Searchable abstracts of presentations at key conferences in endocrinology
Endocrine Abstracts (2015) 37 GP05.06 | DOI: 10.1530/endoabs.37.GP.05.06

ECE2015 Guided Posters Developmental and paediatric endocrinology (10 abstracts)

Comparison of seven LC–MS/MS methods for the simultaneous measurement of testosterone, androstenedione, and DHEA in serum

Rahel M Büttler 1 , Frans Martens 1 , Flaminia Fanelli 2 , Brian G Keevil 3 , Hai T Pham 4 , Mark M Kushnir 5 , Angela E Taylor 6 , Tue Soeborg 7 , Marinus A Blankenstein 1 & Annemieke C Heijboer 1


1Endocrine Laboratory, Department of Clinical Chemistry, VU University Medical Center, Amsterdam, The Netherlands; 2Endocrinology Unit, Department of Medical and Surgical Sciences, Center for Applied Biomedical Research, S. Orsola Malpighi Hospital, University of Bologna, Bologna, Italy; 3Department of Clinical Biochemistry, University Hospital of South Manchester, Manchester, UK; 4Biocrates Life Sciences AG, Innsbruck, Austria; 5Department of Pathology, ARUP Institute for Clinical and Experimental Pathology, University of Utah, Salt Lake City, Utah, USA; 6School of Clinical and Experimental Medicine, Centre for Endocrinology, Diabetes and Metabolism, University of Birmingham, Birmingham, UK; 7Department of Growth and Reproduction, Faculty of Health and Medical Sciences, Rigshospitalet, Copenhagen University Hospital, University of Copenhagen, Copenhagen, Denmark.


Recently, liquid chromatography–tandem mass spectrometry (LC–MS/MS) was stated to be the method of choice to measure sex steroids.1 Information on the mutual agreement of LC–MS/MS methods, however, is scarce. Therefore, we compared seven published LC–MS/MS methods for the simultaneous measurement of testosterone, androstenedione, and DHEA. Fifty-five random serum samples obtained from volunteers were analysed in duplicate by seven published LC–MS/MS methods.2,3,4,5,6,7 We performed a Passing–Bablok regression analysis and calculated a Pearson’s correlation coefficient to assess the agreement of the investigated methods with one of the methods in this study, arbitrarily chosen as ‘reference method’. Moreover, we calculated the intra-assay coefficient of variation (CV) of each method using the duplicate results. Concentrations of testosterone, androstenedione, and DHEA were 0.2–29, 0.4–5.2, and 1.8–18 nmol/l respectively. The slopes of the regression lines calculated by Passing–Bablok regression analysis ranged from 0.92–1.05 to 1.01–1.15 for testosterone values, for the entire data set and testosterone for concentrations below 2 nmol/l respectively. For androstenedione and DHEA the slopes were 0.96–1.28 and 0.96–1.46 respectively. The correlation coefficients ranged 0.987–0.997, 0.926–0.988, 0.925–0.971, and 0.955–0.988 for all testosterone values, testosterone concentrations below 2 nmol/l, androstenedione, and DHEA respectively. The intra-assay CV were 1.2–6.2, 2.9–10, 2.7–6.9, and 4.3–16% for testosterone values higher than 2 nmol/l, testosterone concentrations below 2 nmol/l, androstenedione, and DHEA respectively. In conclusion, in general the investigated LC–MS/MS methods for simultaneous measurement of testosterone, androstenedione, and DHEA, showed a good agreement. However, there appear to be differences in standardisation between some of the assays and a high variation in some of the assays.

References: 1. Handelsman DJ et al. J Clin Endocrinol Metab 2013.

2. Büttler RM et al. Clin Chim Acta 2014.

3. Fanelli F et al. Steroids 2011.

4. Koal T et al. J Steroid Biochem Biol 2011.

5. Kushnir MM et al. Clin Chem 2010.

6. O’Reilly MW et al. J Clin Endocrinol Metab 2014.

7. Soeborg T et al. Clin Chim Acta 2013.

Article tools

My recent searches

No recent searches.