Joshua Buse (Presenter)
University of Calgary
Bio: Dr. Joshua Buse is the first year clinical biochemistry fellow at the University of Calgary and previously was employed as a Laboratory Scientist at Calgary Laboratory Services. His work involved in the development of liquid chromatography-tandem mass spectrometry methods for drugs of abuse, therapeutic drug monitoring and endocrine related compounds. His previous educational experience includes 1) post-doctoral fellowship in structural proteomics at the University of Calgary as well as doctoral studies in pharmacy and undergraduate studies in biochemistry at the University of Saskatchewan. Over this period of time, he has contributed to over twenty-five conference abstracts and seven journal articles as well as received the Canadian Society of Mass Spectrometry’s annual award for best dissertation or thesis involving mass spectrometry in 2014.
Authorship: Joshua D. Buse (1), J. Grace Van Der Gugten (2), Daniel T. Holmes (2), Alex C. Chin (1), S.M. Hossein Sadrzadeh (1), Gregory A. Kline (3)
1) Department of Pathology and Laboratory Medicine Faculty of Medicine University of Calgary 2) St. Paul's Hospital Department of Pathology and Laboratory Medicine Vancouver, BC, V6Z 1Y6 Canada 3) Department of Endocrinology Faculty of Medicine University of Calgary
| Short Abstract Immunoassays (IAs) are invaluable for routine use, but suffer from lower specificity compared to LC/MS-MS. Here we present a case report of a 30 year old woman with no clinical evidence of Cushing Syndrome who had consistent and marked elevation of salivary cortisol by IA, but indeterminate tests for elevation of plasma cortisol. Intensive investigation by LC-MS/MS determined that cortisone was present at supraphysiological concentration and prednisone was present in very high concentrations in the patient’s salivary sample, explaining the high IA results. The limitations of test methods to report false positive results due to interferences can result in erroneous diagnoses or exploited by patients with Munchausen syndrome. |
Long Abstract
Introduction
Clinical chemistry laboratories routinely use immunoassays (IAs) for the analysis of patient specimens because of their high-throughput, cost efficiency and accuracy. However, the ability of the result to accurately present the biological picture is dependent upon the maximization of specimen integrity and minimization of interferences. The increased capacity of interferences to influence results obtained from immunoassays has led to an increasing prevalence of mass spectrometry in the laboratory. Furthermore, the specificity and sensitivity provided by liquid chromatography-tandem mass spectrometry (LC-MS/MS) has made it the gold standard for the analysis of small molecules, including steroid molecules.
Case Presentation
A 30 year old woman presented to her family physician with some non-specific complaints of weight gain and fatigue. Cushing Syndrome was considered by her primary care physician and after the finding of borderline high urine cortisol levels, she was subsequent referred to several endocrinologists. Extensive clinical workup lead toward a diagnosis of cyclical Cushing’s Syndrome or 11-betahydroxysteroid dehydrogenase deficiency type 1 due to the extreme elevation of salivary cortisol levels (300 - 400 nmol/l, ref <3.6) and the non-suppressed Adrenocorticotropic hormone (ACTH) in plasma. However, other clinical features of hypercortisolism were not observed despite several years’ follow up and there was no mutation in the gene coding for the glucocorticoid receptor. In addition, inconsistencies in the clinical picture led the endocrinologist to request that salivary and serum samples to undergo LC-MS/MS investigation for cortisol with a hope to reduce/eliminate the possibility of analytical interference.
Methods
Four midnight salivary specimens collected by the patient over the course of six months underwent analysis; two specimens that were self-collected involved the deposition of saliva in false bottom tubes that were shipped to the laboratory on ice. Patient specimens were directly analyzed simultaneously on both the Roche Cortisol I and Cortisol II assays using a Roche Cobas 8000 instrument (Roche Diagnostics GmBH, Mannheim, Germany) used in accordance with manufacturer’s specifications and monitored daily by analysis of three levels of commercially available quality control material (Bio-Rad Immunoassay Plus, Lot 40880, expiry date 28-Feb-2017).
