Since 1989, four Canadian Consensus Conferences on the Diagnosis and Treatment of Dementia (CCCDTD) have been held.Reference Patterson, Gauthier and Bergman 1 - 4 Of those, the 4th CCCDTDReference Gauthier, Patterson and Chertkow 3 focused on updating previous diagnostic approaches to Alzheimer’s disease (AD) taking into account revised diagnostic criteria proposed by the International Working GroupReference Dubois, Feldman and Jacova 5 , Reference Dubois, Feldman and Jacova 6 and the recommendations made by the National Institute on Aging - Alzheimer Association (AA) workgroups.Reference McKhann, Knopman and Chertkow 7 - Reference Sperling, Aisen and Beckett 9 The focus of CCCDTD4 and of two accompanying papersReference Burhan, Bartha and Bocti 10 , Reference Soucy, Bartha and Bocti 11 was largely on neuroimaging and other biomarkers; nine recommendations on amyloid positron emission tomography (PET) were made, most of which are now outdated. In this paper, we aimed to provide updated Canadian guidelines to dementia care practitioners on proper use of amyloid imaging as formulated by the Specialized Task force on Amyloid imaging in Canada (STAC). We also wished to generate momentum for the industry to submit a new drug proposal to Health Canada so that this revolutionary technique could become part of our clinical armamentarium under precise and well-defined indications.
Amyloid Imaging: From 2004 to Today
Since Klunk’s publication on Pittsburgh compound B (PiB) in 2004,Reference Klunk, Engler and Nordberg 12 PET using amyloid ligands has revolutionized AD research, leading to improved models of disease pathogenesis, providing evidence for a long preclinical disease phase, and stimulating therapeutic trials aimed at delaying or preventing the symptomatic phase of AD.Reference Jack, Wiste and Weigand 13 , Reference Sperling, Rentz and Johnson 14 For example, amyloid imaging has served as a secondary outcome measure in AD clinical trials with disease-modifying agents such as the antiamyloid monoclonal antibodies bapineuzumab and solanezumab.Reference Farlow, Arnold and van Dyck 15 , Reference Rinne, Brooks and Rossor 16 Brain amyloid reduction and slowing of cognitive decline were found after 1 year of treatment with aducanumab, a human immunoglobulin G1 monoclonal antibody against a conformational epitope found on amyloid beta (Aβ).Reference Keller 17 Finally, its added value also lies in subject selection for clinical trials given an approximate 15% amyloid negative rate.
Beyond research, amyloid imaging has demonstrated great potential as a diagnostic tool largely because it allows in vivo detection of amyloid plaques, a core pathologic feature of AD.Reference Klunk, Engler and Nordberg 12 It is now an established technique in the field of neuroimaging of aging and dementia, with data incorporated in the most recent consensus guidelines for the diagnosis of ADReference McKhann, Knopman and Chertkow 7 and predementia AD-related conditions.Reference Albert, DeKosky and Dickson 8 , Reference Sperling, Aisen and Beckett 9 There currently are three US Food and Drug Administration and European Union–approved, fluorine-18–labeled, tracers available for clinical use: florbetapir since 2012,Reference Clark, Pontecorvo and Beach 18 flutemetamol since 2013,Reference Wolk, Grachev and Buckley 19 and florbetaben since 2014.Reference Rowe, Ackerman and Browne 20 Converging evidence on the diagnostic utility of that technique has rapidly accumulated.Reference Ossenkoppele, Prins and Pijnenburg 21 - 29 Still, no tracer has been approved for clinical use in Canada so far.
