Grace van der Gugten (Presenter)
St. Paul's Hospital
Authorship: J. Grace van der Gugten(1), Judy Stone(2), Deborah French(3), Julia C Drees(4)
(1)St. Paul’s Hospital, Department of Pathology and Laboratory Medicine, Vancouver, Canada (2)University of California at San Diego, California (3)University of California at San Francisco California (4)Kaiser Permanente Northern California Regional Laboratory, Berkeley California
| Short Abstract Accurate and reliable measurement of testosterone, particularly at low levels, is clinically important. It is well known that immunoassays perform poorly at low levels when compared to mass spectrometry. LC-MS/MS may be thought of as a “gold standard”, but inter-method biases seen on external quality assurance reports and published reports of interferences prove that LC-MS/MS is by no means infallible. In a manner similar to a recent comparison of published steroid methods, we have undertaken an interlaboratory comparison of testosterone LC-MS/MS assays between 4 laboratories. |
Long Abstract
Serum testosterone levels are essential for diagnosis and/or monitoring of a number of conditions including male hypogandism, polycystic ovarian syndrome, and congenital adrenal hyperplasia [1]. Therefore, accurate and reliable measurement of testosterone, particularly at low levels is clinically important.
It is well known that immunoassays perform poorly at low levels when compared to mass spectrometry [2]. LC-MS/MS may be thought of as a “gold standard” for laboratory steroid analysis, and accordingly, many laboratories have developed in-house methods using this technology. LC-MS/MS is by no means infallible, though, as evidenced by high inter-laboratory biases seen on external quality assurance (EQA) reports and published reports of interferences.
Data from the New York State Proficiency Testing Program for Testosterone run by LC-MS/MS methods in 2006 show inter-laboratory CV’s of up to 20%. Results for this same EQA program in 2011, 1 year after the Centres for Disease Control (CDC) Hormone Standardization Program (HoST) was implemented, show much improved inter-laboratory CV’s of less than 5% [3]. Ideally, all laboratories performing testosterone analysis would take part in the CDC HoST program, but this program is expensive. Standardized reference materials (SRM) are available for testosterone from NIST (SRM 971) as a less expensive alternative to the CDC HoST program as a way to assess in house calibrator accuracy. As well, some external quality control providers have certificates of analysis for their control materials such as the Referenzinstitut für Bioanalytik DGKL program.
In a manner similar to a recent comparison of published steroid methods [4], we compared testosterone concentrations between 4 laboratories. Discarded anonymized aliquots from 102 serum samples previously analysed for Testosterone in Lab_SPH were split, frozen and sent to each participating lab overnight on dry ice.
Method information is summarized as follows:
Lab_SPH: Semi-automated liquid liquid extraction (LLE) in 96 well plates; Shimadzu-SCIEX API5000 LC-MS/MS with a 6 minute run time.
Lab_UCSF: Manual LLE in tubes; Shimadzu-SCIEX 6500 LC-MS/MS with a 7 minute run time.
Lab_Kaiser: Semi-automated LLE in 96 well plates; Shimadzu-SCIEX 5500 LC-MS/MS with a 7 minute run time.
Lab_UCSD: Automated TECAN AC 96-well plate extraction; Waters Acquity-XEVO-TQS LC-MS/MS with a 5.44 minute run time.
The results show excellent agreement between the four laboratories. The results from each laboratory were plotted against the all-method mean for each sample (AMM):
Lab_SPH vs. AMM: slope: 1.005 mean bias: -0.4% median difference: +0.07%
Lab_UCSF vs. AMM: slope: 0.965 mean bias: -3.4% median difference: -2.88%
Lab_Kaiser vs. AMM: slope: 0.993 mean bias: +0.4% median difference: -0.25%
Lab_UCSD vs. AMM: slope: 1.050 mean bias: +3.4% median difference: +4.16%
We will present the results of our Testosterone 4-way inter-laboratory comparison as well as discuss observed biases seen and strategies for harmonization.
References & Acknowledgements:
[1] Vesper HW et al. Interlaboratory comparison study of serum total testoserone measurements performed by mass spectrometry methods. Steroids 2009; 74: 498-503
[2] Herold DA, Fitzgerald RL. Immunoassays for Testosterone in Women: Better Than a Guess? Clin Chem 2003;49:1250-51
[3] Vesper HW, Botelho JC. Testosterone: An Overview of CDC’s Standardization Initiative. AACC Clinical Laboratory News June2012
https://www.aacc.org/publications/cln/articles/2012/june/testosterone
[4] Buttler, RM et al. Comparison of 7 Published LC-MS/MS Methods for the Simultaneous Measurement of Testosterone, Androstenedione, and Dehydroepiandrosterone in Serum. Clin Chem 2015; 61:12: 1475–1483.
| Description | Y/N | Source |
| Grants | no | |
| Salary | no | |
| Board Member | yes | Sci Comm for MSACL EU - for Troubleshooting Forum |
| Stock | no | |
| Expenses | no |
IP Royalty: no
| Planning to mention or discuss specific products or technology of the company(ies) listed above: | no |