The purpose of the study is to compare the concentrations of Vitamin B1 (thiamine), Vitamin B6 (pyridoxal-5-phosphate), folate, Vitamin B12 (cobalamin), Vitamin C (ascorbic acid), Vitamin A (retinol), Vitamin E (alfa-tocopherol), homocystein, uric acid, F2 8-α-isoprostane, 8-deoxyguanosine, retinoids, tau-protein and β-amyloid in spinal fluid, metabolomics, proteomics, m-RNA for DNA repair enzymes and DNA in patients who suffer from mild cognitive impairment (MCI) or mild dementia of Alzheimers type, with healthy controls.

A second aim is to explore the association between vitamin and nutrient reductions, if any, and cognitive function as well as vascular score and possible changes in the MRI.

Dementia of the Alzheimer type (AD) is a serious illness affecting 60-70 000 people in Norway with an estimated cost of 12 Billion NOK (Norwegian kroner) per year. The Department of Social Affairs and Health has initiated psycho-social intervention programs for patients as well as caregivers, however, no research regarding etiology, pathology or prevention of the disease.

In an earlier pilot study conducted on AD patients in a moderate stage, there were significant decrease in the concentrations of the vitamins thiamine, pyridoxal-5 phosphate, ascorbic acid retinol, cobalamin, and increased homocystein, compared with the control group of healthy elderly people. Logistic regression analysis showed that 5 models using different combinations of vitamins and spinal protein, all completely separating patients with dementia from healthy controls.

The current study is a replication of the above mentioned study, and will include up to 120 AD patients with mild cognitive impairment (MCI) or mild dementia (MMSE above 24/30) and up to 60 healthy elderly controls. Measurement of vitamines and nutrients in blood and spinal fluid is the central element of the study. The specimens will therefore be obtained from patients and controls who have abstained from supplements the last four weeks.

Outcome measures will include biochemical analysis of nutrients, expressions of peroxidation, expressions of DNA damage, proteomics and metabolomics of spinal fluid, blood and urine, microarray analysis of different nutritional deficiency subgroups, together with the routine diagnostics investigations undertaken at the memory clinic, and a nutritional assessment.

The patients and the healthy controls will only receive information about results of examinations that are part of the routine investigation at the memory clinic, this according to the Norwegian Biotechnology Law and the permission from the National Committees for Research Ethics in Norway.

In addition to the vitamins listed as primary outcomes, the following measurements will be carried through:

1. Isoprostane measurements: The findings of F2-α-isoprostane-increase in cerebrospinal fluid close to the diseased organ, and not in the plasma, makes it important to confirm these findings. Isoprostane measurement in CNS might become an important parameter in prospective treatment studies where the actual treatment at least should normalize the pathological finding.

2. 8 hydroxy-2-deoxy guanosine increase in cerebrospinal fluid or urine, reveals DNA-damage suggesting further mechanisms for the AD development.

3. Spinal fluid: Nutrients in spinal fluid, expressions of increased peroxidation like F2-α-isoprostane, tau proteins and possibly with LC/MS methodology metabolic pattern (metabolomics) and different proteins (proteomics).

   • Proteomics; determination of protein profiles. The emerging technology of mass spectrometry (MS)-based quantitative proteomics provides a powerful tool to systematically and quantitatively assess quantitative differences in protein profiles. Discovery based proteomics compares the proteome of a diseased sample versus normal at a global scale, and has been widely applied to study various human diseases with the goal of identifying biomarkers and/or reveal the pathogenesis of diseases. In this study, protein samples from control and diseased will be labeled with different stable isotopes, digested to peptides and analyzed by LCMS. Quantitative differences of the proteins from these groups will be performed by state-of-the-art hybrid mass spectrometer, LTQ XL-Orbitrap. The Orbitrap instrument has extremely high resolving power, mass accuracy and sensitivity.
   • Metabonomics: Metabonomics aims at measuring and mathematically modeling changes in the metabolite concentrations found in biological fluids and tissues. Nuclear magnetic resonance (NMR) spectroscopy allows the detection and quantification of compounds in large mixtures, such as the products of metabolism in biological fluids and tissues. There are hundreds of such metabolites to be found in cerebrospinal fluid, plasma, serum and urine. NMR spectroscopy offers the advantage of measuring their concentrations with micromolar sensitivity while conveniently requiring only little sample preparation. In recent studies involving large sets of spectra, the variation of results was shown to be as small as 2 % reflecting a high degree of reproducibility. By identifying patterns in the metabolic fingerprint of the organism, one can study the effects of diet, drugs and disease using multivariate statistics such as principal component analysis and partial least squares attempts to interpret NMR spectra.

4. Genetic analyses: m-RNA measurement with SNP-profiles for

   • DNA repair.
   • Active components in processing av APP and Abeta.
   • Will include if possible, sequencing of genome and/or GWAS.

Inclusion Criteria:

• Cognitive impairment with a MMSE score of 24 or better.
• Depression score below 9 both on MADRS and Cornell.

Exclusion Criteria:

• Acute or chronic disease with CRP above 10.
• Lewy Body Dementia or Frontal Lobe Dementia.
• Dementia due to a known brain vascular disturbance, hemorrhage or thrombosis.
• Patients who had not stopped their intake of vitamins and food additives the last four weeks.

Patients referred consequtively to the Memory Clinic, Oslo University Hospital, Ullevål