Amyloid PET in Cognitively Normal Elderly Individuals and Various Clinical Populations
Cognitively Normal Elderly Individuals
Cognitively normal elderly individuals (CNs) show elevated PiB binding in 10% to 34% of cases, a proportion similar to observed rates of amyloid pathology in autopsy studies.Reference Pike, Savage and Villemagne 30 Increasing age and the presence of the apolipoprotein E ε4 allele are the major predictors of PiB positivity in CN.Reference Rowe, Ellis and Rimajova 31 , Reference Jack, Wiste and Weigand 32 Recent findings in persons without dementia or mild cognitive impairmentReference Albert, DeKosky and Dickson 8 , Reference Petersen 33 , Reference Petersen 34 suggest that amyloid deposition is associated with very subtle cognitive deficits, especially among apolipoprotein E ε4 carriers.Reference Kantarci, Lowe and Przybelski 35 Interestingly, recent data suggest that the cooccurrence of Aβ and neurodegeneration (hippocampus volume and glucose metabolism) is associated with cognitive decline in CNs.Reference Mormino, Betensky and Hedden 36
The significance of a positive amyloid scan in CNs still remains uncertain, but cross-sectional studies have shown “AD-like” brain changes (hippocampal and temporo-parietal atrophy),Reference Dickerson, Bakkour and Salat 37 whereas early longitudinal data have strengthened the notion that many (although probably not all) are in a “preclinical” phase of AD.Reference Sperling, Aisen and Beckett 9 Recently, a group of researchers compared the ability of molecular biomarkers for AD, including amyloid imaging and cerebrospinal fluid (CSF) biomarkers (Aβ1-42, tau, ptau181, tau/Aβ1-42, ptau181/Aβ1-42), to predict time to incident cognitive impairment among cognitively normal adults aged 45 to 88 years and followed for up to 7.5 years. Results indicated that all AD biomarkers studied predicted incident cognitive impairment and supported the hypothesis that biomarkers signal underlying AD pathology at least several years before the appearance of dementia symptoms.Reference Roe, Fagan and Grant 38
From a diagnostic perspective, the significant number of amyloid-positive CNs emphasizes that amyloid positivity is not synonymous with AD, and that amyloid scans cannot replace a detailed clinical evaluation. At present, there is no clinical indication for amyloid imaging in CNs, though this will remain an area of active research in coming years, particularly with the advent of amyloid-lowering therapies, which might be most effective if initiated in the presymptomatic disease stage.Reference Klunk 39 , Reference Ostrowitzki, Deptula and Thurfjell 40
Mild Cognitive Impairment
Current data in mild cognitive impairment (MCI)Reference Albert, DeKosky and Dickson 8 , Reference Petersen 33 , Reference Petersen 34 indicates that amyloid imaging provides prognostic information, presumably by identifying patients with underlying AD pathology.Reference Pontecorvo and Mintun 41 As a group, 52% to 87% of MCI subjects show elevated PiB binding in a regional distribution similar to that observed with AD.Reference Pike, Savage and Villemagne 30 In longitudinal studies, 1-year conversion rates to AD range from 33% to 47% in PiB-positive MCI subjects, whereas virtually no conversions are seen in PiB-negative subjects.Reference Wolk, Price and Saxton 42 In a longitudinal study,Reference Okello, Koivunen and Edison 43 the authors compared baseline amyloid deposition between MCI converters and nonconverters in 31 subjects followed over 3 years. The conversion rate was 82% in those with increased PiB uptake, but only 7% in PiB-negative subjects. Results from the Australian Imaging Biomarkers and Lifestyle study of aging on 87 participants with MCI (age, 73.7±8.27) showed that 59% had progressed to probable AD over 3 years.Reference Ellis, Lim and Harrington 44 Multivariate analysis showed β-amyloid imaging as the variable most strongly associated with progression. Almost all amnestic MCI subjects with a positive amyloid scan developed AD. Altogether, the literature clearly suggests that PiB-positive amnestic MCI patients are likely to have early AD, and amyloid imaging will help in risk stratification and selection of patients who may benefit from experimental disease-specific therapies. Similar to normal aging, positivity of both Aβ and neurodegeneration biomarkers in MCI can further stratify risk of imminent conversion to dementia.
Alzheimer’s Disease
Most studies in ADReference Dubois, Feldman and Jacova 6 , Reference McKhann, Knopman and Chertkow 7 have found very high (90% or greater) PiB-PET positivity, with a pattern that closely mirrors the distribution of plaques found at autopsy.Reference Rowe, Ellis and Rimajova 31 Tracer binding is diffuse and symmetric, with high uptake consistently found in prefrontal cortex, precuneus and posterior cingulate cortex, followed closely by lateral parietal and lateral temporal cortex, and striatum. Studies in atypical clinical presentations of AD have shown that amyloid deposition is more common in the logopenic variant of primary progressive aphasia than in nonfluent or semantic variants,Reference Rabinovici, Jagust and Furst 45 supporting the hypothesis that logopenic variant of primary progressive aphasia is often associated with underlying AD. Others have detected high PiB binding in patients with posterior cortical atrophy, a visuospatial/biparietal clinical syndrome often caused by AD.Reference de Souza, Corlier and Habert 46 Much like Reference Clark, Pontecorvo and Beach 18 F-fluorodeoxyglucose-PET (FDG-PET),Reference Laforce, Buteau, Paquet, Verret, Houde and Bouchard 47 amyloid imaging will probably not add value to the diagnostic workup of patients with straightforward clinical AD, but is likely to be useful in patients with atypical complex presentations or early age-of-onset dementia (see the following two sections).
