June 13, 2011
By John Gever, Senior Editor, Reviewed by Zalman S. Agus, MD; Emeritus Professor
University of Pennsylvania School of Medicine.
A low-fat diet led to improvements in a putative biomarker of Alzheimer’s disease risk in patients with mild cognitive impairment, but it had the opposite effect in healthy older adults, a small, short-term trial found.
In a randomized 49-person trial, mean levels of the 42-amino-acid form of amyloid-beta protein in cerebrospinal fluid (CSF AB42) increased in the cognitively impaired patients after eating the low-fat diet for four weeks, whereas mean levels of this marker declined in healthy adults on the same diet, according to a report in the June issue of Archives of Neurology.
Declines in CSF AB42 have been associated with worsening cognitive function and increased risk of Alzheimer’s disease in some earlier studies.
On the other hand, a separate longitudinal cohort study reported online this week in Archives of Neurology indicated that changes in CSF AB42 levels over three years were not closely correlated with changes in cognitive ability.
Rather, CSF AB42 levels over time showed different patterns depending on whether patients were cognitively normal, mildly impaired, or had overt dementia at baseline.
This stage-dependent pattern was also reflected in the diet study, reported by Suzanne Craft, PhD, of the VA Puget Sound Health Care System in Seattle, and colleagues.
In addition to testing the low-fat dietary intervention, the trial included two other arms evaluating a high-fat diet in patients with mild cognitive impairment and healthy controls.
No change in CSF AB42 levels was seen in the patients, but the normal controls had a substantial increase.
Craft and colleagues suggested that “stage-dependent differences in the trajectory of CSF AB42″ were responsible for the varied responses to the dietary interventions in patients versus controls.
Of note, however, the intervention was relatively brief and the study did not examine effects on cognitive change or progression to Alzheimer’s disease.
Taken together, the two trials suggest a relatively complex relationship between clinical presentations and CSF AB42, which had been among the most promising objective biomarkers for risk of progressive cognitive impairment, as well as the impact of diet.
Diet Study Includes High-, Low-Fat Arms
In the study by Craft and colleagues, 29 patients with amnestic mild cognitive impairment and 20 cognitively normal controls – mean age in both groups was just under 70 – were randomized to either a high- or low-fat diet.
The low-fat diet was designed to have 25% of calories from fat, with less than 7% from saturated fat and a glycemic index of less than 55. The high-fat diet had 45% of calories from fat with saturated fat accounting for more than 25% of calories and a glycemic index of more than 70.
In addition to CSF AB42, levels of other factors such as insulin, tau protein, and apolipoprotein E in CSF were measured, along with blood lipids and insulin.
Effects of the two diets on serum cholesterol and insulin profiles were as expected, with relatively unhealthy changes seen with the high-fat diet in both groups of participants, versus improvements with the low-fat diet.
No significant differences in these parameters were seen in the cognitively impaired patients relative to controls.
On the other hand, changes in CSF AB42 and insulin both differed significantly between the two participant groups.
In response to the low-fat diet, cognitively impaired patients had increases in CSF insulin and AB42, whereas decreases were seen in the healthy controls (P≤0.05 for CSF insulin, P≤0.001 for CSF AB42).
The high-fat diet seemed to affect the CSF biomarkers only in the normal controls. CSF insulin declined dramatically in controls with this diet but no mean change was seen in the patients. A similar pattern was seen for CSF levels of apolipoprotein E.
Relative to those on the high-fat diet, those on the low-fat diet – among both cognitively impaired patients and controls – improved their performance on a test of delayed visual memory (P≤0.05). Little change from baseline was seen in either group with the high-fat diet, but scores increased substantially from baseline in patients and controls.
Summarizing the results, Craft and colleagues wrote, “our study supports further investigation into the possibility that consumption of a diet high in saturated fat and simple carbohydrates may contribute to pathologic processes in the brain that increase the risk of Alzheimer’s disease. Conversely, diets low in saturated fat and simple carbohydrates may offer protection against Alzheimer’s disease and enhance brain health.”
However, they cautioned that the four-week study was not long enough to evaluate effects on cognition generally. They also noted that individuals with high baseline cholesterol levels and statin use were excluded.
