. Amyloid-β imaging with PET in Alzheimer's disease: is it feasible with current radiotracers and technologies?. Eur J Nucl Med Mol Imaging. 2012 Feb;39(2):202-8. PubMed.


Please login to recommend the paper.


  1. This story gives a fair and correct account of this battle. I personally believe that Alavi et al. are wrong with most of their statements, while Villemagne et al. are right with most of their rebuttal. The interview with Bengt Långström points to the remaining core of uncertainty about our understanding of amyloid imaging—he is probably right, but I would see the remaining uncertainty as a minor point. I have published my view on clinical amyloid imaging in a recent short review for Lancet Neurology (Herholz and Ebmeier, 2011).

    This debate is probably just "the tip of the iceberg" with respect to a much larger debate: whether it makes sense to test healthy individuals (or people with just some moderate memory deficit) for amyloid, assuming that a positive scan indicates that they already have an early stage of Alzheimer's disease and that we should aim for prevention of dementia in these subjects. PET is the obvious tool for doing this—thus, we now have the debate on PET, but it is not really the imaging science that is being discussed; it is the aim and implications of what it is going to be used for. This has potentially huge societal implications. Obviously, an effective strategy for early diagnosis and intervention to prevent dementia will need industry to pursue it with appropriate clinical trials, but with industry, big money and conflicts of interest also get involved quite heavily. Both we academics and patients need the pharmaceutical industry (just like they are needed to combat cancer), even though we don't love them. We need to be careful that things are being done right when engaging in early diagnosis of Alzheimer's disease.


    . Clinical amyloid imaging in Alzheimer's disease. Lancet Neurol. 2011 Jul;10(7):667-70. PubMed.

    View all comments by Karl Herholz
  2. This is an interesting and necessary debate. Interesting because things have been going so quickly: It took more than 20 years for hippocampal atrophy to be included as part of the Alzheimer’s disease diagnostic workup, while Aβ PET started being considered in revised diagnostic guidelines only a few years after the first PIB publication. Necessary, because the conflict of interest that inevitably accompanies the considerable economic dimension of this technique raises questions in the community’s mind. Some may have been wondering if this remarkable speed is because Aβ imaging truly represents a revolution in our field or because of its economic aspect. We need such debates to think about the use of this technique and its ethical aspects before its release, so that we can help prevent any abusive use or negative consequences. Debates such as this remind us of our duty, as scientists and clinicians, to be prudent.

    This provocative editorial, although some of its arguments were clumsy, is worth talking about. Let’s take the time to defuse some of the potential serious ethical issues that may be associated with Aβ imaging. Let’s make clear what Aβ imaging does and what it does not allow. That is how I interpret the spirit behind the paper by Moghbel et al., though I would personally not have used the same arguments. For example, we should acknowledge the limitation of Aβ imaging in that it only binds to fibrillar but not the soluble form of Aβ, but I don’t think we can question the ability of Aβ imaging to visualize plaque or refute that Aβ imaging is a revolution in the field of AD research. It fills the gap between neuropathologists and neuroimagers; it fills the gap between molecular, cellular, or ex-vivo experimentation and human in-vivo “macroscopic” research. It provides the most direct neuroimaging measure to date for a neuropathological hallmark of AD.

    This is very exciting. In spite of this, we clinician-scientists have the responsibility to proceed thoughtfully. I don’t agree with the arguments by Moghbel et al., but I salute their initiative, because it may calm down the industry race and because it gives Victor Villemagne and coauthors the opportunity to argue, answer, detail, explain, justify several critical points related to Aβ imaging. And they did that extremely well. They provide a systematic, objective, and documented answer to each of the issues raised by Moghbel et al. This enlightenment is extremely useful and was crucial.

    I fully agree with what I consider the most important point of the reply (as it is the most important confusion with regard to Aβ imaging). That is, we should not ask Aβ imaging to do what it is not supposed to do. It is not a tool to diagnose AD, but it is a tool to visualize Aβ deposition. As such, it is immediately a considerable breakthrough to further our understanding of AD pathology and progression. As for the clinical application, it would certainly be useful for diagnosis. My opinion is that we probably need to know a bit more about what information Aβ imaging provides, and what is the relative risk associated with an amyloid-positive scan (and, inversely, the percent chance of developing AD when someone has a negative scan). There are not enough studies to make firm conclusions about this yet. And we should think about the ethical issue. Absent a treatment, what do we do with the information provided by the amyloid scan? What information should be released to the patient?

