Background
Liquid chromatography-tandem mass spectrometry (LC-MS/MS) is increasingly becoming the method of choice in the clinical laboratory for the measurement of low molecular weight analytes. The major advantage that LC-MS/MS possesses relative to conventional laboratory techniques such as immunoassay is its higher specificity (and often sensitivity, although this is compound specific) and its ability to measure multiple compounds in a single run (multiplexing). LC-MS/MS thus provides the opportunity for more accurate and precise biochemical diagnosis and monitoring of human disease. One example of the increasing adoption of LC-MS/MS by clinical laboratories is the measurement of steroid hormones in various matrices (serum, saliva, urine).
Steroid metabolism
All steroids share a cyclopentanoperhydrophenanthrene nucleus, with individual species varying according to the presence of different functional groups attached to this four-ring structure, as well as by the oxidation state of the rings. Cortisol structure is given as an example in Figure 1. In humans, the major sites of steroid hormone production are the adrenal gland and the gonads. Steroids are synthesized from cholesterol via a series of enzyme-catalysed steps (Fig. 2), which are under tight regulation in healthy individuals by feedback mechanisms involving the hypothalamus and anterior pituitary. Steroids have a wide range of physiological functions which are summarized in Table 1.
Adrenocortical carcinoma – a diagnostic challenge
There are many endocrine disorders that result in the improper synthesis of steroids, and one of the rarest and most severe is adrenocortical carcinoma (ACC). ACC is a malignancy of the adrenal cortex with an annual incidence of 1 or 2 cases per million [1]. The majority of ACC cases are sporadic and occur in the fifth or sixth decade of life and more commonly in women; although ACC can be associated with several familial syndromes including Li-Fraumeni, Beckwith-Wiedemann, Lynch syndrome and multiple endocrine neoplasia type 1 [2]. Functional steroid hormone-producing tumours occur in around two-thirds of cases [3], presenting with varied signs and symptoms of steroid overproduction, most commonly Cushing’s syndrome (cortisol excess) and hyperandrogenism. ACC can progress rapidly in some patients, therefore it is vital that it is distinguished from benign adrenal adenomas, as ACC has a 5-year survival rate of <50% [2]. A surgical cure is only possible if the carcinoma is detected in its localized stage, otherwise the median survival period is <15 months [4].
The diagnosis of ACC is challenging as there is no single diagnostic tool that is able to distinguish ACC from other adrenal masses, including benign adenomas with glucocorticoid or mineralocorticoid excess, phaeochromocytoma and non-functioning adenomas. Imaging alone is insufficient for diagnosis, as although patients with ACC almost always present with tumours ≥4 cm, the presence of a large mass only has a clinical specificity of 61% [5]. Additionally, whereas up to two-thirds of tumours are functional, less than half of ACC cases present with clinical signs of steroid overproduction [3], with a further proportion presenting with other symptoms including abdominal pain. However, a significant proportion are discovered incidentally [2].
The European Network for the Study of Adrenal Tumours (ENSAT) currently recommends that the initial biochemical work-up for suspected ACC includes measurement of serum cortisol (both basal and assessment of suppression after dexamethasone), dehydroepiandrostenedione sulphate (DHEAS), androstenedione, testosterone, 17-hydroxyprogesterone, estradiol and aldosterone (if the patient is hypokalemic or hypertensive). An alternative approach is to measure steroid metabolites in urine using gas chromatography-mass spectrometry (GC-MS); increases in the excretion of metabolites of the steroid precursors 11-deoxycortisol, 17-hydroxypregnenolone and pregnenolone have been shown to provide particularly high diagnostic utility in ACC. Unfortunately, urine steroid profiling is not commonly available in clinical laboratories owing to lengthy sample preparation and complex result interpretation. Further, serum 11-deoxycortisol, 17-hydroxypregnenolone or pregnenolone measurements are rarely performed either because of lack of demand, or specificity of the available immunoassays which may be subject to significant levels of cross-reactivity.