The Frontotemporal Lobar Degeneration Spectrum of Disorders
Considering that the frontotemporal lobar degeneration spectrum of disorders (FTLD) and AD are the leading causes of early age-of-onset dementia,Reference Ratnavalli, Brayne, Dawson and Hodges 48 that distinguishing the two during life can be clinically challenging but is also important,Reference Alladi, Xuereb and Bak 49 and that Aβ plaques are not part of the FTLD pathologic spectrum, several authors have argued for a valuable role of amyloid imaging in the differential diagnosis of these conditions. Small case series have reported low rates of PiB positivity (0% to 15%) and florbetaben positivity (9%) in FTLD.Reference Villemagne, Ong and Mulligan 50 Recently, results from the largest study currently available on the diagnostic utility of amyloid PET in FTLD showed, in 62 AD and 45 FTLD patients, that PiB visual reads had a higher sensitivity for AD than FDG-PET, with similar specificity.Reference Rabinovici, Rosen and Alkalay 23 PiB outperformed FDG in classifying patients with known histopathology, and visual reads showed higher inter-rater reliability and agreement than for FDG, suggesting it was the more accurate and precise technique.
Complex, Atypical Patients With an Uncertain Diagnosis
The clinical diagnosis of AD has only moderate sensitivity and specificity when compared with the pathological cause of dementia as determined at autopsy.Reference Beach, Monsell, Phillips and Kukull 51 Misdiagnosis rates are even higher in complex, atypical patients with an uncertain diagnosis, approaching 30%.Reference Grundman, Pontecorvo and Salloway 26 A growing body of literature supports the clinical utility of amyloid imaging for the differential diagnosis of atypical patients with an uncertain diagnosis.Reference Ossenkoppele, Prins and Pijnenburg 21 , Reference Sanchez-Juan, Ghosh and Hagen 22 , Reference Wolk, Price and Madeira 24 - 29 , Reference Ossenkoppele, Jansen and Rabinovici 52 This has major implications for a cohort of individuals who are often younger than 65 years of age and still active in the workforce. Indeed, several dementia experts have argued that an accurate diagnosis helps direct therapy (i.e. avoid unnecessary or undesired cholinesterase inhibitors or memantine prescriptions), determine a better care plan (which considers patient safety and minimizes the risk of preventable complications), and enable patients to participate in legal and financial planning.
Recently, a group of Canadians researchers investigated the clinical utility of amyloid PET in the differential diagnosis of atypical cases and its impact on caregivers in the context of a tertiary memory clinic. 29 Using the amyloid tracer Reference Clark, Pontecorvo and Beach 18 F-NAV4694,Reference Cselényi, Jönhagen and Forsberg 53 - Reference Rowe, Pejoska and Mulligan 55 they prospectively scanned 25 patients (mean age, 59.3 years; standard deviation [SD], 5.8; and mean Mini-Mental State Examination, 21.6; SD 6.2) with an atypical syndrome as determined by dementia experts. All patients had a full workup (i.e. history, examination, blood tests, neuropsychology, magnetic resonance imaging [MRI], and FDG-PET), yet no certain diagnosis could be arrived at following that investigation. Amyloid PET was either positive or negative based on qualitative and quantitative reads by two qualified independent expert nuclear medicine specialists. Physicians rated whether amyloid PET was associated with a change in diagnosis and altered management. They also reported their degree of confidence in diagnosis before and after amyloid PET. Caregivers were met 3 months after having been told of the diagnosis and completed a 21-item Likert scale questionnaire along with a 1-hour interview designed to assess the impact of the amyloid scan. The cohort was 48% amyloid positive and 52% amyloid negative. Inter-rater reliability was 100%. Amyloid PET was associated with a diagnostic change in 36% (9/25) of cases (24% changed from AD to non-AD and 12% from non-AD to AD). There was a significant increase (40%) in diagnostic confidence following the scan. Altogether, this study corroborated recent findings and suggested an additive role for amyloid PET in atypical cases with an unclear diagnosis despite the detailed workup of a tertiary memory clinic. Amyloid PET increased diagnostic confidence and generated significant alterations in management in almost three-quarters of cases. Furthermore, the overall process was very well received by caregivers, reducing anxiety and depressive symptomatology as well as increasing quality time spent with their loved ones.
Other Clinical Conditions
Other clinical conditions studied with amyloid PET include vascular cognitive impairment, cerebral amyloid angiopathy, Parkinson’s disease dementia, and dementia with Lewy bodies (DLB). In one study on vascular cognitive impairment, authors found that 69% of patients were PiB-negative.Reference Lee, Kim and Kim 56 High PiB binding rates were found in nondemented patients with cerebral amyloid angiopathy.Reference Johnson, Gregas and Becker 57 Most studies showed higher amyloid plaques in DLB than in Parkinson’s disease dementia or nondemented PD patients and, in some, PiB positivity was associated with more rapid disease progression.Reference Maetzler, Liepelt and Reimold 58 The high frequency of plaques and high rates of positive scans in DLB suggest that amyloid PET is unlikely to be helpful in differentiating DLB from AD.
