A 2-arm (sequence), 2-period, 2-treatments, single blinded (outcome assessor), randomized crossover-trial (12+12 weeks with immediate contrast) comparing a low-carbohydrate-high-fat diet (LCHF) with a high-carbohydrate-low-fat diet (HCLF) among individuals with prodromal Alzheimer's disease.

The impact of macronutritional composition on cognitive health is not fully understood. On one hand, the World Health Organization (WHO) guidelines propose a limit of total fat intake at 30% of total energy intake (E%), implying that carbohydrates provide at least 50 E%. On the other hand, some pilot studies on ketogenic diets (strict carbohydrate restriction, ≤10 E%) have shown promising results-while liberal carbohydrate restriction has not been investigated in a clinical trial among individuals with Alzheimer's disease or mild cognitive impairment (MCI). It is unclear how important the metabolic state ketosis is for driving potential effects of ketogenic diets on cognitive health outcomes, and our previous observational analyses suggest that even macronutritional changes in the non-ketogenic range might impact cognitive function-although estimated effects differed between sub-samples. This pilot study evaluates the potential of liberal carbohydrate restriction, alternatively fat restriction, as targets for future large scale trials. Participants must be diagnosed with prodromal Alzheimer's disease, which means MCI in combination with biologically validated Alzheimer-pathology-but absence of dementia.

The aim of this trial is to generate a contrast within participants regarding a diet parameter of special interest: the carbohydrate/fat-ratio (CFr). In a randomized order, participants will be exposed to 12 weeks with a low CFr diet (LCHF) and 12 weeks with a high CFr diet (HCLF). In LCHF, sustained ketosis is not an aim but transient mild ketosis may appear in some participants. The following strategies will be used to enhance adherence:

• A mandatory supportive study partner.
• Delivery of one daily meal.
• Delivery of some key ingredients for self-prepared meals.
• Individualized guidance by a dietitian, with consideration of preferred protein sources and complexity of cooking.

Beyond a dichotomized comparison between the diet phases, the study is expected to generate data for a substantial number of observational panel analyses where individual continuous CFr-levels assessed at 5 timepoints may be used as the predictor variable. Those CFr-data will be assessed in parallel with health outcomes including neurodegenerative biomarkers in blood, metabolic biomarkers, and Continous Glucose Monitoring (CGM). Cognitive performance is measured only at 3 timepoints to minimize learning effects. The sample size is adapted to assess feasibility and trends in health outcomes. Due to the limited statistical power there is a considerable risk for type-II errors; therefore, p-values >0.05 should not be interpreted as absence of a clinically meaningful effect in this pilot. Effect modification will be explored by one pre-specified stratification: 1. Apolipoprotein E (APOE) genotypes epsilon-3/4 and 4/4; 2. All other APOE-genotypes.

A cross-over design with immediate contrast (no "wash-out" period) is applied, since it is not possible to define a wash-out value of CFr or reliably keep all participants on a particular CFr between the diet periods. Period 1 (and 2) may have a carry-over effect from pre-study CFr and period 2 may have a carry-over effect from period 1. Primary comparisons against baseline are at the end of each period (weeks 12 & 24) when carry-over effects are assumed to be relatively low.

Inclusion Criteria:

• Ability to fully understand written and verbal information regarding the study and provide signed and dated informed consent

• Prodromal Alzheimer's disease, as defined by Mild Neurocognitive Disorder due to Alzheimer's disease (AD) according to the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) criteria, and evidence for underlying AD pathology by either:

  • Cerebrospinal fluid (CSF) β-amyloid 1-42/1-40x10 ratio < 1 and/or total tau and/or phospho-tau and/or β-amyloid 42 based on local cut-offs OR
  • Magnetic Resonance Imaging (MRI) evidence for medial temporal lobe atrophy (MTA score 1 or higher [mesiotemporal atrophy]) OR
  • Abnormal Fludeoxyglucose F18 (FDG) Positron Imaging Tomography (PET) and/or Pittsburgh Compound-B (PiB) PET compatible with AD type changes.

(When the diagnosis prodromal AD is confirmed from medical record, no cognitive testing or renewed assessment of biological AD-pathology is needed for fulfilling this criterion.)

• Montreal Cognitive Assessment (MoCa) ≥20.
• Availability of a study partner with sufficient contact with the participant, willing and able to give follow-up information on the participant as well as supporting the participant throughout the study.
• Self-reported expected motivation and ability to prepare most weakly meals according to given instructions, with support from the study partner.
• Accept plant-based food, plus food from at least one of the following categories: A. Fish; B. Meat; C. Eggs and dairy
• Ability to reliably undergo a cognitive test in Swedish

Exclusion Criteria:

• Major Neurocognitive Disorder (dementia) according to DSM-5
• Body-mass Index (BMI) < 18 or BMI > 35
• Diagnosed Diabetes Mellitus.
• Ongoing treatment with Metformin, Glucagon-Like Peptide 1 (GLP-1)-analog, or Sodium-Glucose Transport Protein 2 (SGLT-2)-inhibitors
• Diagnosed Familial Hypercholesterolemia
• Untreated or unstable Hypertension
• Alcohol or Substance abuse (current or within 2 years)
• A concomitant serious disease (e.g., cancer, or major psychiatric disorder or other neurological disorder than AD) as judged by study physician
• Major depression or Suicidal ideations (current or within 2 years)
• History of Stroke or Myocardial infarction during the last 5 years.
• Subjects with brain MRI (or CT) scan clinically significant infarct, intracranial macro bleeding, mass lesion or Normal Pressure Hydrocephalus. Those subjects with an MRI scan demonstrating minimal white matter changes (Fazekas scale for white matter lesions classification of 2 or below) and up to 2 lacunar infarcts which are judged to be clinically insignificant are allowed.
• Severe loss of vision or communicative ability
• Conditions preventing cooperation as judged by the study physician.
• Participation in any other intervention trial within 30 days (or, if applicable, 5 half-lives of the relevant drug if longer) before baseline and along the study period.
• Any planned changes in cognitive enhancers (e.g., ginkgo, cholinesterase inhibitors), statins, antidepressants, sleeping pills, supplements like medium-chain triglycerides, or any medication expected to influence cognitive function. Such medications are accepted if taken on a stable dose ≥3 months prior to baseline assessment and should, if possible, remain on the same dose during the study. Unplanned changes that take place within the study do not imply exclusion but should be registered in the case report form (CRF).
• Deviations from habitual diet within 1 month before study start. A carbohydrate-restricted or fat-restricted diet, as well as any time-restricted eating, is accepted as habitual diet if stable (and the participant is open to change).

Since this is a feasibility study with limited statistical power for analyses on health outcomes, there is no prospective plan to share individual participant data (IPD). This may be reconsidered if feasibility can be established and data sharing can be scientifically, ethically, and legally justified.

Assessors of disease monitoring outcomes (defined as Secondary outcomes below) are blinded. One of the primary outcomes (Adherence) assessed by dietitian which cannot be blinded.