    In my opinion, the real question at this point in time is not whether Aβ imaging is valid to visualize amyloid, but, What do we do with the information it provides beyond the evident interest for research?

    View all comments by Gael Chetelat
  3. I disagree with Moghbel et al. It is well established that PIB has been a very useful diagnostic tool, and most likely other experimental compounds, including those seeking FDA approval, are, too. The specificity of binding to amyloid fibrils issue may be relevant for early detection of Alzheimer's disease when smaller aggregates are predominantly present; however, additional experimental studies should be able to address this issue instead of asking for radical dismissal of the entire trace compound approach. I don't believe the radiotracers or dye tracers currently in development are fake. These compounds have very promising potential and are rather highly useful for AD diagnostics.

    View all comments by Olivera Mitrasinovic
  4. The recent editorial by Moghbel et al. has the merit of bringing to light a number of valid concerns that have permeated the interpretation of "amyloid imaging" scans for too long already. The purported amyloid specificity of these probes has been predicated on the basis of in vitro determinations certainly suitable for staining of fibrillar neuroaggregates, but not appropriate for identification of other tissue targets. Now we know that amyloid probes structurally related to the 6-hydroxybenzothiazole (and related) family (e.g., PIB, flutemetamol or 3’-fluoroPIB and others) are not amyloid-specific but are also targeted in vivo in brain by estrogen sulfotransferase (SULT1E1) (Cole et al., 2010), a labile low abundance cytosolic enzyme, first thought to be absent in the human brain (Mathis et al., 2004) As expected, the enzyme is present in all human tissues and also is extensively expressed in other animal species (Miki et al., 2002). What this means is that these amyloid PET probes can be retained in brain in vivo as their 6-O-sulfate, similar to the tissue retention of FDG as FDG 6-phosphate (the basis of the concept of metabolic trapping of PET probes), and not as a result of the presence of amyloid aggregates. This possibility has been demonstrated in rodent brain (Cole et al, 2010) and suggested already as a likely explanation for the multiple inconsistencies reported in the literature between positive PET scans and neuropathology localization in humans (Shin et al., 2011; Phelps and Barrio, 2011).

    We do not yet know what these purported amyloid probes detect in vivo: for instance, the PIB family of probes were first considered specific for neuritic plaques and insensitive to diffuse amyloid (Cairns et al., 2009; Reiman et al., 2009); later it was reported that PET positive signals were correlated with diffuse amyloid and not neuritic plaques (Kantarci et al., 2010), following reports that PIB binding correlated to total insoluble Abeta (Abeta40 plus Abeta42) (Ikonomovic et al., 2008). Most recently however it was reported that the PIB PET signal in a PIB(-) case was only correlated to Abeta42, and not Abeta40 plaques or diffuse amyloid, (Ikonomovic et al., 2012) but in a PIB(+) case it was correlated to both Abeta42 and Abeta40 plaques.

    Equally confusing is the report that Florbetapir PET amyloid positive imaging matches with autopsy cortical neuropathologic findings of frequent neuritic and diffuse plaques, as well as frequent amyloid angiopathy, in neocortex but not with frequent neuropathologically demonstrated diffuse plaques and amyloid angiopathy in cerebellum (Sabbagh et al., 2011). Then, what is it? What do these probes label in vivo in brain? It is probably none of the above.