Amyloid PET in Clinical Practice Today: Are We Ready Yet?
The translation from the research setting into the clinic has progressed steadily.Reference Johnson, Minoshima and Bohnen 59 - Reference Johnson, Minoshima and Bohnen 61 For example, a group recently determined the sensitivity and specificity of amyloid PET with flutemetamol using neuropathologically determined neuritic plaque levels and showed high sensitivity and specificity in an end-of-life population.Reference Clark, Pontecorvo and Beach 18 , Reference Curtis, Gamez and Singh 62 , Reference Sabri, Sabbagh and Seibyl 63 In an effort to guide clinicians, the Society of Nuclear Medicine and Molecular Imaging (SNMMI) and the AA jointly published criteria for the appropriate use of amyloid PET.Reference Johnson, Minoshima and Bohnen 59 , Reference Johnson, Minoshima and Bohnen 61 Over the past 3 years, almost 10 reports on the practical clinical applications of amyloid imaging have been published, and their conclusions indicated a significant role in orienting treatment (i.e. deciding whether to initiate or discontinue AD symptomatic medications) and a positive impact on caregivers.Reference Ossenkoppele, Prins and Pijnenburg 21 , Reference Sanchez-Juan, Ghosh and Hagen 22 , Reference Wolk, Price and Madeira 24 - 29 , Reference Ossenkoppele, Jansen and Rabinovici 52 Finally, much is expected from the soon-to-be launched Imaging Dementia – Evidence for Amyloid Scanning (IDEAS) Study, a $100 million open-label longitudinal cohort effort on approximately 18,500 US Medicare beneficiaries. In this venture, diagnostically uncertain cases of MCI and atypical dementia will be scanned to determine whether knowledge of amyloid status leads to significant changes in patient management and if this translates into improved medical outcomes.
In light of the outstanding worldwide momentum surrounding the utility of amyloid imaging in clinical practice, and the fact that Canada lags behind on approval of the technique by health authorities, we first aimed to provide updated Canadian guidelines to dementia care practitioners on proper use of amyloid imaging. Second, we wished to generate momentum for the industry to submit a new drug proposal to Health Canada.
Methods
The STAC met in Montreal, QC, Canada on May 14, 2015, to update the 2012 Canadian consensus guidelines,Reference Gauthier, Patterson and Chertkow 3 which incorporated nine recommendations on amyloid PET, most of which are now outdated. The meeting included all members of the STAC (a group of local, national, and international dementia experts and imaging specialists; see Appendix A) as well as consensus meeting participants (clinical and academic; see Appendix A), and leading representatives from the molecular imaging and imaging software industries. Technical and regulatory considerations pertaining to Canada were discussed with a member of Health Canada who attended the meeting. Peer-reviewed and published literature between January 2004 and May 2015 was searched before the meeting. A survey of current diagnostic practices in Canadian dementia centers was also presented to allow focused discussions on Canadian medical practices. In brief, we discussed indications and, just as importantly nonindications, of amyloid imaging based on clinical and nonclinical scenarios with variables including symptoms (typical and atypical), clinical settings, clinical contexts, evidence of cognitive deficits, family history, knowledge of AD genetic risk, and age.
Although previous CCCDTDs had used the evidence grading system developed by the Canadian Task Force on Preventive Health Care, for this iteration we attempted to follow, where possible, the Grading of Recommendations Assessment, Development and Evaluation system in keeping with current recommendations for the conduct of consensus conferences.Reference Guyatt, Oxman and Vist 64 Each participant was allowed to take part in the discussion. Consensus was defined as 80% or more conference participants being in agreement with a recommendation. Partial consensus was defined as 60% to 79% being in agreement. Recommendations reaching consensus are listed in Tables 1 and 2. Recommendations reaching only partial or no consensus are only mentioned in the Discussion section.
Recommendations for clinicians on behalf of the Canadian Consensus Conference on the Use of Amyloid Imaging
AD=Alzheimer’s disease; MCI=mild cognitive impairment; MRI=magnetic resonance imaging; PET=positron emission tomography.
* Subjects with diagnostic uncertainty after a comprehensive tertiary memory clinic evaluation are also named “complex/atypical cases” and often include AD variants, non-AD dementias (e.g. frontotemporal lobar degeneration), nonprogressing AD, or patients with comorbid and nondegenerative conditions (e.g. depression, substance abuse, atypical bipolar disorder) (see Laforce & RabinoviciReference Laforce and Rabinovici 66 for a detailed discussion).
