THU214 Exploratory Study Of The Association Of Genetic Factors With Recovery Of Hypothalamic-Pituitary-Adrenal Axis In Cushing Disease

Disclosure: M.H. Nguyen: None. W. Zhang: None. N. Pankratz: None. J. Lane: None. P. Chittiboina: None. F.R. Faucz: None. A. Liu: None. J.L. Mills: None. C.A. Stratakis: None. C. Tatsi: None. Successful treatment of endogenous Cushing syndrome (CS) is often followed by a period of adrenal insufficiency of variable duration. Recovery of the hypothalamic-pituitary-adrenal (HPA) axis may range from 6 months to more than 2 years. Previous studies have investigated potential clinical and biochemical factors involved in this process with variable results. We performed an exploratory study on genetic factors potentially involved in HPA axis recovery in patients with Cushing disease (CD). Ninety (90) patients with CD, who were treated with transsphenoidal surgery (TSS), had initial remission, and a follow up of at least 3 months postoperatively, were included in the study. Recovery of the HPA axis was defined as morning cortisol level or peak cortisol during a standard dose (250 mcg) IV corticotropin stimulation test of >18mcg/dL. Genetic data was obtained by whole exome sequencing (WES). Variants 1) in genes potentially involved in the HPA axis, 2) rare in the general population and 3) predicted as possibly or probably damaging in silico, were selected from WES. Statistical analysis was performed for each variant or for each gene (calculating all variants present in the same gene) using a Cox proportional hazard model. Correction for multiple comparisons was assessed with the false discovery rate (FDR) method.We identified 97 variants in 47 genes of interest that met the above criteria. Although five variants showed a trend toward association with shorter duration of postsurgical adrenal insufficiency (AI), no significant association was noted after correction for multiple comparisons. Furthermore, each variant was present in only one patient. At the gene-level, the BAG1 gene showed significant correlation with shorter postsurgical duration of AI (Hazard ratio: 28.7, 95% Confidence Interval: 5.4-151.1, FDR q-value=0.004), but both patients with a BAG1 variant later experienced recurrent disease, which could suggest an independent predictor of shorter recovery time. When we excluded patients with recurrence, no significant association was reported on the gene-level. In this exploratory study, we did not identify a significant genetic modifier of HPA recovery time. Other factors, clinical or surgical, may play a more important role in the recovery process or larger cohort studies may be required. Presentation: Thursday, June 15, 2023