COPENHAGEN – Selected antidiabetes medications appear to blunt the increased risk of dementia associated with type 2 diabetes, according to a Danish national case control registry study.
Boarding1Now/Thinkstock
This benefit applies to the newer antidiabetic agents – specifically, the dipeptidyl peptidase 4 (DPP4) inhibitors, the glucagon-like peptide 1 (GLP1) analogs, and the sodium-glucose transport protein 2 (SGLT2) inhibitors – and metformin as well, Merete Osler, MD, PhD, reported at the annual congress of the European College of Neuropsychopharmacology.
In contrast, neither insulin nor the sulfonylureas showed any signal of a protective effect against development of dementia. In fact, the use of sulfonylureas was associated with a small but statistically significant 7% increased risk, added Dr. Osler, of the University of Copenhagen.
Elsewhere at the meeting, investigators tapped a Swedish national registry to demonstrate that individuals with type 1 diabetes have a sharply reduced risk of developing schizophrenia.
Type 2 diabetes medications and dementia
Dr. Osler and colleagues are among several groups of investigators who have previously shown that patients with type 2 diabetes have an increased risk of dementia.
“This has raised the question of the role of dysregulated glucose metabolism in the development of this neurodegenerative disorder, and the possible effect of antidiabetic medications,” she noted.
To further explore this issue, which links two great ongoing global epidemics, Dr. Osler and coinvestigators conducted a nested case-control study including all 176,250 patients with type 2 diabetes in the comprehensive Danish National Diabetes Register for 1995-2012. The 11,619 patients with type 2 diabetes who received a dementia diagnosis were matched with 46,476 type 2 diabetes patients without dementia. The objective was to determine associations between dementia and ever-use and cumulative dose of antidiabetes drugs, alone and in combination, in logistic regression analyses adjusted for demographics, comorbid conditions, marital status, diabetic complications, and year of dementia diagnosis.
Patients who had ever used metformin had an adjusted 6% reduction in the likelihood of dementia compared with metformin nonusers, a modest but statistically significant difference. Those on a DPP4 inhibitor had a 20% reduction in risk. The GLP1 analogs were associated with a 42% decrease in risk. So were the SGLT2 inhibitors. A dose-response relationship was evident: The higher the cumulative exposure to these agents, the lower the odds of dementia.
Combination therapy is common in type 2 diabetes, so the investigators scrutinized the impact of a variety of multidrug combinations. Combinations including a DPP4 inhibitor or GLP1 analog were also associated with significantly reduced dementia risk.
Records of glycemic control in the form of hemoglobin A1c values were available on only 1,446 type 2 diabetic dementia patients and 4,003 matched controls. An analysis that incorporated this variable showed that the observed anti-dementia effect of selected diabetes drugs was independent of glycemic control, according to Dr. Osler.
The protective effect appeared to extend to both Alzheimer’s disease and vascular dementias, although firm conclusions can’t be drawn on this score because the study was insufficiently powered to address that issue.
Dr. Osler noted that the Danish study confirms a recent Taiwanese study showing an apparent protective effect against dementia for metformin in patients with type 2 diabetes (Aging Dis. 2019 Feb 1;10(1):37-48).
“Ours is the first study on the newer diabetic drugs, so our results need to be confirmed,” she pointed out.
If confirmed, however, it would warrant exploration of these drugs more generally as potential interventions to prevent dementia. That could open a whole new chapter in the remarkable story of the SGLT2 inhibitors, a class of drugs originally developed for treatment of type 2 diabetes but which in major randomized clinical trials later proved to be so effective in the treatment of heart failure that they are now considered cardiology drugs first.
Asked if she thinks these antidiabetes agents have a general neuroprotective effect or, instead, that the observed reduced risk of dementia is a function of patients being treated better early on with modern drugs, the psychiatrist replied, “I think it might be a combination of both, especially because we find different risk estimates between the drugs.”
Dr. Osler reported having no financial conflicts of interest regarding the study, which was funded by the Danish Diabetes Foundation, the Danish Medical Association, and several other foundations.
The full study details were published online shortly before her presentation at ECNP 2019 (Eur J Endocrinol. 2019 Aug 1. pii: EJE-19-0259.R1. doi: 10.1530/EJE-19-0259).
Type 1 diabetes and schizophrenia risk
Kristina Melkersson, MD, PhD, presented a cohort study that utilized Swedish national registries to examine the relationship between type 1 diabetes and schizophrenia. The study comprised 1,745,977 individuals, of whom 10,117 had type 1 diabetes, who were followed for a median of 9.7 and maximum of 18 years from their 13th birthday. During follow-up, 1,280 individuals were diagnosed with schizophrenia and 649 others with schizoaffective disorder. The adjusted risk of schizophrenia was 70% lower in patients with type 1 diabetes. However, there was no difference in the risk of schizoaffective disorder in the type 1 diabetic versus nondiabetic subjects.
The Swedish data confirm the findings of an earlier Finnish national study showing that the risk of schizophrenia is reduced in patients with type 1 diabetes (Arch Gen Psychiatry. 2007 Aug;64(8):894-9). These findings raise the intriguing possibility that autoimmunity somehow figures into the etiology of the psychiatric disorder. Other investigators have previously reported a reduced prevalence of rheumatoid arthritis in patients with schizophrenia, noted Dr. Melkersson of the Karolinska Institute in Stockholm.
She reported having no financial conflicts regarding her study.
SOURCE: Osler M. ECNP Abstract P180. Melkersson K. Abstract 81.