† In accord with Canadian Consensus Conferences on the Diagnosis and Treatment of Dementia 4,Reference Gauthier, Patterson and Chertkow 3 structural brain imaging using MRI means a “Head MRI – Dementia Protocol,” which includes: (1) coronal T2/fluid-attenuated inversion recovery and (2) axial susceptibility-weighted imaging, diffusion-weighted imaging, as well as three-dimensional T1 sequences. Structural brain imaging using MRI should always be performed before amyloid imaging as a positive scan in typical AD distribution in a patient with dementia can be secondary to cerebral amyloid angiopathy.Reference Ducharme, Guiot, Nikelski and Chertkow 67
‡ Dementia expert: a physician with substantial clinical experience and practice in dementia care. Expertise in dementia is acquired through formal training and clinical experience in neurology, psychiatry, and geriatric medicine; however, not all dementia experts have expertise in amyloid imaging and/or work in conjunction with nuclear medicine specialists qualified in amyloid imaging, hence the recommendation to refer to a dementia center with expertise in this technique.
§ Safety issues and the notion of major consequences: this is determined on a case-by-case basis with all significant clinical information. An example of this could be an emergency medical technician, paramedic, or a lead squad firefighter, etc.
Recommendations for research and translational development to clinical care on behalf of the Canadian Consensus Conference on Use of Amyloid Imaging
AD=Alzheimer’s disease.
Results
A set of updated guidelines were agreed on that define the types of patients and clinical circumstances in which amyloid PET could be used in Canada (Table 1). Future research directions were also outlined, notably the importance of studies that would assess the pharmaco-economics of such diagnostic procedure (Table 2).
Survey of the Availability and Use of Biomarkers in Canada
Three months before the meeting, a survey was sent to all Canadian dementia centers. A total of 27 respondents (British Columbia, 3; Alberta, 3; Saskatchewan, 1; Ontario, 7; Quebec, 10; Maritimes, 3) provided answers mainly to two questions: (1) How many early-onset atypical dementia cases do you see per month, and (2) Which advanced diagnostic techniques do you use in practice beyond clinical history, physical examination, standard laboratory tests, and basic computed tomography imaging (i.e. MRI, hippocampal volumetry, molecular imaging, CSF Aβ1-42 and tau). Results indicated that an average of 5,8 (SD: 5,4) early-onset atypical dementia cases per clinician were seen per month. More than 80% use MRI. Only clinicians from Quebec use FDG-PET in clinical practice, whereas single-photon emission computed tomography is used by a majority of clinicians outside Quebec, because of regionally specific provincial reimbursement issues. Only 20% of clinicians use CSF measures. Less than 11% prescribe acetylcholine-esterase inhibitors to atypical cases with uncertain diagnoses. Finally, in 85% of cases, clinicians reported that knowledge of the amyloid status of their atypical patient would change their therapeutic approach. Some clinicians also reported sending patients to the United States for an amyloid scan.
Discussion
Amyloid PET is now an established neuroimaging technique with data incorporated in the consensus guidelines on AD and predementia AD-related conditions. So far, three different fluorine-18–labeled agents have been approved for clinical use in a variety of jurisdictions around the world. Despite these major advancements, Canada is not yet one of those jurisdictions. Publication of appropriate use criteria by the AA and the SNMMI has paved the way for other countries to adopt a standardized model reinforcing proper use of amyloid imaging in medical practice. This paper is derived from discussions of the STAC, a group of local, national and international dementia experts and imaging specialists who revisited the scenarios in which amyloid PET could be used appropriately in Canada. The final product is an updated set of guidelines to the 2012 CCCDTD4 effort,Reference Gauthier, Patterson and Chertkow 3 which also factors in the results of a survey of current practices in Canadian dementia centers. Furthermore, it is tailored to the Canadian reality and wishes to promote future development of amyloid imaging in our country. We hope this paper will generate momentum for the industry to submit a new drug proposal to Health Canada so that regulatory bodies approve the technique and discussions about provincial reimbursement can begin.
Canadian Challenges
This consensus effort allowed members to realize that only two Canadian provinces (Alberta and Quebec) have access to neurological FDG-PET examinations, a well-established molecular imaging technique in the field of dementia. Indeed, use of FDG-PET has been supported by a wealth of literature (see Bohnen et alReference Bohnen, Djang, Herholz, Anzai and Minoshima 68 for a review). In the diagnosis of AD, authors have showed that FDG-PET is superior to a baseline clinical evaluation and similar to an evaluation performed 4 years later.Reference Jagust, Reed, Mungas, Ellis and Decarli 69 The addition of FDG-PET to the investigation of atypical/unclear cases of dementia helps generate a more accurate diagnosis and initiate earlier treatment.Reference Laforce, Buteau, Paquet, Verret, Houde and Bouchard 47 Several studies have indicated that it is a cost-effective technique in the differential diagnosis of dementiaReference Mosconi 70 - Reference Silverman, Gambhir and Huang 73 ; yet, it appears that the vast majority of Canadian dementia centers do not take advantage of this technique.
