In This Issue of Diabetes

A new mechanism that explains some of the negative metabolic effects of glucocorticoids is explored by Vali et al. (p. 211). Specifically, it appears that glucocorticoids and the corresponding glucocorticoid receptor (GR) restrain the expansion and vascularization of adipose tissue, resulting in inflammation and insulin resistance. Moreover, this suppression is achieved by decreasing the angiogenic vascular endothelial growth factor A (VEGFA) and its regulator hypoxia-inducible factor 1α (HIF-1α), suggesting multiple leads for therapies that would likely focus on antagonizing GRs in adipocytes. The findings come from a combination of studies with adipocyte GR-deficient mice and a small number of patients with Cushing syndrome. They extend prior observations of the authors that GR-deficient mice treated with corticosterone experience large adipose tissue expansion, but this expansion is paradoxically associated with improved metabolic health. They found that corticosterone treatment in control mice resulted in large decreases in Vegfa and Hif-1α mRNA levels in adipose tissue and adipocytes, but in the GR-deficient mice the effect was almost completely prevented. This suggests the previously observed expansion of adipose tissue was actually due to an increase in the vascular network. Notably, when the same mice were treated with corticosterone and aflibercept (which blockades VEGFA), the authors found that adipose tissue expansion was prevented along with the development of the metabolically healthy profile. In a final step, the authors stratified 20 individuals with Cushing syndrome according to VEGFA expression levels in subcutaneous adipose tissue. They found that individuals with high cortisone levels and lower VEGFA expression had a worsened metabolic profile than patients with higher VEGFA expression. Accordingly, the authors note there is a combined action of VEGFA and the glucocorticoid/GR pathway in adipose tissue and metabolic homeostasis. Commenting further, author Marthe Moldes said, “We believe that innovative approaches to selectively antagonize the glucocorticoid receptor, specifically in adipocytes, will be relevant to prevent the adverse metabolic effects of glucocorticoids such as insulin resistance and hepatic steatosis.”

An association exists between canagliflozin-induced glycosuria and protection against multiple cardiorenal outcomes that were included in the Canagliflozin and Renal Events in Diabetes with Established Nephropathy Clinical Evaluation (CREDENCE) trial, according to Ferrannini et al. (p. 250). Specifically, subjects treated with canagliflozin who were in the top quartile of urine glucose-to-creatinine ratios were significantly protected against the primary composite end point of the trial (renal disease or death from renal or cardiovascular disease), hospitalization for heart failure, and all-cause death. Although prior studies with sodium–glucose cotransporter 2 (SGLT2) inhibitors (including canagliflozin) have shown pronounced reductions in cardiorenal outcomes, the mechanisms underlying the benefit are still uncertain, prompting the study. Using urine samples collected at the time of the study, the authors determined urine concentrations of glucose, creatinine, and ketones at baseline and after 1 year in ~2,600 individuals who received placebo or canagliflozin 100 mg/day. Participants had type 2 diabetes, chronic kidney disease, and albuminuria. The authors looked for associations between glycosuria and the primary and secondary end points of the trial. They found that canagliflozin treatment increased urinary glucose excretion from ~3% at baseline to approximately 30% at year 1. For comparison, those on placebo moved from 3% to ~5% over the same period, although in all cases the variance was considerable. The patients receiving canagliflozin and in the top quartile of urine glucose-to-creatinine ratios had significantly reduced hazards for the primary and secondary outcomes, and this persisted after multiple adjustments for all the usual risk factors. While they describe their interpretation as complex, they conclude that “glycosuria is the sine qua non (i.e., indispensable step) and the trigger of most, if not all, changes that follow SGLT2 inhibition.” They underline that more studies are needed. “Importantly, abundant glycosuria in response to SGLT2 inhibition occurs also in subjects without diabetes and in a quota of patients with markedly reduced renal function,” said author Ele Ferrannini. “This explains why cardiorenal protection persists in these patients. Furthermore, glycosuria can be used to check adherence to treatment with SGLT2 inhibitors.”

A long-acting agonist of the glucose-dependent insulinotropic polypeptide receptor (GIPR) appears to counter the weight gain and appetite issues associated with rosiglitazone while also improving its insulin-sensitizing properties, according to Furber et al. (p. 292). The study in mice also provides insights into the mechanisms behind GIPR agonism, which will have considerable relevance to the application of other drugs, such as tirzepatide, which targets GIPR and glucagon-like peptide 1 receptor (GLP-1R). The findings come from a series of genetic and pharmacological studies with obese insulin-resistant mice and was designed to investigate whether GIPR agonism influences the pharmacological actions of rosiglitazone. According to the authors, success with pairing GIPR activation with GLP-1R agonism has prompted the search for additional therapeutic partners for the approach as well as interest in how GIPR agonism mediates efficacy. Using a loss-of-function approach, they found that rosiglitazone does not require activation of GIPR to drive either weight gain or improved insulin tolerance despite increased mRNA expression of Gipr in adipose tissue. Chronic administration of rosiglitazone and the long-acting GIPR agonist then prevented the usual weight gain and excess adiposity, likely via the prevention of the appetite stimulant effects of the drug. The authors also found that the combination improved insulin sensitivity, suppressed endogenous glucose production, and enhanced overall glycemic control independently of changes in adiposity. Finally, turning to RNA sequencing, they found that the combination appeared to promote expression of proteins that drive metabolic health in white and brown fat. Accordingly, the authors suggest that the findings have clinical relevance in terms of obesity and type 2 diabetes management, as they provide insights into how GIPR agonism works. They note that further studies are needed in terms of mechanisms and that clinical validation (presumably including safety studies) will be needed in adult humans with obesity and diabetes.

No genetic evidence could be found for a causal association between the consumption of the low-calorie sweetener erythritol and cardiometabolic disease, according to Khafagy et al. (p. 325). Specifically, the Mendelian randomization study did not find evidence that increased erythritol was causally associated with type 2 diabetes, coronary artery disease, chronic kidney disease, BMI, or HbA1_c. There was also no evidence of reverse causation where, for example, cardiometabolic traits have an impact on erythritol concentrations. Consumption of low-calorie sweeteners is increasingly popular as a substitute for sugar, and some positive health benefits have been attached to the consumption of such sweeteners. However, there are also persistent concerns around the adverse cardiovascular effects based on evidence from observational studies. As the authors point out, however, observational studies can be biased by confounding and reverse causality, which prompted them to turn to Mendelian randomization to investigate the relationship. They found no evidence that increased erythritol intake increases coronary artery disease or any other glycemic and renal traits. There was only weak evidence that erythritol might decrease BMI or increase waist-to-hip ratio adjusted for BMI. As for the previous studies that did document an association between erythritol and higher cardiometabolic risk, they note the possibility of reverse causality, where there is higher erythritol consumption in individuals with higher cardiometabolic risk anyway. In terms of limitations, the authors note that erythritol can be endogenously produced, and while they assumed a linear relationship between erythritol and cardiometabolic disease, it has not been established whether the risk starts above a certain threshold. On that basis, they suggest that further studies are needed to establish whether a nonlinear relationship exists. “Given the previous data, there is some concern about using erythritol as a low-calorie sugar substitute,” said author Satya Dash. “Collectively, these findings are reassuring, but further prospective randomized controlled studies are needed to definitively confirm these findings.”