Hypercortisolism in Type 2 Diabetes with Treatment-Resistant Hyperglycemia: A Treatable Metabolic Phenotype?

The primary goal of diabetes treatment is to prevent diabetes-related complications and improve long-term outcomes, and appropriate glycemic control remains central to this aim [1]. In recent years, diabetes management has placed greater emphasis on individualized treatment strategies, including the selection of glucose-lowering agents according to comorbidities and the adjustment of glycemic targets based on patient characteristics [1,2]. Nevertheless, for most adults with type 2 diabetes who do not have severe comorbidities or substantial hypoglycemia risk, a glycated hemoglobin (HbA1c) target of <7.0%, or more stringently <6.5%, remains a commonly used therapeutic goal [1]. Globally, an estimated 589 million adults, corresponding to approximately 11.1% of the adult population, are living with diabetes [3]. However, many patients still do not achieve recommended glycemic targets [4]. This treatment gap is also evident in the Republic of Korea. In a nationwide population-based study using data from the National Health Insurance Service, 60.6% of Korean patients with diabetes achieved HbA1c <7.0%, leaving approximately 40% above this target. In addition, HbA1c ≥8.0% was observed in 16% of patients [5], indicating that many patients have poor glycemic control and may require more intensive evaluation and treatment strategies.

This glycemic treatment gap is unlikely to have a single cause. Behavioral factors, including poor medication adherence, suboptimal dietary control, and physical inactivity, should be assessed first [6,7]. Clinical inertia and treatment mismatch—such as continued reliance on oral glucose-lowering agents despite progressive β-cell failure or overbasalization in the setting of predominant postprandial hyperglycemia—may also contribute to failure to achieve glycemic targets [8,9]. Accordingly, type 2 diabetes with treatment-resistant hyperglycemia should not be viewed simply as a condition that requires the addition of more glucose-lowering medications. Instead, it should be understood as a complex clinical phenotype in which behavioral factors, β-cell dysfunction, insulin resistance, and treatment mismatch overlap. When poor glycemic control persists despite careful assessment and correction of these factors, overlooked but potentially treatable secondary endocrine causes should be considered [10].

Among these secondary endocrine causes, hypercortisolism has long been suspected to be associated with diabetes and obesity [11–13]. However, it has received relatively little attention in patients without overt Cushing syndrome. Abdominal obesity, hyperinsulinemia, and chronic metabolic stress can activate the hypothalamic-pituitary-adrenal axis, leading in some patients to increased cortisol levels or inadequate cortisol suppression after the 1-mg dexamethasone suppression test (DST) [14]. Therefore, in patients with diabetes or obesity who lack typical Cushing features, such findings have often been interpreted as false-positive DST results, pseudo-Cushing states, or functional hypercortisolism rather than pathologic cortisol excess [15]. This traditional interpretation may have underestimated the clinical relevance and therapeutic potential of non-overt hypercortisolism.

In this context, the CATALYST study provides a useful basis for reevaluating the clinical significance of hypercortisolism in type 2 diabetes with treatment-resistant hyperglycemia. CATALYST was conducted in two parts (Table 1) [16,17]. In the first prevalence phase, patients with difficult-to-control type 2 diabetes, defined as HbA1c 7.5%–11.5% despite multiple glucose-lowering therapies, underwent a 1-mg overnight DST. Hypercortisolism was defined as a post-DST cortisol level >1.8 μg/dL after common causes of false-positive DST results were excluded. Among 1,057 evaluable participants, 252 (23.8%) met the criteria for hypercortisolism. In addition, computed tomography imaging was performed in 219 patients after hypercortisolism was confirmed, and 34.7% had an adrenal abnormality, suggesting that an adrenal source, such as mild autonomous cortisol secretion, may have contributed in some patients [16]. Earlier studies also reported cortisol abnormalities in patients with diabetes or obesity. However, the prevalence of confirmed occult or subclinical hypercortisolism was generally lower, and the study populations were not specifically selected for treatment-resistant hyperglycemia [11–13]. In this respect, the CATALYST findings suggest that hypercortisolism may be more than a nonspecific finding observed in type 2 diabetes and may instead be a potential contributor to the pathophysiology of type 2 diabetes with treatment-resistant hyperglycemia.

The key question, then, is whether hypercortisolism observed in type 2 diabetes with treatment-resistant hyperglycemia represents a treatable metabolic target. CATALYST Part 2 addressed this question by evaluating the glucocorticoid receptor antagonist mifepristone in 136 patients with DST-confirmed hypercortisolism. After 24 weeks, mifepristone significantly reduced HbA1c compared with placebo, and this effect was observed regardless of whether adrenal imaging abnormalities were present. Mifepristone was also associated with reductions in body weight, body mass index, and waist circumference, suggesting that glucocorticoid receptor blockade may affect obesity-related metabolic phenotypes as well as glycemic control [17]. These findings indicate that the hypercortisolism observed in type 2 diabetes with treatment-resistant hyperglycemia may represent a biologically active and therapeutically relevant target. More broadly, they support a shift from focusing only on overt Cushing syndrome to considering a wider spectrum of hypercortisolism, including patients previously categorized as having pseudo-Cushing state or functional hypercortisolism. Recent adrenal incidentaloma guidelines reflect a similar approach: they recommend assessing cortisol-related comorbidities, such as hypertension and type 2 diabetes, in patients with mild autonomous cortisol secretion, with surgery considered as one therapeutic option when these comorbidities are not adequately controlled [18].

Earlier studies attempted to target cortisol-related pathways in type 2 diabetes through 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1) inhibition, which reduces local glucocorticoid action in metabolic tissues by blocking the conversion of inactive cortisone to active cortisol. INCB13739, an 11β-HSD1 inhibitor, was evaluated in a 12-week phase 2 study of 302 patients with type 2 diabetes treated with metformin alone, with a mean baseline HbA1c of 8.3%. However, the overall glycemic effect of INCB13739 was modest: even in the high-dose group, the HbA1c reduction was approximately 0.47% [19], and the agent did not achieve clinical success as a diabetes therapy. Other 11β-HSD1 inhibitors were also investigated across various metabolic diseases, but none became established as diabetes therapies [20]. How, then, should the modest effects observed in earlier 11β-HSD1 inhibitor studies be interpreted? In light of the CATALYST findings, the issue may have been patient selection rather than the cortisol-targeted approach itself. Even in the INCB13739 study, greater glycemic effects were observed in subgroups with higher baseline HbA1c or higher body mass index, with HbA1c reductions of approximately 0.65%–0.76% [19]. Earlier studies mainly enrolled general patients with type 2 diabetes or obesity and did not select for cortisol-related phenotypes such as DST non-suppression [20], potentially diluting treatment effects in the overall study population.

In conclusion, hypercortisolism in type 2 diabetes with treatment-resistant hyperglycemia should be viewed as a potentially relevant metabolic phenotype. Recent evidence suggests that non-overt hypercortisolism may, in some patients, contribute to treatment-resistant hyperglycemia. This interpretation moves abnormal cortisol findings away from being viewed solely as false-positive or functional responses and toward recognizing hypercortisolism as therapeutically relevant in a selected subset of patients. Nevertheless, cortisol-targeted therapy is not ready for broad application. Safety concerns [17], including hypokalemia, edema, fatigue, and nausea, together with the need for accurate patient selection, remain major challenges. Future studies should define which patients have clinically meaningful cortisol-driven dysglycemia and determine how cortisol-targeted therapy can be used safely and effectively. If these challenges are addressed, cortisol-targeted approaches may become a useful addition to the therapeutic strategy for type 2 diabetes with treatment-resistant hyperglycemia.