One could argue that amyloid imaging, when approved by regulatory bodies, will be in no different situation than FDG-PET because its use will also depend on provincial reimbursement policies. In 2013, our American colleagues faced a similar situation with regards to reimbursement for amyloid imaging. The Centers for Medicare and Medicaid Services (CMS) National Coverage Decision on amyloid PET imaging in dementia and neurodegenerative disease (CAG-00431N) ruled not to cover the scans because “the evidence is insufficient to conclude that the use of amyloid PET imaging is reasonable and necessary for the diagnosis or treatment of illness or injury or to improve the functioning of Medicare beneficiaries with dementia or neurodegenerative disease.” CMS questioned the ability of amyloid PET to lead to improved health outcomes, such as avoidance of futile treatment or tests, improving or slowing the decline of quality of life, and survival. However, CMS did find sufficient evidence that the use of amyloid PET is promising: (1) to exclude AD in narrowly defined and clinically difficult diagnoses and (2) to enrich clinical trials seeking better treatments or prevention strategies for AD. The soon-to-be launched IDEAS study (discussed previously) was developed partly in response to this decision and it is hypothesized that it will demonstrate that knowledge of amyloid status leads to significant changes in patient management and improved medical outcomes, particularly in diagnostically uncertain cases of MCI and atypical dementia.
We believe time has come for Canadian dementia experts to take leadership in defining the role of molecular imaging in the differential diagnosis of dementia. Reimbursement issues surrounding the two PET techniques currently available should be discussed with appropriate provincial health authorities. Both techniques (FDG-PET and amyloid imaging) are now supported by a solid body of evidence, and politicians properly informed of their benefits should be open to fund programs where FDG-PET and amyloid PET are reimbursed in specific clinical situations along the guidelines detailed previously (e.g. differential diagnosis of AD vs FTLD).
Amyloid Imaging versus CSF
Amyloid imaging is not the sole biomarker which may signal underlying AD pathology. A decade or more before the appearance of dementia symptoms, and possibly before amyloid accumulation is detectable by PET, CSF changes can appear and correlate with brain atrophy in cognitively normal elderly.Reference Jack, Wiste and Vemuri 74 - Reference Fagan, Head and Shah 76 As suggested in the CCCDTD4 paper,Reference Gauthier, Patterson and Chertkow 3 authors have compared amyloid imaging with CSF AD biomarkers in the same study and found that CSF Aβ1-42 analyzed consecutively in routine clinical practice at an accredited laboratory can be used with high accuracy to determine whether a patient has normal or increased cortical Aβ deposition and so can be valuable for the early diagnosis of AD.Reference Palmqvist, Zetterberg and Blennow 77 Other groups replicated these findings using cross-sectional and longitudinal designs.Reference Landau, Lu and Joshi 78 Interestingly, abnormal flutemetamol retention levels correlated with disease stage in patients with mild cognitive symptoms, but this was not the case for CSF Aβ1-42.Reference Palmqvist, Zetterberg and Blennow 77
The utility of CSF in diagnosing unclear dementing syndromes such as those on the FTLD spectrum is less clear.Reference Hampel, Bürger, Teipel, Bokde, Zetterberg and Blennow 79 - Reference Rosa-Neto, Hsiung and Masellis 81 Although tau fibrils and aggregates are pathological hallmarks of several FLTD subtypes, total CSF tau (t-tau) appears to be a general marker of neurodegeneration, whereas phosphorylated tau (i.e. p-tau-231 and p-tau-181) are useful in discriminating between AD and frontotemporal dementia.Reference Hampel and Teipel 82 Increased ratio of tau/Aβ1-42 can also distinguish AD from FTLD,Reference Bian, Van Swieten and Leight 83 - Reference Kapaki, Paraskevas and Papageorgiou 85 and a low CSF p/t-tau ratio may distinguish FTLD-TDP from FTLD-tau.Reference Hu, Watts and Grossman 86 Another study suggests that total and p-tau in CSF were elevated in primary progressive aphasia relative to the behavioral variant frontotemporal dementia.Reference Bibl, Mollenhauer and Lewczuk 87
At the moment, however, CSF variability across techniques and centers is such that it limits proper confident interpretation of the results. Because of the absence of appropriate laboratory infrastructure in Canada, or consensus as to where the samples should be sent for analysis, Aβ1-42, t-tau, and p-181-tau have no clinical utility in Canada (not recommended for clinical practice), although they are part of research protocols in observational and therapeutic studies. Current international efforts to standardize CSF AD biomarkers, 88 notably with more reliable enzyme-linked immunosorbent assay techniques, are currently under way.