    If the in vivo brain signal with these biomarkers is (totally or partially) related to estrogen sulfotransferase or other estrogen targets, then the PET positive signal could be related to inflammatory processes—well known to be involved with AD (Akiyama et al., 2000). It is also well-established that the stilbene-based Florbetapir and related probes are structurally related to estrogen analogs and alternative explanations associated with inflammation can be made for these probes also. An inflammation target would more easily explain the white matter ‘non-specific binding’ observed with these probes, or their ‘affinity for myelin’ (Stankoff et al., 2011), the 30 percent positive PET scans of controls subjects, the ‘on and off’ PET scan profile of MCIs, the negative PET scans in cases of autopsy validated AD cases (Cairns et al., 2009), the lack of cortical PET positive signal in autopsy validated familial AD cases with significant cortical neuropathology deposition (Klunk et al., 2007); the elevated signal in some brain cortices, e.g., precuneus with no more amyloid accumulation than other brain areas, including the medial temporal lobe (Nelson et al., 2009) and also the modest (if any) decline in positive-"amyloid scans" in subjects under anti-amyloid therapies. If the PET-positive signal is related to estrogen or estrogen sulfotransferase, then it is not accident that PIB accumulation would match the PET results of other markers of inflammation with the same patients, as shown in Bengt Langstrom’s recent article with C-11-L-deuteriodeprenyl as a marker of astrocytosis (Santillo et al., 2011).

    The "amyloid imaging" results need critical analyses, not a monotonic amyloid explanation pathway or consensus science justification (Barrio, 2009). The in vivo target specificity of these probes ought to be addressed, not dismissed. An ample debate is needed, not to win the argument, but to arrive at the correct scientific answer. These concerns need to be addressed because a possible erroneous interpretation of positive PET imaging, as indicative of the presence of brain amyloid particularly in asymptomatic controls, is not a mere academic issue. We would all agree that unnecessarily administering current anti-amyloid therapies to these subjects could have devastating medical consequences. Our patients, our profession and our scientific commitment in search for the truth should not permit it.


    . Inflammation and Alzheimer's disease. Arch Pharm Res. 2010 Oct;33(10):1539-56. PubMed.

    . Consensus science and the peer review. Mol Imaging Biol. 2009 Sep-Oct;11(5):293. PubMed.

    . Absence of Pittsburgh compound B detection of cerebral amyloid beta in a patient with clinical, cognitive, and cerebrospinal fluid markers of Alzheimer disease: a case report. Arch Neurol. 2009 Dec;66(12):1557-62. PubMed.

    . Specific estrogen sulfotransferase (SULT1E1) substrates and molecular imaging probe candidates. Proc Natl Acad Sci U S A. 2010 Apr 6;107(14):6222-7. PubMed.

    . Post-mortem correlates of in vivo PiB-PET amyloid imaging in a typical case of Alzheimer's disease. Brain. 2008 Jun;131(Pt 6):1630-45. PubMed.

    . Early AD pathology in a [C-11]PiB-negative case: a PiB-amyloid imaging, biochemical, and immunohistochemical study. Acta Neuropathol. 2012 Mar;123(3):433-47. PubMed.

    . Ante mortem amyloid imaging and β-amyloid pathology in a case with dementia with Lewy bodies. Neurobiol Aging. 2010 Oct 18; PubMed.

    . Amyloid deposition begins in the striatum of presenilin-1 mutation carriers from two unrelated pedigrees. J Neurosci. 2007 Jun 6;27(23):6174-84. PubMed.

    . Species-dependent metabolism of the amyloid imaging agent [C-11]PIB. J Nucl Med. 2004;45(Suppl):114P.

    . Systemic distribution of steroid sulfatase and estrogen sulfotransferase in human adult and fetal tissues. J Clin Endocrinol Metab. 2002 Dec;87(12):5760-8. PubMed.

    . Alzheimer's-type neuropathology in the precuneus is not increased relative to other areas of neocortex across a range of cognitive impairment. Neurosci Lett. 2009 Feb 6;450(3):336-9. PubMed.

    . Correlation of brain amyloid with "aerobic glycolysis": A question of assumptions?. Proc Natl Acad Sci U S A. 2010 Oct 12;107(41):17459-60. PubMed.

    . Fibrillar amyloid-beta burden in cognitively normal people at 3 levels of genetic risk for Alzheimer's disease. Proc Natl Acad Sci U S A. 2009 Apr 21;106(16):6820-5. PubMed.

    . Positron emission tomography and neuropathologic estimates of fibrillar amyloid-β in a patient with Down syndrome and Alzheimer disease. Arch Neurol. 2011 Nov;68(11):1461-6. PubMed.

