Endocrine Abstracts

The Endocrine Society guidelines recommend initial testing for Cushing’s syndrome (CS) can be based on non-invasive late-night salivary cortisol measurement (NSC). In the BHSCT NSC (11pm), measured using the IBL ELISA kit has been found to be highly discriminative in identifying patients with CS. However it is a labour intensive test and the need for analysing samples in batches delays turnaround time, limiting its use in the routine work-up for CS. Roche provide an automated competitive electrochemiluminescence immunoassay for salivary cortisol standardised against IRMM/IFCC-451panel using ID-GCMS. The aim of this project was to evaluate Roche’s automated assay for NSC. NSC samples were obtained from 52 patients (8 CS+, 44 CS−). Cortisol was measured in each sample using the ELISA and Roche assays and results correlated. An optimal cut-off for the Roche assay was determined. Between batch imprecision of the Roche assay was 8.7% at 11.5 nmol/l and 5% at 29.2 nmol/l and within batch was <1.95% at 8.0 and 26.7 nmol/l. The assay was shown to be linear to approximately 2 nmol/l. Measurement Uncertainty (MU) was determined at a level of 11.46 nmol/l to be 1.99 nmol/l (9.48–13.45 nmol/l) and for a level of 29.17 nmol/l the MU was 2.97 nmol/l (26.21–32.14 nmol/l). Correlation between test kits was demonstrated with r²=0.933 and y=0.5835x+0.8152. ROC curve analysis (Roche) showed area under curve 0.956 (P <0.001) with an optimal cut-off 7 nmol/l to identify CS (sensitivity 100%, specificity 93.2%). This correlates well with the cut-off provided by Roche of <7.56 nmol/l and <11.3 nmol/l, 95th and 97.5th percentiles respectively. In conclusion the Roche automated assay meets performance requirements and will be introduced into clinical practice. Further evaluation of the diagnostic usefulness of the assay as a routine test is planned. Genomic thyroid hormone action is mediated via receptor subtypes (TRα, TRβ) with differing tissue distributions. Resistance to Thyroid Hormone, due to mutations in thyroid hormone receptor α (RTHα), is characterised by features of hypothyroidism in specific tissues, but near-normal thyroid function tests. Our identification of potentially pathogenic TRα variants in genome databases suggests that RTHα is more common but underdiagnosed. TRα mutants inhibit wild type receptor function via a dominant negative mechanism, involving constitutive inhibition of target gene expression by a mutant receptor-transcriptional corepressor (TR-CoR) complex. Beneficial patient response to thyroxine therapy correlates with mutant TRα whose dysfunction is T3-reversible. Thyroid hormone analogues, which promote dissociation of the TR-CoR complex, may have therapeutic utility in RTHα associated with severe, T3-refractory, receptor defects.