Potential Transcriptional Biomarkers to Guide Glucocorticoid Replacement in Autoimmune Addison's Disease

Åse Bjorvatn Sævik; Anette B. Wolff; Sigridur Björnsdottir; Katerina Simunkova; Martha Schei Hynne; David William Peter Dolan; Eirik Bratland; Per M Knappskog; Paal Methlie; Siri Carlsen; Magnus Isaksson; Sophie Bensing; Olle Kämpe; Eystein S Husebye; Kristian Løvås; Marianne Øksnes


J Endo Soc. 2021;5(3) 

In This Article

Abstract and Introduction


Background: No reliable biomarkers exist to guide glucocorticoid (GC) replacement treatment in autoimmune Addison's disease (AAD), leading to overtreatment with alarming and persistent side effects or undertreatment, which could be fatal.

Objective: To explore changes in gene expression following different GC replacement doses as a means of identifying candidate transcriptional biomarkers to guide GC replacement in AAD.

Methods: Step 1: Global microarray expression analysis on RNA from whole blood before and after intravenous infusion of 100 mg hydrocortisone (HC) in 10 patients with AAD. In 3 of the most highly upregulated genes, we performed real-time PCR (rt-PCR) to compare gene expression levels before and 3, 4, and 6 hours after the HC infusion. Step 2: Rt-PCR to compare expression levels of 93 GC-regulated genes in normal versus very low morning cortisol levels in 27 patients with AAD.

Results: Step 1: Two hours after infusion of 100 mg HC, there was a marked increase in FKBP5, MMP9, and DSIPI expression levels. MMP9 and DSIPI expression levels correlated with serum cortisol. Step 2: Expression levels of CEBPB, DDIT4, FKBP5, DSIPI, and VDR were increased and levels of ADARB1, ARIDB5, and POU2F1 decreased in normal versus very low morning cortisol. Normal serum cortisol levels positively correlated with DSIPI, DDIT4, and FKBP5 expression.

Conclusions: We introduce gene expression as a novel approach to guide GC replacement in AAD. We suggest that gene expression of DSIPI, DDIT4, and FKBP5 are particularly promising candidate biomarkers of GC replacement, followed by MMP9, CEBPB, VDR, ADARB1, ARID5B, and POU2F1.


In autoimmune Addison's disease (AAD), patients suffer from deficiency of glucocorticoids (GCs) and mineralocorticoids due to autoimmune destruction of the adrenal cortex, a fatal condition if left untreated.[1] Current treatment strategies rely on replacement of cortisol (the main GC) and aldosterone (the main mineralocorticoid), usually twice or thrice oral hydrocortisone (HC) or cortisone acetate in combination with once daily oral fludrocortisone and salt. Conventional treatment strategies fail to restore good health in patients with AAD, evident by lower quality of life,[2,3] increased prevalence of cardiovascular disease, metabolic syndrome, and infections as well as overall higher mortality rates.[4–7]

In addition to the lethal threat of acute adrenal crisis, current evidence points to 2 main causes of the deleterious health outcomes in AAD. First, none of the currently used treatment modalities perfectly mimic the physiological circadian rhythm of GC production. Second, there is no biomarker available to aid physicians in determining the correct GC replacement dosage for each individual.[8,9] Instead, GC dosages are adjusted based on the patient's symptoms, signs, and general well-being. Overtreatment may be especially challenging to identify, as clinical clues of too-high GC replacement are nonspecific and slow to develop, including weight gain, metabolic syndrome, hypertension, and sleep disturbances.[10]

Despite numerous attempts, measurement of GCs and adrenocorticotropic hormone (ACTH) in blood, urine, and saliva fall short to guide GC replacement.[11] In the last decade, hair cortisol concentration has emerged as a promising tool for assessing GC exposure over time.[9] On the downside, this approach requires a relatively large hair sample (1 cm thick), and is therefore not suited for frequent assessment, and in bald patients, not at all. Although measurement of GC levels in body fluids or tissue will provide information on the bioavailability of the exogenous GC in AAD, it does not reflect the actual physiological effect.[12]

Gene expression is an alternative avenue that allows for evaluation of GC effects at a transcriptional level in the individual patient. Through binding to the GC receptor, GCs regulate the expression of several hundred genes involved in vital physiological processes, including metabolic hemostasis, stress response, and immunity.[13] Previous studies suggest that expression levels of GC-regulated genes align with the reported use and dosage of exogenous GC.[14] Yet, its potential use for guiding GC treatment in AAD remains an unexplored landscape.

We conceived this study in an aspiration to improve health and quality of life in patients with AAD by personalizing GC replacement dosages. In order to foresee positive and adverse effects of different GC dosages, we need access to biomarkers that can act as sensors of GC treatment. Here, we set out to identify candidate GC biomarkers among GC-responsive genes, characterized by change in gene expression levels as a response to different levels of GC exposure.