Action Plan
An action plan was developed modeled on appropriate use criteria by Johnson and colleagues.Reference Johnson, Minoshima and Bohnen 59 , Reference Johnson, Minoshima and Bohnen 61 These recommendations cover issues that could be disseminated to Canadian health care professionals and dementia organizations through knowledge translation activities: (1) who should be referred for an amyloid scan; (2) education (patients and families, health care professionals, dementia organizations); (3) amyloid PET scanning technique, interpretation (visual vs quantitative), translation into a clinical decision; and (4) proper disclosure of results.
Who Should Be Referred for an Amyloid Scan?
This decision should rely on the dementia expert guided by current Canadian guidelines. A dementia expert is a physician with substantial clinical experience and practice in dementia care. Expertise in dementia is acquired through formal training and clinical experience in neurology, psychiatry, and geriatric medicine; however, not all dementia experts have expertise in amyloid imaging and/or work in conjunction with a nuclear medicine specialist (NMS) qualified in amyloid imaging, hence the recommendation to refer to a dementia center with expertise in this technique, when appropriate.
Education (Patients, Families, Health Care Professionals, Dementia Organizations)
Dementia specialists in Canada are committed to disseminating information to the public and dementia organizations (e.g. information on amyloid imaging already exists on the Alzheimer Society’s website at http://www.alzheimer.ca), to assisting colleagues in appropriate use of amyloid imaging, and in providing clarifications on how to incorporate amyloid PET in medical practice. Knowledge translation activities should be organized in all Canadian provinces to introduce the most recent guidelines on amyloid imaging.
Amyloid PET Scanning, Interpretation, and Translation Into a Clinical Decision
Amyloid PET Scanning
Imaging procedures should be performed by qualified nuclear medicine technologists and NMS with national certification in nuclear medicine and appropriate qualification in amyloid imaging. It should be performed in an imaging facility certified by Canadian accrediting agencies. Procedure guidelines for amyloid PET (SNMMI and European Association of Nuclear Medicine) should be followed.
Interpretation
The nuclear medicine team (i.e. the technologist and the NMS) performing the scan must be familiar with brain anatomy and must have adequate specific training in amyloid PET interpretation because amyloid PET imaging can be technically challenging and should be performed with strict attention to quality control.Reference Johnson, Minoshima and Bohnen 59 , Reference Johnson, Minoshima and Bohnen 61 Web-based instruction programs have been developed and validated and should be completed successfully by all amyloid imaging teams before reading scans. Interpretation of amyloid PET imaging should be communicated to the referring physician by the NMS by way of a written report according to a standard diagnostic imaging practice as outlined in the SNMMI General Imaging Guideline. At this time, the final reading should indicate whether Aβ was present (amyloid positive) or was not present (amyloid negative). The protocol for the qualitative read that determines positivity or negativity must be standardizedReference Landau, Mintun and Joshi 89 and must conform to a specific guideline provided by the manufacturer. Indeterminate results may arise as a result of technical or physiological factors and should be reported as such.
Translation Into a Clinical Decision
The NMS report should not equate amyloid positivity with AD dementia (amyloid positivity is not synonymous with AD, and amyloid scans cannot replace a detailed clinical evaluation). Upon receiving the NMS report, the dementia expert should proceed to a comprehensive review of the clinical assessment (history, physical examination) and test results (laboratories, neurocognitive testing, MRI, FDG-PET) and incorporate amyloid results into a clinical decision process, always considering that amyloid imaging is an evolving modality and that image interpretation criteria, clinical significance of positive and/or negative scans, and technical imaging considerations are evolving.Reference Ossenkoppele, Jansen and Rabinovici 52
Proper Disclosure of Results
That moment can be highly stressful for patients and families. To maximize safety and effectiveness of disclosing results, we recommend consulting the sequence recently developed by Harkins et alReference Harkins, Sankar and Sperling 65 in cognitively normal older adults participating in AD prevention studies. This process included: (1) an educational session, in which participants receive verbal and written information covering what is known and unknown about amyloid imaging, including possible results and their meaning; (2) screening for anxiety and depression to determine suitability to receive amyloid imaging information; (3) checking comprehension and recognizing distress; (4) conducting imaging on a separate day from consent, and disclosing results on a separate day from imaging; 5) proceeding to disclosure in person, with time for questions (at disclosure, physicians should assess mood and willingness to receive results); and 6) offering resources for support (brochures, follow-up call). The latter were developed for normal older adults participating in AD prevention studies. This may not be entirely possible in clinical practice, but can serve as general principles to guide proper disclosure.
