Dysglycemia, the primary biochemical abnormality of type 2 diabetes and pre-diabetes, has been repeatedly linked to late-life cognitive dysfunction, as well as increased risk for cognitive impairment syndromes and dementia.1,2 Despite the consistency in this observed relationship, meta-analytic evidence suggests that the strength of the association is in fact, weak, with one review of 86 studies suggesting that dysglycemia accounts for only 10 % to 15 % of the variance in cognitive outcomes.3 Thus, while dysglycemia may, in part, contribute to poorer cognition,4 other pathophysiological events that occur alongside dysglycemia in the development and evolution of type 2 diabetes could themselves play a contributing role. In particular, insulin sensitivity and secretion, major regulators of glucose metabolism, have garnered significant attention in the literature as an important factor in the risk for cognitive impairment syndromes and dementia.5

Dysglycemia in pre-diabetes and type 2 diabetes is driven by the co-evolution of loss of peripheral insulin sensitivity (ISens) and impaired insulin responses (IResp) to an oral or intravenous glucose challenge as a manifestation of pancreatic islet β-cell dysfunction.6,7 A growing body of observational studies has shown that lower ISens (i.e., greater insulin resistance) in midlife and old age, in people with or without pre-diabetes or diabetes, is associated with late-life cognitive impairment, as well as Alzheimer's Disease (AD) and vascular dementias.5 Lower ISens may also be associated with a higher burden of AD neuropathology among those carrying the most prevalent genetic risk factor for sporadic AD (the APOEε4 allele).8 In vitro cell models of AD have also suggested that applying treatments that increase ISens (such as metformin) may protect cultured neurons from AD neuropathological insult.9 Together, these data support the concept that impaired ISens contributes to cognitive dysfunction and risk for cognitive impairment syndromes.

Independent contributions of β-cell dysfunction, the major pathophysiologic component of dysglycemia along with loss of ISens, is less well understood in the context of cognitive function and risk for cognitive impairment syndromes and AD-like brain pathology.10, 11, 12 Human studies have been limited, but animal studies have consistently shown that β-cell ablation is associated with greater AD neuropathology burden.13, 14, 15, 16, 17 Conversely, in animal models of AD, pharmacological treatment with agents that stimulate an insulin response (IResp) from β-cells, like liraglutide,18,19 have reduced AD neuropathology burden. Impaired insulin response(s) as manifestations of β-cell dysfunction may therefore be another, potentially independent, factor that increases the risk for late-life cognitive dysfunction. However, the evidence remains mixed, with some studies reporting that greater IResp may be associated with poorer cognitive outcomes.20,21

Despite initial evidence that both IResp and ISens may be associated with late-life cognitive function, few studies to date in people at high risk for type 2 diabetes have examined the independent longitudinal effects of ISens and IResp on cognitive dysfunction and risk for cognitive impairment syndromes. Examination of both factors alongside dysglycemia is important to understand the independent contribution of each to cognitive health. To address this scientific gap, we examined ISens and IResp in people with pre-diabetes, by definition at high risk for developing type 2 diabetes, who participated in the prospective longitudinal Diabetes Prevention Program Outcomes Study (DPPOS). Here, we hypothesized that a greater insulin response and lower insulin sensitivity, would be associated with poorer cognitive health.