Alzheimer's Tied to Insulin Resistance

by John Gever for MedPage Today:Insulin resistance may be a worthwhile treatment target to prevent or delay Alzheimer's disease, several studies reported here suggested.Patients with Alzheimer's disease or mild cognitive impairment receiving 40 IU/day of intranasal insulin detemir in a randomized trial showed improvements in memory relative to the placebo group, but only if they had high levels of insulin resistance at baseline, reported Suzanne Craft, MD, of Wake Forest University in Winston-Salem, N.C.Also, analysis of Kaiser Permanente health system records indicated that diabetic members who started metformin were less likely to develop dementia later on (including forms other than Alzheimer's disease) compared with those initiating other types of anti-diabetic drugs, according to Rachel Whitmer, PhD, of Kaiser's research division in Oakland, Calif.Both studies were reported at the Alzheimer's Association International Conference.The latter finding implicates insulin resistance because metformin, in contrast to other anti-diabetic drugs, is an insulin sensitizer.Earlier reports at the meeting also tied Alzheimer's disease to diabetes and glucose dysregulation. One had found high rates of previously undiagnosed diabetes in patients with mild to moderate Alzheimer's disease while another indicated that diabetic women were at significantly increased risk for cognitive impairment.Another suggestion of a link has been in preliminary findings that the diabetes drug pioglitazone (Actos) may slow or prevent onset of Alzheimer's disease. A large randomized trial called TOMMORROW (sic) is in the final planning stages, to be funded by Takeda and Zinfandel Pharmaceuticals.SNIFF-Long StudyThe insulin detemir trial was prompted by earlier observations that inhaled conventional insulin appeared to improve cognition and certain biomarkers of Alzheimer's disease pathology.Craft explained that the pharmacokinetics of insulin detemir (sold as Levemir for subcutaneous injection), compared with natural insulin, were more attractive for an Alzheimer's disease treatment. It remains active in the body for up to 24 hours after a single dose, peaking about 5 hours after dosing -- thus mimicking the natural basal secretion of insulin in the pancreas.The intranasal route helps insulin get into the central nervous system without having to penetrate the blood-brain barrier, by traveling through nerve channels connecting nasal passages to the brain, she said.Called SNIFF-Long (Study of Nasal Insulin to Fight Forgetfulness-Long Acting Detemir), the study randomized 60 patients with diagnoses of mild Alzheimer's disease or amnestic mild cognitive impairment to 20 or 40 IU/day of intransal insulin detemir or placebo for 3 weeks.Patients were tested at baseline and at the end of treatment with two verbal memory tests with results combined into a composite score, as well as the Benton Visual Retention Test recognition task and a dot-counting test of working memory.At baseline, patients' mean age was about 72, with Mini-Mental State Examination scores averaging 26.Overall, neither dose of insulin was more effective than placebo on the primary outcome, which was change from baseline in memory composite scores.But when patients in each treatment arm were stratified into those with low versus high Homeostasis Model of Assessment-Insulin Resistance (HOMA-IR) values for insulin resistance, the two subgroups in the 40-IU arm both differed significantly from the stratified placebo groups -- in opposite directions.Patients with high insulin resistance taking 40 IU showed an improvement of about 0.75 points in the memory composite, compared with an increase of less than 0.1 point in the placebo patients with high insulin resistance. The 40-IU group patients with low insulin resistance, meanwhile, had a decrease of 1 point in memory score, compared with a decrease of about 0.2 points in the low-resistance placebo subgroup (P=0.015 for both placebo-versus-40 IU comparisons).A similar pattern was seen when the treatment arms were stratified by APOE genotype, with carriers of the epsilon-4 allele benefiting significantly (P=0.025) from 40 IU of insulin detemir relative to placebo, while noncarriers showed worsened memory compared with placebo.In all analyses, results with 20 IU of insulin detemir were similar to placebo.Kaiser Study of Diabetes Drugs and Alzheimer RiskWhitmer presented results from nearly 15,000 Kaiser Permanente members 55 and older who started diabetes drug treatment for the first time from 1999 to 2001, with their subsequent records from 2002 to 2007 checked for new diagnoses of dementia, including Alzheimer's disease, vascular dementia, and other forms.Drugs were classed as metformin, sulfonylureas, thiazolidinediones, and insulin.Across the entire cohort, 9.9% developed incident dementia during the follow-up period, Whitmer reported. The raw data indicated little difference in this rate among patients grouped by drug type. But when adjusted for patient age, race, education level, and diabetes duration, metformin users showed clearly lower rates compared with the other groups.Hazard ratios (HRs) for incident dementia among these other groups relative to metformin users were as follows:

• Sulfonylureas: HR 1.24, 95% CI 1.1-1.4
• Thiazolidinediones: HR 1.18, 95% CI 1.1-1.4
• Insulin: HR 1.28, 95% CI 1.1-1.6

Whitmer reported that adjusting further for baseline glycated hemoglobin (HbA1c) did not alter the results. She also said that restricting the analysis to Alzheimer's disease cases had no effect. Metformin appeared to be protective against vascular dementia as well, with hazard ratios similar to those seen in the overall analysis.Kathleen Welsh-Bohmer, PhD, of Duke University, who is helping to lead the upcoming pioglitazone trial, told MedPage Today that mitochondrial activity in brain neurons appears to be the link between metabolic dysfunction and dementia.Put simply, insulin resistance and diabetes appear to starve mitochondria of their fuel -- glucose -- which in turn causes neurons to cease functioning and eventually die.But the study results also paint a confused picture. Pioglitazone has appeared promising in some early studies, yet thiazolidinediones (the class that includes pioglitazone) were not protective in the Kaiser study. Metformin does appear to protective -- both in the Kaiser study and earlier analyses, but whether it is due to its insulin sensitizing effect remains uncertain.A placebo-controlled trial of metformin in 80 patients with mild cognitive impairment has reportedly been conducted by researchers at Columbia University in New York City to determine if the patients' cognitive decline was slowed or stopped with the drug, but no results have been published.Source: http://bit.ly/15WUvVZ