European Alzheimer Disease Consortium (EADC) – Neuroimaging study group

FDG-PET data sharing in patients with MCI and normal controls:

a spontaneous study.

FDG-PET is a recognized biomarker for the early diagnosis of AD before dementia.

The hypometabolic pattern is focussed around the posterior cingulate cortex; then, several other cortical regions have been highlighted to compose the AD pattern, with discrepancies among studies. The main candidates are the parietal precuneus, the superior and inferior parietal lobules, the medial temporal lobe (MTL), the lateral frontal cortex, the orbitofrontal cortex.

Such discrepancies inevitably derive from the limited number of patients in each research, reaching a maximum value of 50-60 patients in the few multicentre studies. MCI-converters in these more numerous studies are just about 15.

The proposal could overcome this main limitation of FDG-PET studies in MCI by joining data from several centres in the frame of the EADC, an organisation that guarantees high-standard diagnostic quality. The issues of different equipments can be overcome by the participation of a group of healthy controls together with the patients groups in each centre. Moreover, each centre can be taken into account as a possible confounding variable both in voxel-based analysis (SPM) and in other VROI-based, multivariate analysis (such as PCA).

The idea is to converge toward a common data elaboration centre for statistical analysis.

The operative proposal includes some steps:

  1. Centres where a NM Unit-physician who is in tight contact with a Dementia Unit and regularly uses FDG-PET.
  2. MCI patients must have a baseline neuropsychological assessment including standard tests for -at least- verbal memory (with delayed recall), visuoconstruction, language, executive functions. Scores must be given both as raw scores and as Z scores on the published normative (preferred in the original language), corrected for age and education. The MMSE is mandatory. Patients must also have standard questionnaires excluding dementia, such as ADL, IADL, CDR. Other scales, such as depression scales and NPI, are desiderable.
  3. MCI patients must have at least a 1-year follow-up period, but longer periods are desiderable. At least at 1-year interval, they should be re-visited and at least MMSE, CDR, ADL and IADL must be available. A every-year neuropsychological assessment is highly desiderable because it allows the study of MCI decliners and non-decliners
  4. Patients must be diagnosed according to current standards and classified as amnestic MCI (aMCI: single or multi-domain) and non amnestic-MCI (naMCI: single or multi-domain). The study is manily aimed to investigate the AD pattern, so that amnestic MCI is highly preferred to non amnestic MCI. These last patients may therefore be included but they should represent a marginal part of the whole sample. The follow-up time for AD converter patients is stopped when the diagnosis of dementia is made.
  5. All patients must undergo MRI (or CT, if MRI unavailable or not applicable) at the time of diagnosis. All neuro-psychiatric diseases (Parkinson, head trauma, stroke, psychosis, etc.) or systemic diseases (heapitc or renal failure, etc.) that may account for cogntive deficit are exclusion criteria. White matter hyperintensities and lacunae are not exclusion criteria.
  6. the availability of digital MRI is highly desiderable, as the availability of other markers, including CSF biormarkers, ApoE genotype, digital EEG, ect.
  7. Each centre should contribute with a minimum of 20 MCI patients and a minimum of 10 controls. Higher numbers are obviously desiderable.
  8. During the meeting in Thessaloniki, it has been proposed and accepted to add also a subgroup of patients with early-mild AD at baseline (i.e., MMSE>20; CDR=0.5-1) for the groups that have this availability.

The aim is to reach a number of about 200 MCI patients and 100 controls. Among MCI patients, at least 50-100 should convert (or already converted, if retrospective cases are considered) to AD in a couple of years.

The PET files should be converted into Analyse format for further processing. This can be easily done by means of free softwares, such as Mricro.

Myself and the elaborating centre can help the file transformation with a sample case.

Some rules for this cooperative study can be discussed during the meeting in Thessaloniki, but I believe some points can be put forward:

-the data of all centres will be available to each participating group, under request, if that group wants to utilise them for a common analysis and subsequent publication. This means that each group may request to download the data, perform statistical and/or clinical analysis and prepare abstracts for congress presentation or full manuscript for publication in scientific Journals. Any submission must be discussed, viewed and approved by the other centres. Obviously, data will be uploaded without any reference to the patient identity.

Other uses are not permitted. In any case, any utilisation of the data must be authorized in advance by all the components of the study group.

-data sharing via a FTP service on the web server of the University of Padua is ready and functioning.

At the moment, digital MRI and digital EEG, as well as other potential markers (evoked potentials, fMRI) available in some of the centres are only reported with the date when they were acquired. Files might be shared in the future when an ad-hoc proposal will be put forward by one of the participating centres.