. PET of brain amyloid and tau in mild cognitive impairment. N Engl J Med. 2006 Dec 21;355(25):2652-63. PubMed.


Please login to recommend the paper.


Make a Comment

To make a comment you must login or register.

Comments on this content

  1. I want to congratulate the authors for this interesting, complex, well-performed study. It is stimulating to see that great effort is undertaken in the field of the PET technology to detect in vivo, pathological changes occurring in neurodegenerative diseases.

    In general, I agree with the comments by Professors Rowe, Blennow, and Nordberg.
    In my opinion, the strongest potential of FDDNP is the capability to detect neurofibrillary tangles, whereas PIB remains as a strong amyloid marker with a high level of discrimination between subjects with amyloid deposition and subjects without it (90 percent difference in parietal cortex!). More study must be done to clarify the future role of these two novel tracers in clinical routine as well as in the follow-up of different treatments. Meanwhile, here are some reflections:

    1. PET is not a precision tool. For a PET tracer in clinical routine, its level of discrimination is an important factor. The long scanning time and the small percent difference between groups represent to me disadvantages of FDDNP compared with PIB.

    2. For differential diagnosis and for monitoring drug treatments, the specificity of binding of the tracer is also important. The new approaches for treatment of AD concentrate on anti-amyloid drugs or drugs blocking the production of hyperphosphorylated tau. Because of this, it is important to differentiate between amyloid and neurofibrillary tangles. It might be a disadvantage for FDDNP to bind to both substances. But FDDNP might be useful to detect general signs of neurodegeneration. However, the binding to tangles in people older than the controls who participated in the study must be studied.

    3. In my opinion, the comparison between FDDNP and FDG should have been done using the regional cerebral metabolic rate of glucose (rCMR glc) instead of SUVr. Variations in plasma glucose produce important changes in uptake.

    My conclusion is that the study by Small et al. confirms that we are at the beginning of a new imaging era, in which we may be able to understand for the first time complicated pathological processes in vivo that we are today just scratching on the surface.

  2. This comment was co-authored by Gary Small, Henry Huang, Vladir Kepe, and
    Jorge Barrio

    We appreciate the interest that our paper has generated and wish to comment on some of the previous observations and clarify some points. In our New England Journal of Medicine paper (2006;355;2652-2663), we used a full 120 minute scan time for the purpose of validation of the quantification method, but we have also evaluated successfully the reduction of the FDDNP scan time from 120 to 60 minutes. For routine applications, we have found that a late scan reading between 30 to 60 minutes is not only sufficient but makes the procedure very easy to tolerate.

    As presented in the paper, we have performed a brain autopsy in one of the patients scanned with FDDNP. Whereas tangles were found predominantly in the medial temporal lobe, plaques were the predominant pathology in the rest of the brain, confirming data in the literature. With knowledge of the distribution of plaques and tangles in a degenerating brain, FDDNP-PET could be used as a surrogate marker to monitor either anti-plaque or anti-tangle treatments or both. For an anti-tangle treatment, investigators would predict a greater effect size in the medial temporal region; for an anti-plaque treatment, the effect size would be greater in other cortical areas. Thus, FDDNP offers potentially great versatility as a surrogate marker tool.

    Our approach to recruitment was far from unusual but rather typical for obtaining a convenience sample for study. The fact that our older controls had mild memory complaints is not unusual; in fact, nearly all individuals note subjective slowness in retrieval and learning with age. As pointed out in the paper, their self-acknowledged age-related memory complaints could have led to higher FDDNP-PET binding values since concern about memory complaints could be a subtle indication of pre-symptomatic disease in some controls. Despite this potential bias, we found that FDDNP-PET binding values for the control group were significantly lower than those for the MCI and the AD groups. Thus, these controls were clearly a distinct group from the MCI subjects who had more advanced objective cognitive decline. It is possible that some subjects were misclassified because of cholinesterase inhibitor treatment; however, when we eliminated subjects taking memory enhancing drugs from the analysis, our results did not change.

    The statement in one of the comments that only two of 28 subjects converted to MCI is incorrect since we did not have available follow up data on all 28 subjects. Of the four MCI subjects available for follow up, two converted AD, which is consistent with the expected conversion rate over the follow-up period (30 months) in this study for the MCI subjects.

    We agree with Dr. Blennow that a combination of biomarkers may be the best approach. Although we are using a combination of molecular imaging probes to properly evaluate the many variables associated with AD (e.g., neuropathological aggregates, neuronal losses and dysfunction), the added cost of multiple markers will have to be weighed against the added value of each measure when the approaches become routine clinical tools.

    The word “non-invasive” classically describes all PET imaging approaches. Brain biopsy or sampling of CSF in patients are more invasive approaches to document brain amyloid and tau in living patients than a PET scan.

    We agree that studies of FDDNP and PIB in the same subjects will further help us to understand these novel imaging approaches in living subjects, and such studies are currently underway at several sites.

This paper appears in the following:


  1. PET Probe Lights Up MCI, AD
  2. Salzburg: Amyloid Imaging Update
  3. Tau Tracer May Light Up All Tauopathies