ORIGINAL ARTICLE Corticosteroid-binding globulin regulates cortisol pharmacokinetics I. Perogamvros*, L. Aarons†, A. G. Miller‡, P. J. Trainer* and D. W. Ray* *Department of Endocrinology, Christie Hospital and Endocrine Sciences Research Group, Manchester Academic Health Science Centre, †School of Pharmacy and Pharmaceutical Sciences, University of Manchester, Manchester and ‡Department of Biochemistry, University Hospital of South Manchester, Wythenshawe, UK Summary Objective Corticosteroid-binding globulin (CBG) is the principal carrier for cortisol in the circulation. Variations in CBG-binding capacity are predicted to alter total serum cortisol disposition, but free serum cortisol is believed to be unaffected. Unbound cortisol pharmacokinetics (PK) have not been studied in the context of CBG changes. We aimed to assess the regulation of cortisol PK by CBG. Design and subjects Women on oestrogens [oral contraceptive pill, (OCP)], patients homozygous for a nonfunctioning CBG vari- ant (CBG null) and healthy controls (HV) were studied before and after IV and oral administration of hydrocortisone 20 mg. Measurements PK parameters were studied for total serum corti- sol (SerF), free serum cortisol (FreeF) and cortisone (FreeE), and salivary cortisol (SalF) and cortisone (SalE): area under the curve (AUC), clearance (CL), half-life and volume of distribution (V d ). Results Following IV hydrocortisone, AUC and half-life of SerF were significantly higher in the OCP group and lower in the CBG null. SerF CL and V d were significantly lower in the OCP group and increased in the CBG null, compared to HV. PK parameters for FreeF and the salivary biomarkers were not different between the CBG null and HV, although OCP patients still had higher AUC compared to HV and prolonged half-life. These findings were con- firmed following oral hydrocortisone, but concentration–time pro- files were highly heterogeneous and SalF interpretation was problematic because of oral contamination. Conclusions We have demonstrated that CBG has a distinct effect on cortisol PK. When CBG binding is disrupted, FreeF retains normal PK characteristics, although CBG null patients lack a CBG-bound pool of readily releasable cortisol. Women on oes- trogens may have altered free serum cortisol kinetics and thus may be potentially overexposed to glucocorticoids. (Received 3 August 2010; returned for revision 16 September 2010; finally revised 29 September 2010; accepted 7 October 2010) Introduction Circulating free cortisol is in dynamic equilibrium with the hor- mone moiety that is bound to plasma proteins, mainly corticoste- roid-binding globulin (CBG) and albumin. CBG has an affinity for cortisol that is several orders of magnitude higher than that of albumin. 1 In addition to its traditional role as a carrier molecule, CBG may also be a dynamic regulator of glucocorticoid bioavail- ability. It provides a pool of circulating cortisol, available to be released both systemically and at a tissue level, which is typified by its function as a ‘negative’ acute phase protein during acute inflammation 2,3 and the interaction with proteinases, such as neu- trophil elastase, which result in the tissue-specific release of corti- sol. 4,5 Variations in CBG binding capacity are relatively common and most frequently result from acquired disorders of the liver and kid- ney, 6 as well as commonly prescribed medications, most notably oestrogens. 7 Genetic CBG defects are rarely described and include mutations that cause undetectable circulating levels or binding capacity of CBG. 8,9 Changes in circulating CBG have a significant impact on total serum cortisol levels, but free serum (FreeF) or sali- vary cortisol (SalF) are believed to be unaffected and they could thus provide a more physiological biomarker for glucocorticoid availability. 6,10 Therefore, accurate measurement of unbound glucocorticoids in serum or saliva is a potentially useful means of diagnosing hypoadrenalism, 11 and also monitoring glucocorticoid replacement. 12 Routine assessment of the hypothalamic–pituitary–adrenal (HPA) axis involves total serum cortisol measurement and current practice indicates that patients on oestrogens may have to stop their treatment for 6 weeks before their serum cortisol is measured. It is thus assumed that the CBG-specific increase in measured corti- sol in these patients is a practical inconvenience rather than an indication of glucocorticoid overexposure. Conversely, low total serum cortisol concentration in the absence of CBG binding may Correspondence: Prof. David W. Ray, Endocrine Sciences Research Group, Department of Endocrinology, Manchester Academic Health Science Cen- tre, The University of Manchester, 3rd Floor AV Hill Building, Oxford Road, Manchester M13 9PT, UK. Tel.: +44 1612755655; Fax: +44 1612755958; E-mail: david.w.ray@manchester.ac.uk Clinical Endocrinology (2011) 74, 30–36 doi: 10.1111/j.1365-2265.2010.03897.x 30 Ó 2010 Blackwell Publishing Ltd