Baseline Impairment Has Role in Second Study
The other study reported online in Archives of Neurology was written by Raymond Lo, MD, of the University of California Berkeley, and colleagues on behalf of the large Alzheimer’s Disease Neuroimaging Initiative.
It followed 819 individuals 55 to 90 years old for three years beginning in 2005, with CSF AB42 and tau protein measurements, PET scans with fludeoxyglucose F18 tracers (indicating metabolic activity in the brain), and MRI measures of hippocampal volume.
Participants included 229 with normal cognition at baseline, 397 with mild cognitive impairment, and 193 with Alzheimer’s disease.
Lo and colleagues found that CSF AB42 levels declined from baseline in all three participant groups, with the greatest numerical declines in those with normal cognition at baseline. The rate of decline in these individuals was about 50% greater than in those with baseline impairments.
CSF tau levels remained stable in the normal cognition and mildly impaired groups, while declining more rapidly in those with Alzheimer’s disease at baseline.
The PET and MRI results both showed progressively greater declines in metabolic activity and hippocampal volume, respectively, with the degree of baseline impairment.
Similarly, increases in scores on the cognition subscale of the Alzheimer’s Disease Assessment Scale, reflecting growing impairment, were smallest in those with normal baseline cognition and greatest in those with Alzheimer’s disease.
Another factor the researchers considered was APOE genotype status. Individuals with impairment or dementia at baseline who were positive for the high-risk epsilon-4 allele showed accelerated hippocampal atrophy, but not in those with normal baseline cognition.
But APOE genotype did not appear to affect the PET results or changes in CSF AB42 during the study.
Although the relationships were complex, Lo and colleagues indicated that serial measurement of these biomarkers “provides a potential approach for early diagnosis of Alzheimer’s disease.”
In an accompanying commentary, Archives of Neurology editor Roger Rosenberg, MD, indicated that the study by Lo and colleagues supports the emerging view that amyloid protein deposition “occurs when individuals are still cognitively normal, leading after a lag period to neuronal dysfunction, metabolic impairment, and neurodegeneration.”
“My belief and experience say that therapeutic progress will depend on biomarker detection of risk of Alzheimer’s disease, with prompt initiation of safe preventive therapies before the patient is diagnosed with the disease,” Rosenberg wrote.
But both of the new Archives studies involved repeated testing with invasive and/or expensive diagnostic technologies. Neither Rosenberg nor Lo and colleagues addressed the practicalities of conducting these evaluations on a population-wide basis.
The diet study was funded by the National Institute on Aging, with additional support from the VA and the Nancy and Buster Alvord Endowment.
The longitudinal biomarker study was funded by the National Institute on Aging and the National Institute of Biomedical Imaging and Bioengineering with contributions from Abbott, AstraZeneca, Bayer Schering Pharma, Bristol-Myers Squibb, Eisai Global Clinical Development, Elan, Genentech, GE Healthcare, GlaxoSmithKline, Innogenetics, Johnson and Johnson, Eli Lilly, Medpace, Merck, Novartis, Pfizer, F. Hoffman-La Roche, Schering-Plough, Synarc, the Alzheimer’s Association, the Alzheimer’s Drug Discovery Foundation, and the FDA.
Craft and colleagues declared they had no relevant financial interests.
Lo declared no relevant interests. Other co-authors intellectual property interests in biomarker technology as well as relationships with Pfizer, Novartis, Elan, Janssen, NeuroPhage; serves as a consultant to Elan, Wyeth, Eisai, Neurochem, Schering- Plough, Bristol-Myers Squibb, Eli Lilly, NeuroPhage, Merck, Roche, Amgen, Genentech, Abbott, Medivation, and other commercial entities.
Rosenberg declared he had no relevant financial interests.
Primary source: Archives of Neurology
Bayer-Carter J, et al “Diet intervention and cerebrospinal fluid biomarkers in amnestic mild cognitive impairment” Arch Neurol 2011; 68: 743-752.
Additional source: Archives of Neurology
Lo R, et al “Longitudinal change of biomarkers in cognitive decline” Arch Neurol 2011; DOI:10.1001/archneurol.2011.123.
Additional source: Archives of Neurology
Rosenberg R “Treat Alzheimer disease before it is symptomatic” Arch Neurol 2011; DOI:10.1001/archneurol.2011.135.