Subsequent LC-MS/MS analysis required preparation of the specimen prior to analysis. A 200 µL aliquot of salivary sample was added to 50 µL of the internal standard mixture (Cortisol-D4, 17-hydroxyprogesterone-D8, 11-deoxycoritsol-D5, 21-deoxycortisol-D8, androstene-3,17-dione-13C3, each at 5 µg/L) in 2.0 mL Eppendorf microcentrifuge tubes. A 1000 µL aliquot of Methyl-tert-butyl-ether was added to each sample, which was vortexed for 5 minutes and subsequently centrifuged at 12000 x g for 10 minutes at 20 °C. An 800 µL aliquot of the organic phase was placed into a clean glass test tube and evaporated under a stream of nitrogen at ambient temperature until dry. Each sample was reconstituted in 100 µL of acetonitrile containing 50% ammonium formate buffer (5 mM at pH 4.0), with 5 µL of sample being injected into the LC-MS/MS system.
The LC-MS/MS system consisted of an Agilent series 1290 binary pump with an online degasser, auto sampler and column oven (Agilent Technologies, Mississauga, ON, Canada) coupled to a 6460 triple quadrupole MS/MS via a JetStream(R) electrospray interface. Chromatographic separation of the analytes was achieved by using a Restek Raptor Biphenyl column (2.7 µm particles, 100 x 2.1 mm) protected by a Phenomenex HPLC KrudKatcher Ultra Column In-line filter. The mass spectrometry grade acetonitrile mobile phase contained 0.01% formic acid, while the 5 mM ammonium formate buffer (pH 4.0) was made using Millipore water. The six and a half minute chromatographic method employed a gradient flow of 500 μL/min, with starting conditions of 80% 5 mM ammonium formate buffer (pH 4.0) and ending conditions of 100% acetonitrile containing 0.01% formic acid. The column was maintained at a temperature of 50 °C during the run.
The Agilent 6460 triple quadrupole MS/MS utilized a nebulizer flow of 35 psi and a sheath gas flow of 11 psi, with the sheath gas temperature of 250 °C. The capillary voltage was set at 3000 V. The mass spectrometer utilized multiple reaction monitoring run in positive mode to monitor the 14 analytes (Table 1); each MRM transition monitored relied upon a dwell time of 20 ms. Subsequent untargeted MS and MS/MS analysis utilized the same LC-MS parameters as above; with the exception of modifications to the LC gradient and MS collision energy being operated at 0, 15, 30, 45 and 60 V. Data evaluation was performed using the Agilent Mass Hunter Software (B 07.01).
Table 1 – MS/MS parameters utilized for targeted analysis of 9 steroid analytes
Analyte Prec. Ion (m/z) Prod. Ion (m/z) Frag [V] (CE [V])
Cortisol(-D4) 363 (367) 121, 91 (331, 121, 91) 105 (24, 60, 16)
Cortisone 361 162, 121 120 (20, 32)
21-Deoxycortisol(-D8) 347 (355) 311, 125 (319, 125) 97 (14, 30)
11-Deoxycortisol(-D5) 347 (352) 109, 97 (113, 100) 95 ( 28, 28)
17-Hydroxyprogesterone(-D8) 331 (339) 109, 97 (113, 100) 100 (27, 31)
Pregnenolone 317 159, 112 102 (4, 16)
Androstenedione(-13C3) 287 (290) 109, 97 (112, 100) 100 (17, 25)
DHEA 289 271, 231 95 (2, 20)
Aldosterone 361 332, 190 105 (6, 14)
Quantification of cortisone and confirmation of prednisolone was performed on the specimen through extraction with supported liquid extraction on Biotage Isolute SLE400+ plates and eluted with methyltertbutyl ether as described in [1]. Cortisone quantification relied upon the MRMs 361.23 → 163.0 and 361.23 → 121.1 in positive ESI mode [2], while prednisone was detected (M+1 = 359.1) and analyzed by product ion scanning and compared to a solution of prednisone in solution (Sigma, Oakville, ON).
