OR25-06 Morning ACTH Levels as a Reliable Biomarker for Excluding Autonomous Cortisol Secretion in Incidetally Discovered Adrenal Adenomas. A Prospective Cohort

Abstract

Adrenal incidentalomas are common with a prevalence of 3-10% and in up to 30% of cases may have probable autonomous cortisol secretion. Hypercortisolism is associated with substantial cardiometabolic morbimortality and can physiologically decrease ACTH levels. Objective: To determine the sensitivity, specificity, and positive and negative predictive values of ACTH levels in evaluating autonomous cortisol secretion in a prospective cohort of incidentally discovered adrenal adenomas. Methods: We prospectively evaluated 224 consecutive adult subjects with incidentally discovered adrenal masses on computed tomography. Finally, 168 participants with radiographic adenoma criteria underwent systematic hormonal assessment, including measurements of morning cortisol and ACTH on day 1, and a 1 mg dexamethasone suppression test (DST) on day 2. Hypercortisolism was excluded if the DST was < 1.8 mcg/dL. Autonomous cortisol secretion was defined as a DST > 5.0 mcg/dL and DST levels of 1.8-5.0 mcg/dL were considered to be possibly autonomous hypercortisolism. We evaluated the correlation of ACTH levels with clinical, radiographic, and endocrine variables. In order to identify the most sensitive threshold value for diagnosing autonomous cortisol secretion, we determined ROC curves and negative likelihood ratio (NLR). Concordance of repeated ACTH was assessed using Bland Altman analysis. Results: The characteristics of the cohort were mean age 56 (+/- 11.8) years, 76% female, adenoma size 19 (+/- 7) mm, and 13% bilateral adenomas. Mean ACTH was 15 (+/- 11) pg/ml (range 5-72) and the mean DST was 2.2 (+/- 3.0) ug/dL (range 0.4-25.9). Fifty-four (32%) participants had a DST ≥1.8mcg/dL and 13 (8%) a DST≥5.0 mcg/dL. We found no correlation between ACTH levels and age, gender or body mass index. ACTH was inversely associated with adrenal adenoma diameter (r=-3.3 p=0.002) and volume (r=-2.9 p=0.008). There was an inverse association between ACTH and DST values (r=-3.1 p=0.01). In the subgroup of patients with a second ACTH measurement we found high concordance, with mean difference of 0.16+/-3.6 pg/ml (p=0.83). ROC analysis showed that an ACTH ≥20 pg/ml had a sensitivity of 98% to exclude hypercortisolism, with a negative predictive value of 97% and a negative likelihood ratio of 0.06. The only case with DST≥1.8 and ACTH≥20 had Cushing′s phenotype with both an adrenal adenoma and a pituitary ACTH-producing adenoma. Systematic evaluation of morning cortisol and ACTH allowed the detection of 5 cases of false negative low DST values due to the use of non-oral corticosteroids. Conclusion: In this cohort, an ACTH ≥20pg/ml excluded autonomous cortisol secretion with excellent sensitivity and negative predictive value, providing strong reassurance that there is no clinically relevant hypercortisolism. Therefore, subjects with a normal DST and ACTH ≥20pg/ml should be candidates for relaxed surveillance.