    . In vivo imaging of astrocytosis in Alzheimer's disease: an ¹¹C-L-deuteriodeprenyl and PIB PET study. Eur J Nucl Med Mol Imaging. 2011 Dec;38(12):2202-8. PubMed.

    . Multimodal Imaging of Alzheimer Pathophysiology in the Brain's Default Mode Network. Int J Alzheimers Dis. 2011;2011:687945. PubMed.

    . Imaging central nervous system myelin by positron emission tomography in multiple sclerosis using [methyl-¹¹C]-2-(4'-methylaminophenyl)- 6-hydroxybenzothiazole. Ann Neurol. 2011 Apr;69(4):673-80. PubMed.

    View all comments by Jorge Barrio
  5. I concur with the rebuttal by Villemagne et al. It is already clear that imaging amyloid will make a significant contribution to the understanding of this devastating disease. Even if there are multiple issues which remain to be elucidated further, the progressive use of biomarkers, such as offered with these compounds, will become paramount.

    Provided that sufficient rigor is seen in the evaluation of data, I see no conflict with "business" taking over. We should welcome it.

    View all comments by Peter Ell
  6. Dr. Barrio raises scientific questions regarding the precise relevance of PIB and related derivatives as a new generation of compounds in imaging diagnostics. His comment looks more like an article on its own, and answering each individual detail would go beyond the scope of this forum. I agree with some of the specificity concerns. However, that is all the more reason to continue investments in studies using these compounds, rather than shutting them down. The imaging field is not standardized experimentally for routine medical applications, and the variety of protocols performed by different groups create a variety of often conflicting results, such as contribution of inflammation, or estrogen and sulfo-enzymes, or cortical versus cerebellum specificity to the labeling probe. Therefore, more studies are needed, and yes, they should be performed with the highest ethics by all involved, particularly pharmaceutical companies. It is important to keep in mind, though, that imaging is to be used in conjunction with other diagnostic testing approaches rather than as exclusive and conclusive methodology. A number of amyloid angiopathies never even progress to AD, and certainly no one wants to see unnecessary administration of "preventive" therapies. Yet, some individual scientific articles reporting opposing results should not halt a field where a certain degree of consensus opinion has been established. When supported by creatively planned, detailed studies from different groups of investigators, which fully address all questions and provide solid answers, consensus science is tremendously useful, with or without politics.

    View all comments by Olivera Mitrasinovic
  7. The comment above by Dr. Barrio is certainly provocative, but it is from time to time inaccurate and incomplete, and there are several major weaknesses in the argumentation. For example, Dr. Barrio argues that a positive PIB signal may be due to inflammatory processes in the brain. This immediately raises the question of why most patients with FTD or other forms of dementia show negative PIB scans, as there is involvement of inflammation in other neurodegenerative diseases as well. Furthermore, to support the claim that PIB binding reflects a neuroinflammatory response, Dr. Barrio refers to a paper by Santillo et al. (2011) that showed overlap between PIB and carbon-11-deuteriodeprenyl (DED, a PET tracer with affinity to bind to MAO-B enzymes, which are spatially related to astrocytes) in AD patients. They do not refer to Carter et al. (2012), however, who used the same tracers and found that average DED retention was highest in MCI patients, whilst PIB uptake was highest in AD patients. The dissociation between these tracers strongly argues against the author’s statement that PIB binding reflects neuroinflammation.


    . In vivo imaging of astrocytosis in Alzheimer's disease: an ¹¹C-L-deuteriodeprenyl and PIB PET study. Eur J Nucl Med Mol Imaging. 2011 Dec;38(12):2202-8. PubMed.

    . Evidence for astrocytosis in prodromal Alzheimer disease provided by 11C-deuterium-L-deprenyl: a multitracer PET paradigm combining 11C-Pittsburgh compound B and 18F-FDG. J Nucl Med. 2012 Jan;53(1):37-46. PubMed.

    . Use of florbetapir-PET for imaging beta-amyloid pathology. JAMA. 2011 Jan 19;305(3):275-83. PubMed.

    View all comments by Rik Ossenkoppele

Make a Comment

To make a comment you must login or register.