Treating Alzheimer's Disease by Treating Insulin Resistance?

Insulin resistance (IR) is a well-known underlying mechanism in such metabolic disorders as diabetes, hypertension, obesity, and dyslipidemia, all of which increase the risk of Alzheimer’s disease (AD). Understanding precisely how these comorbid conditions contribute to the pathophysiology of AD is challenging because of their heterogeneous etiologies and complex, interdependent relationships. Suzanne Craft, PhD, of the University of Washington School of Medicine/VA Puget Sound Medical Center (Seattle, WA) discussed what is known about the role of insulin resistance in AD and the treatment of insulin resistance as a novel therapeutic strategy in her plenary session presentation at the 2009 Alzheimer's Association International Conference on Alzheimer's Disease, held in Vienna, Austria.

Insulin plays many roles in normal brain function, including the regulation of amyloid precursor protein (APP) release and enhancement of memory at optimal doses. However, at high levels, insulin can impair memory. Insulin resistance is usually paired with hyperinsulinemia.

Animal studies in recent years have shown that insulin and beta-amyloid appear to exert reciprocal effects. For example, synthetic beta-amyloid oligomers induce the removal of insulin receptors from the dendritic plasma membrane, which reduces insulin signaling. Insulin, on the other hand, is protective against beta-amyloid oligomer toxicities, including the loss of insulin receptors. Furthermore, insulin also controls Abeta trafficking. In other studies, oligomers have been shown to reduce long-term potentiation (LTP), a process that is required for memory formation. Insulin, however, can be used to rescue these neurons from the LTP deficit. Although less is known about the effects of insulin on tau, several studies have shown that insulin reduces tau phosphorylation.

Recent work has linked impaired vascular function to reduced insulin signaling in a specific pathway (PI3K) and blurred the lines between AD and vascular dementia. Microvascular lesions are a common pathology in type II diabetes, which is associated with insulin resistance (unlike type I, which is associated with inadequate insulin production). Microvascular lesions have been found to be associated with dementia in adults with diabetes.

These findings led to studies of diabetics with dementia, which had intriguing findings. Untreated diabetics with dementia demonstrated the expected beta-amyloid plaque pathology. However, in diabetics with dementia whose diabetes was treated, the AD pathology was much reduced, but vascular pathology was increased. This finding suggested to Dr. Craft's group that brain insulin signaling and transport was impaired in the untreated group and that reciprocal effects of insulin on beta-amyloid pathology prevented the development of AD pathology in that group.

They devised a therapeutic strategy based on normalizing insulin signaling in the CNS. Intranasal administration of insulin was selected in order to bypass systemic effects. Validation of a beneficial effect of intranasal insulin on memory in a diabetic mouse model led to implementation of this approach in 25 patients with mild cognitive impairment or mild AD. The trial design is simple: cognitive testing is conducted 15 minutes after intranasal administration of insulin. A 50% improvement in memory, improved attention, increased Abeta40/Abeta42 ratio, and a dose-response curve have been observed. The trial will be completed this fall.

Dr. Craft noted in the Q&A that her study holds hope that increasing insulin signaling may be a viable treatment avenue in AD. Disappointing results were presented earlier in the week by GlaxoSmithKline, which reported that a large phase III trial with its PPAR-gamma agonist, rosiglitazone (Avandia) found that the drug was ineffective in the treatment of mild to moderate AD. The mechanism of action of PPAR-gamma agonists in diabetics is to reduce insulin resistance (versus treatment with insulin to overcome insulin resistance).