Results
Elevated concentrations of cortisol were detected in two midnight salivary samples provided by the patient when measured by the Roche Cortisol I (323.2, 222.4 nmol/L) and Cortisol II (4101, 2243 nmol/L) (by dilution) immunoassays. The discontinuity between the grossly elevated salivary cortisol concentrations and the absence of cushingoid symptoms led to investigation of both the patient’s and her two children's salivary samples by LC-MS/MS. Cortisol was absent in the patient’s salivary sample and measured at normal levels in her children's when analyzed by LC-MS/MS. However, a large peak with a shoulder was detected in the patient’s cortisone and aldosterone MRM chromatograms. Although this peak was an inappropriate match for cortisone, incorrect MRM ratio, and aldosterone, incorrect retention time as well as MRM ratio, its mass to charge (m/z) of 361 and similarities it held in its fragment ions in comparison to cortisone and aldosterone warranted its further evaluation. Chromatographic separation of the peaks with both MS analysis and MS/MS analysis with ramping of the collision energy produced MS/MS spectra that were compared to the MassBank database. Cortisone (m/z 361) and prednisone (m/z 359) were determined to be the respective matches for the peaks; additional full spectrum analysis led to discovery of a potential prednisone metabolite. Further analysis with standards for both cortisone and prednisone confirmed their presence in the patient’s salivary specimens, with quantitative analysis of cortisone in one specimen yielding a concentration of 137.6 nmol/L.
Discussion
The susceptibility of the cortisol immunoassays to interference from cortisone and prednisone led to the reporting of multiple falsely elevated salivary cortisol levels. Roche’s cortisol I and cortisol II cross relativities for cortisone (0.3% and 6.6% at 10 µg/mL; respectively) and prednisone (0.3% and 2.2% at 10 µg/mL; respectively) demonstrated how salivary cortisol levels could be falsely elevated. Although the salivary cortisol levels did not correlate with the patient’s clinical presentation, extensive investigations were undertaken in attempts to identify a unifying diagnosis – such as cyclical Cushing’s Syndrome or 11-betahydroxysteroid dehydrogenase deficiency type 1. Clinical and biochemical inconsistency with Cushing Syndrome led to the request for cortisol quantification by LC-MS/MS analysis on salivary and serum specimens. The absence of cortisol with the salivary specimens motivated the use of both untargeted and targeted analysis of salivary specimens, which led to the discovery of cortisone and prednisone. The absence of cortisol and prednisolone in the salivary specimens, a non-suppressed ACTH plasma concentration and confirmation of saliva in the specimen conclusively demonstrated that the salivary specimens were adulterated through direct addition of the prednisone and cortisone contaminants. In self-collected specimens physicians should be aware of the higher potential for nefarious alteration of the specimen and results. In this case, the LC-MS/MS analysis yielded conclusive results for the caring physicians to properly diagnosis and provide treatment to their patient; the current working diagnosis is Munchausen’s Syndrome.
References & Acknowledgements:
References
1. Van Der Gugten JG, Holmes DT. Quantitation of Aldosterone in Serum or Plasma Using Liquid Chromatography-Tandem Mass Spectrometry (LC-MS/MS). Clinical Applications of Mass Spectrometry in Biomolecular Analysis: Methods and Protocols. 2016:37-46.
2. Perogamvros I, Owen LJ, Newell-Price J, Ray DW, Trainer PJ, Keevil BG. Simultaneous measurement of cortisol and cortisone in human saliva using liquid chromatography–tandem mass spectrometry: application in basal and stimulated conditions. Journal of Chromatography B. 2009 Nov 1;877(29):3771-5.
| 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 |