Hair Cortisol and Cushing's Disease

Selective surgical removal of the pituitary corticotroph adenoma is the ideal treatment of Cushing's disease. However, surgery may not be feasible or is unsuccessful in roughly 25% of patients. In addition, a recurrence of the disease is observed after a transient remission in 15 to 25% of patients. Several therapeutic alternatives are available, amongst which medical treatment is commonly used. Drugs that are available in France to control hypercortisolism target the pituitary adenoma secretion (pasireotide and cabergoline) or inhibit adrenal steroidogenesis (ketoconazole and metyrapone). Usual criteria to monitor the treatment and titrate the drug dosage include evaluation of relevant clinical endpoints and measurement of UFC. However, limitations of UFC for this purpose, include difficulties in obtaining a complete 24 urine collection and the fact that UFC assess only short-term cortisol in a disease characterized by high variability over time in the intensity of hypercortisolism. Elsewhere, a mild to moderate hypercortisolism may persist despite a normal UFC. Several groups, including ours, have shown that LNSC is useful tool to diagnose overt and mild hypercortisolism and may be more sensitive than UFC to diagnose mild hypercortisolism. Despite being more convenient to collect than 24h urine, LNSC also suffers from only measuring time-point cortisol. Rare studies have examined whether LNSC could be an adequate biomarker for monitoring response to medical therapy in patients with CD and its usefulness to monitor drug treatment in CD is yet unknown (one study comparing UFC to LNSC in CD patients treated with pasireotide-LAR has been presented but is not yet published). Accordingly, the 2015 guidelines of the endocrine society recommend future research to accurately monitor patients for their response to medical therapy to guide dose optimization. More recently, the measurement of salivary cortisone in 3 saliva samples (SCx3) drawn at approximately 8 hours intervals and starting at 7-8 am have been shown to be a reliable estimate of cortisol production over 24h in normal subjects. Whether this parameter could be used as a substitute of UFC in patients treated with anticortisolic drugs remains unstudied.

Hair cortisol concentration is a non-invasive way to measure cortisol exposure over much longer periods of time (weeks and months) than previously possible with samples of blood, saliva or urine. Several studies have shown that measurement of cortisol in a single scalp hair sample has a diagnostic accuracy for CS similar to currently used first-line tests and may also be used to identify overtreatment in patients receiving hydrocortisone replacement for adrenal insufficiency. To date, no data is available concerning hair Cortisol measurement in comparison with other usual biological tools in patients with CD receiving a medical treatment.

Since 2016, the departments of endocrine biology and clinical endocrinology of Bordeaux university hospital (CHU) have developed the measurement of cortisol and cortisone in scalp hair using LC-MS/MS and have established normative values using several cohorts of control patients.

The purpose of the study is to take advantage of the ability of hair cortisol to measure long-term cortisol exposure to assess the response to medical therapy in patients with CD. More specifically, the working hypothesis is that some patients with a normal UFC may still suffer from an occult mild hypercortisolism that will be identified only by hair cortisol. To study this hypothesis, the investigators will compare the results of hair cortisol to that multiple measurement of UFC, LNSC and SCx3 during a three-month period in patients with CD already treated with medical treatments and considered as "controlled" on the basis of previous UFC measurements.