Cancer incidence in endogenous hypercortisolism.

PURPOSE: Evidence regarding incident malignancy in endogenous hypercortisolism remains limited. We evaluated cancer risk across the spectrum of cortisol excess and identified predictors of malignancy. METHODS: This retrospective single-center cohort study included 347 patients with endogenous hypercortisolism (138 overt Cushing's syndrome and 209 mild autonomous cortisol secretion). Patients with active malignancy at baseline were excluded. Incident cancers diagnosed after hypercortisolism diagnosis were identified through hospital records. Observed malignancies were compared with age- and sex-specific national cancer incidence rates to calculate standardized incidence ratios (SIRs). Cox regression analyses were performed to assess predictors of incident cancer. RESULTS: A total of 42 incident malignancies (12.1%) occurred during a median follow-up of 87 months (IQR, 35-129). Overall cancer incidence was significantly higher than expected (42 observed vs. 12.27 expected; SIR 3.42, 95% CI 2.47-4.63). Risk was increased in overt Cushing's syndrome (SIR 5.12, 95% CI 3.07-7.97), adrenal Cushing's syndrome (SIR 4.94, 95% CI 2.26-9.37), Cushing's disease (SIR 5.27, 95% CI 2.53-9.69), and mild autonomous cortisol secretion (SIR 2.68, 95% CI 1.70-4.02). Thyroid cancer was the most frequent malignancy (31.0%). Excess risk persisted after excluding thyroid cancers (overall SIR 2.52, 95% CI 1.69-3.62). In univariable analyses, diabetes mellitus, severe obesity, and HbA1c ≥ 64 mmol/mol were associated with incident malignancy in overt Cushing's syndrome, while bilateral adrenal tumors were associated with incident malignancy in adrenal Cushing's syndrome. CONCLUSIONS: Endogenous hypercortisolism is associated with increased incident cancer risk, particularly in overt Cushing's syndrome. Long-term oncologic vigilance may be warranted.