see Apr 2012 news story). Doraiswamy and colleagues set out to assess whether florbetapir predicts decline in a similar way, and if it could be used to identify people who will progress fastest for clinical trials. 

The group first conducted a cross-sectional study and found that florbetapir binding correlated with the severity of disease (see Johnson et al., 2013). They followed 152 of those people longitudinally. Sixty-nine participants were cognitively normal, 52 had MCI, and 31 had been clinically diagnosed with AD. According to florbetapir imaging, 14 percent of controls, 37 percent of those with MCI, and 68 percent of AD patients tested positive for brain amyloid. At baseline and 18 months, the researchers tested cognition, using the ADAS-Cog, MMSE, digit symbol substitution (DSS) test, and a verbal fluency test. The 18-month data suggested that a greater amyloid signal at baseline foretold decline on several tests (see Doraiswamy et al., 2012). 

In the present paper, Doraiswamy and colleagues report 36-month data on 51, 37, and 16 people from the cognitively normal, MCI, and AD groups. Those cognitively normal people who were Aβ-positive and those with MCI who had an Aβ-positive scan deteriorated an average of 3.2 and 5.8 points on the ADAS-Cog, respectively, while people in both those clinical groups who were amyloid-negative improved slightly. Scores in the other tests changed in parallel. Florbetapir-positive AD patients gained 8.88 points on average (higher scores indicate worse cognition on this test). In contrast, those who had received a diagnosis of AD but turned out to test negative for brain amyloid worsened by 3.81 points, an insignificant change from baseline. Six of the Aβ-positive MCI patients progressed to AD dementia after three years, as opposed to four with no evidence of amyloid. 

Regardless of their cognitive status, people who tested positive for brain Aβ declined faster than those whose scan was negative. While the authors did not address in this paper whether amyloid load changed in any of these participants, they did re-scan them at 36 months and will report results in a later publication. 

Christopher Rowe of Austin Hospital, Melbourne, Australia, pointed out that the numbers of plaque-positive MCI and AD patients fall well below the 60 and 90 percent, respectively, that his group had previously found in AIBL, the Australian Imaging, Biomarker and Lifestyle Flagship Study of Ageing (see Villemagne et al., 2011, and Pike et al., 2007). Roe added that 30 percent of controls usually test positive for amyloid, so the 14 percent in the current study is also small by comparison. Doraiswamy said that these values are different because while previous research groups quantified the signal strength, his group read the scans as either positive or negative to simulate the way scans are read in the clinic. That meant they used higher cutoff values to designate a positive test. 

Rowe noted the rates of progression to AD in the MCI group seemed low as well. In AIBL, 86 percent of amyloid-positive MCI subjects progressed to AD over three years. However, the Australian study uses a stricter clinical definition of MCI, setting low cutoff scores on cognitive tests. Doraiswamy included patients with included patients with milder MCI, a form community doctors would likely see. “This illustrates the need for a consistent and objective definition of MCI, otherwise its prognostic implications cannot be accurately stated to patients,” said Rowe. David Knopman, Mayo Clinic, Rochester, Minnesota, noted in an email (see full comment below) that the inclusion of neurodegenerative markers can improve the prognostic ability of amyloid scans. For instance, his group previously reported that evidence of hippocampal atrophy on MRI can indicate whether amyloid pathology is accompanied by neuron loss (see Knopman et al., 2012). 

Several amyloid imaging-based studies have reported that some patients with clinically diagnosed AD show no evidence of amyloid accumulation in the brain (see Aug 2013 news story). In the current study, this population declined much more slowly, sometimes not at all, compared with their plaque-positive counterparts. Rowe hypothesized that these patients have a slower-progressing, non-amyloid pathology.—Gwyneth Dickey Zakaib


  1. This study represents the 36-month follow-up of a group of individuals in the AD spectrum who initially underwent Amyvid imaging as part of an Avid Pharmaceuticals-sponsored trial. As was expected from many studies using 11C PiB, the authors found that cognitively normal (CN) individuals who were amyloid positive (Aβ+) had a faster rate of cognitive decline than those who were amyloid negative (Aβ-). The authors did not report on whether Aβ+ CN participants had a higher rate of conversion to MCI, but presumably not (or they would have said so). Similarly, those with MCI had a higher rate of progression to AD dementia and a lower rate of regression to CN if Aβ+. These observations replicate many others that have been reported in MCI.

    The authors also report that persons with dementia who were Aβ+, i.e., intermediate likelihood, or higher, AD dementia also worsened cognitively more than Aβ- dementia patients. It is of interest that in this cohort, presumably recruited and diagnosed by skilled clinicians, that the rate of Aβ+ was lower in both MCI (38 percent) and especially dementia (68 percent) than most other studies. This does not invalidate their findings, but it raises questions about the stringency of the diagnostic process at enrollment. The main weakness of this study is that it does not break much in the way of new ground, except to the extent that it shows that Amyvid behaves like PiB. In addition, the use of amyloid imaging alone, without biomarkers for neurodegeneration (ND), clearly gives an incomplete picture of prognosis. As in our work (Knopman et al., 2012; Petersen et al., 2013) and that of others (Prestia et al., 2013), the distinction between Aβ+ND- and Aβ+ND+ is quite relevant for prognosis.


    . Short-term clinical outcomes for stages of NIA-AA preclinical Alzheimer disease. Neurology. 2012 May 15;78(20):1576-82. PubMed.

    . Criteria for mild cognitive impairment due to alzheimer's disease in the community. Ann Neurol. 2013 May 20; PubMed.

    . Prediction of dementia in MCI patients based on core diagnostic markers for Alzheimer disease. Neurology. 2013 Mar 12;80(11):1048-56. Epub 2013 Feb 6 PubMed.

    View all comments by David Knopman

Make a Comment

To make a comment you must login or register.


News Citations

  1. FDA Approves Amyvid for Clinical Use
  2. Suspected Non-Amyloid Pathology (SNAP)—Not an Open or Shut Case

Paper Citations

  1. . Conversion of amyloid positive and negative MCI to AD over 3 years: an 11C-PIB PET study. Neurology. 2009 Sep 8;73(10):754-60. PubMed.
  2. . Amyloid imaging in mild cognitive impairment subtypes. Ann Neurol. 2009 May;65(5):557-68. PubMed.
  3. . Florbetapir (F18-AV-45) PET to assess amyloid burden in Alzheimer's disease dementia, mild cognitive impairment, and normal aging. Alzheimers Dement. 2013 Oct;9(5 Suppl):S72-83. PubMed.
  4. . Amyloid-β assessed by florbetapir F 18 PET and 18-month cognitive decline: a multicenter study. Neurology. 2012 Oct 16;79(16):1636-44. PubMed.
  5. . Longitudinal assessment of Aβ and cognition in aging and Alzheimer disease. Ann Neurol. 2011 Jan;69(1):181-92. PubMed.
  6. . Beta-amyloid imaging and memory in non-demented individuals: evidence for preclinical Alzheimer's disease. Brain. 2007 Nov;130(Pt 11):2837-44. PubMed.
  7. . Short-term clinical outcomes for stages of NIA-AA preclinical Alzheimer disease. Neurology. 2012 May 15;78(20):1576-82. PubMed.

Further Reading

Primary Papers

  1. . Florbetapir F 18 amyloid PET and 36-month cognitive decline: a prospective multicenter study. Mol Psychiatry. 2014 Sep;19(9):1044-51. Epub 2014 Mar 11 PubMed.