Howard Horng (Presenter)
University of California, San Francisco
Bio: As a scientist in both academia and industry, I am experienced in analytical chemistry and pharmaceutical sciences. My graduate studies at UCSF, focused on The Bioactivation of Carboxylic Acid Drugs, which are associated with toxicities and one hypothesis is the formation of reactive metabolites. As a postdoctoral fellow for the FDA, I focused on determining whether in vitro measures can be used to obviate the need for PK studies in patients with renal impairment for drugs primarily eliminated by the liver. Being the director of the Hair Analytical Lab (HAL), I manage all hair studies, including the development, validation, and running therapeutic drug monitoring studies for antiretrovirals (ARVs).
Authorship: Howard Horng, Monica Gandhi, Alan HB Wu, and Kara Lynch
University of California, San Francisco(UCSF)/San Francisco General Hospital (SFGH)
| Short Abstract Measuring hair cortisol levels in humans to determine long term systemic exposure is increasingly being used as a biomarker of chronic stress. Since human hair has a fairly uniform growth rate (~1 cm/month), one can investigate periods of time in which stress is absent, present, or most pertinent. The objective of this study was to develop and validate an LC-MS/MS method for the measurement of cortisol in hair and compare the method to the Siemens Centaur cortisol competitive immunoassay. |
Long Abstract
Introduction
Measuring hair cortisol levels in humans to determine long term systemic exposure is increasingly being used as a biomarker of chronic stress. Human hair has historically been used to monitor exposure to various xenobiotics and reflects uptake of exogenous and endogenous compounds from systemic circulation over extended periods of time. As a result, there are a number of advantages of measuring hair cortisol levels as a biomarker of chronic stress. Since human hair has a fairly uniform growth rate (~1 cm/month), one can investigate periods of time in which stress is absent, present, or most pertinent. However, the means to which various laboratories measure hair cortisol levels, immunoassay versus liquid chromatography-mass spectrometry (LC-MS/MS), has been shown to yield different results. The objective of this study was to develop and validate an LC-MS/MS method for the measurement of cortisol in hair and compare the method to the Siemens Centaur cortisol competitive immunoassay.
Methods
Sample preparation consisted of finely cutting human hair into pieces followed by pulverization using the BioTage Bead Ruptor 24. The resulting powder was then weighed (~50 mg), placed into a culture tube, and incubated in methanol (1.0 ml) with cortisol-d4 (internal standard (IS)) at 37oC overnight in a shaking water bath. The resulting supernatant was evaporated to dryness and reconstituted with an aqueous solution containing 0.1% formic acid (1.0 ml), followed by extraction with ethyl acetate (3.0 ml). The organic layer was then transferred to another culture tube followed by evaporation to dryness. Samples to be analyzed via immunoassay and LC-MS/MS were reconstituted with saline (250 µl) and methanol (100µl), respectively, and injected for analysis. LC-MS/MS analysis were carried out using an Agilent 1200 HPLC system coupled to an AB Sciex 3200 QTRAP triple quadrupole mass spectrometer outfitted with electrospray (ESI) positive ionization. Liquid chromatography conditions utilized a reverse phase column (Thermo C-18, 5 µm, 4.6 x 150 mm) and a gradient system of aqueous solution (0.1% formic acid) and acetonitrile (0.1% formic), 10% ACN to 90%, over 10 minutes at a flow rate of 0.800 ml/min. Cortisol quantitation was performed via multiple reaction monitoring (MRM) using the mass transitions MH+ m/z 363 to m/z 121 and MH+ m/z 367 to m/z 121 for cortisol and cortisol-d4, respectively. Standard curve and quality control samples were spiked with known amounts of cortisol in synthetic blank hair. The linear concentration range for cortisol was 5.00 to 2000 ng of cortisol/g hair and three quality control samples were also spiked with the following concentrations: 1600 ng/g (high), 200 ng/g (medium), and 40 ng/g (low). Untargeted analysis was carried out using a Shimadzu LC-20AD HPLC coupled to a AB Sciex Triple TOF 5600 high resolution mass spectrometry (HRMS) utilizing ESI positive ionization. HRMS analysis was performed on a reverse phase column (Phenomenex Kinetex, 2.6 µm, 50 x 3.0 mm) and a gradient system of 5.0 mM ammonium formate/0.050% formic acid in water (A) and 50% acetonitrile/50% methanol ( B) over 14 minutes at a flow rate of 0.4 ml/min. Data was collected in full mass scan mode (MH+ m/z 50 to 700) with IDA-triggered acquisition of product ion scan. For immunoassay analysis, reconstituted hair samples were injected onto a Siemens Centaur cortisol competitive immunoassay utilizing direct chemiluminescent technology.
Results
Cortisol concentrations from 10 human subjects were analyzed for comparison. Hair cortisol concentrations for each subject determined via LC-MS/MS were 13.44, 22.85, 20.27, 4.32, 17.40, 14.52, 39.15, 7.68, 20.3, and 14.21 ng/g, while the cortisol values determined via immunoassay for the same subjects were 40.55, 22.30, 28.34, < 21.05, 23.16, 28.22, 35.13, 28.08, <21.05, 27.57 ng/g, respectively. In addition, the concentration of the spiked quality control samples were also measured using the two instruments yielding a concentration of 1466.00 and 1587.62 (high), 187.40 and 198.45 (medium), and 43.20 and 39.69 (low) ng/g for the LC-MS/MS and immunoassay respectively. Precision, defined as percent coefficient of variation (%CV), and accuracy, defined as percent relative error, of the standard curve and quality control samples were all within 15%. The lower limit of cortisol quantitation via LC-MS/MS and immunoassay are 5 and 21.05 ng/g, respectively.
Conclusion
The comparison of human hair cortisol concentrations using an immunoassay and LC-MS/MS resulted in different results. The values determined via immunoassay were higher than those derived from LC-MS/MS. The cortisol levels derived from the immunoassay in the human subjects experienced an average 1.90-fold increase compared to those values derived from LC-MS/MS. In addition, the interindividual increase in values was highly variable (46.50 %CV). The quality control sample created from spiking cortisol in synthetic blank resulted in an average 1.02-fold increase in value and a smaller interindividual variation (8.69 %CV). Untargeted analysis comparing real human hair to synthetic blank hair utilizing HRMS revealed the presence of a number of endogenous compounds at concentrations higher than that of the synthetic solution, potentially resulting in matrix effects and cross reactivity. This discrepancy may be the reason why the variation in quality control samples is much lower than that of the human hair and the high variability in cortisol levels derived from the immune assay versus LC-MS/MS in human subjects. Ultimately, these results suggest that a correction factor would be insufficient to correlate cortisol hair values obtain via immunoassay and LC-MS/MS.
References & Acknowledgements:
| Description | Y/N | Source |
| Grants | no | |
| Salary | no | |
| Board Member | no | |
| Stock | no | |
| Expenses | no |
IP Royalty: no
| Planning to mention or discuss specific products or technology of the company(ies) listed above: | no |