Future Challenges
Amyloid imaging has been approved for widespread clinical use by leading health authorities in the United States and the United Kingdom. Despite this, several unknowns about its diagnostic utility remain and future studies should particularly focus on (1) its sensitivity and specificity as compared with pathology in practice-based settings (as opposed to the hospice studies), (2) technical and patient factors that could lead to false positives and false negatives, (3) the relative contribution of both diffuse and neuritic plaques’ binding to the in vivo signal, (4) interpretation of the test as a dichotomous result versus assessing binding intensity and spatial distribution, (5) inter- and intra-rater reliability of visual and/or quantitative interpretations, (6) the optimal quantitative threshold for defining a positive scan for each of the available tracers, (7) whether the threshold for PiB positivity should be adjusted based on demographic factors such as age or genetic variables, and (8) cost effectiveness issues. Such issues are relevant for any diagnostic test, and should be addressed as research continues to target key variables associated with amyloid imaging. Recent longitudinal initiatives such as the IDEAS study should help answer several of these questions.
Conclusions
Amyloid imaging represents a major breakthrough in the evaluation of dementia that will doubtlessly translate into better clinical care and ultimately help guide the development of molecular-based therapies for these devastating illnesses. An impressive body of research has already been generated in the field, and studies of practical clinical applications are emerging with a specific indication in patients with objectively confirmed cognitive impairments where diagnostic uncertainty remains even after a comprehensive clinical evaluation in a tertiary memory clinic. This technique should always remain an adjunct imaging tool that is part of a comprehensive clinical evaluation in which a dementia expert determines that having a more accurate clinical diagnosis will alter management. Fundamentally, amyloid imaging detects a brain histological state, and is not a clinical diagnosis. Used in isolation, it cannot diagnose AD, MCI, or differentiate normal from abnormal aging. Clinical availability of new tracers in Canada would represent a major advancement for the many Canadians affected by an unclear dementing condition who suffer in silence while being exposed to unnecessarily prolonged delays before diagnosis, repeated and pointless visits and diagnostic tests with inferior sensitivity and specificity than amyloid imaging, and inappropriate treatments or lack thereof when indicated. In the end, we wish this effort to generate momentum for the industry to submit a new drug proposal to Health Canada so that regulatory bodies approve the technique and approval of provincial reimbursement can follow its proper course.
Acknowledgements and Funding
Sincere thanks to the staff of the McGill Center for Studies in Aging who helped during the meeting in Montreal. We also thank the Molson Foundation and the Molson Lecture Series (http://aging.mcgill.ca/talks.php) for their support and dissemination of knowledge. Financial support for the meeting was obtained from Le réseau des cliniques mémoire du Québec and the Ethical, Legal, Social Impact committee of the Canadian Consortium on Neurodegeneration in Aging funded by the Canadian Institutes of Health Research. Finally, sincere thanks to Société Alzheimer de Québec, Fondation du CHU de Québec, and Fonds sur la maladie d’Alzheimer de l’Université Laval.
Disclosures
PR-N is an independent investigator and has a research Contract with Navidea Biopharmaceuticals. J-PS is a speaker and has received a speaker fee from Novartis. GR has been an invited speaker and received honoraria from GE Healthcare and Piramal Imaging; been a principal investigator/independent investigator and received research support from Avid Radiopharmaceuticals; and been a principal investigator and received research support from the Alzheimer’s Association, NIH, Tau Consortium, and French Foundation. BD has been a consultant and received consulting fees from Eli Lilly. SG has been an advisor and received consulting fees from Lilly, TauRx, Ever Pharma, and Schwabe; and been a speaker and received speaker fees from Lundbeck and Schwabe. RL has nothing to disclose.
Appendix A: Members
A. Specialized Task Force on Amyloid imaging in Canada (STAC)
1. Organizing Committee Members
Serge Gauthier, MD (chair)
Robert Laforce Jr, MD, PhD
Pedro Rosa-Neto, MD (co-chair)
Jean-Paul Soucy, MD
2. Invited International Leaders
Bruno Dubois, MD
Gil D. Rabinovici, MD
B. Consensus meeting participants
Clinical and Academic
Howard Chertkow, MD
Guy Lacombe, MD
Alain Robillard, MD
Sylvia Villeneuve, PhD
C. Member from Canadian Regulatory Bodies (Health Canada)
Agnes Klein, MD
D. Leading representatives from the molecular imaging and imaging software industries
Susan De Santi, PhD
Benoît Galarneau
Rick Hiatt
René Rebeaud
Dwight Tapp
Appendix B: Relationships With Industry and Management of Conflicts of Interest
The group rigorously attempted to avoid any actual, perceived, or potential conflicts of interest that might have arisen as a result of an outside relationship or personal interest of the writing committee members. We reviewed our own industry relationship policies to ensure that the ensuing process adhered to current standards. Members were required to provide disclosure statements of all relationships that might be perceived as real or potential conflicts of interest. These statements were reviewed by the chair and senior author of this paper.