As a result of these limitations, the use of LC-MS/MS is increasingly being adopted to provide more specific steroid hormone measurements. An approach we have taken in our laboratory is to develop and fully evaluate a multiplexed LC-MS/MS method panelling 13 steroids in serum [6] to include many of the steroid synthetic pathway intermediates currently not available for ACC work-up.
Use of a serum steroid panel
The steroids included in our serum panel are highlighted in Figure 2 and are as follows:
- androstenedione
- corticosterone
- cortisol
- cortisone
- 11-deoxycorticosterone
- 11-deoxycortisol
- 21-deoxycortisol
- DHEAS
- 17-hydroxypregnenolone
- 17-hydroxyprogesterone
- pregnenolone
- progesterone
- testosterone.
Samples are prepared for analysis by an initial protein precipitation step to remove steroids from their binding proteins, followed by liquid-liquid extraction in order to cleanly extract the steroids from remaining matrix components. Prepared extracts are then analysed by LC-MS/MS in which steroids are first resolved on a reverse phase C18 column by gradient elution followed by MS/MS detection using positive atmospheric pressure chemical ionization (APCI) operated in multiple reaction monitoring mode. Chromatographic separation of several isobaric (same mass to charge ratio) steroids is essential, as is the use of deuterated internal standards for all steroids in the method.
When we applied our method to adrenal tumour samples [6], we were able to show that between 4 and 7 steroids were elevated in all ACC cases in comparison to non-ACC adrenal tumours where a maximum of 1–2 steroids were abnormal. The cortisol precursor 11-deoxycortisol was most useful in the discrimination between ACC and non-ACC adrenal lesions, whereas other steroids markedly elevated in ACC included 17-hydroxypregnenolone and pregnenolone. Indeed, all steroids except testosterone in males and corticosterone and cortisone in both sexes were of use in discriminating ACC. This validates the use of a panelling approach when investigating adrenal masses.
Our findings compare well with urine steroid profiling studies. Although urine steroid profiling using 24-hour collections may offer greater clinical sensitivity compared to a single blood measurement owing to diurnal rhythms of steroid production, urine measurements rely on accurately timed collections that are often performed incorrectly and are inconvenient to the patient. Advantages of our LC-MS/MS serum panel compared to urine steroid profiling by GC-MS include a less labour intensive sample preparation, as well as less expertise required for the interpretation of complex profiles, as the serum method only targets selected steroids rather than the large number of their metabolites in urine.
Use of our LC-MS/MS serum steroid panel in ACC patients has further demonstrated the limitations of assessing serum steroids by immunoassay. We observed evidence of notable interference in ACC patients in the cortisol, progesterone, 17-hydroxyprogesterone and androstenedione immunoassays, inferred to be due to elevated concentrations of structurally related steroid precursors.
Future work
Currently, our 13-steroid serum panel has been used to study a relatively small number of ACC patients (because of the rarity of the disease), and clearly larger prospective studies are required to more fully determine the diagnostic utility of our panel in ACC. Further work is also required to clarify the effects of age, sex and diurnal variation on serum steroid panelling; nonetheless the most useful markers of ACC are markedly elevated above variation attributable to these biological factors. In addition to the complexity of interpreting biomarker panels, it is not only important to consider specific reference ranges, but to also consider the patterns in results which require an omics-based analysis approach to interpretation. The challenge surrounding this, as well as the requirement for clear presentation and reporting of results to clinicians requires close involvement of clinical colleagues for the development and introduction of such testing strategies.
The analysis of steroid panels by LC-MS/MS can also undoubtedly be used in other conditions including inborn errors of steroid metabolism such as congenital adrenal hyperplasia (CAH) and polycystic ovarian syndrome (PCOS).
Although we have demonstrated the advantages of our LC-MS/MS steroid panel compared to routine immunoassays, there are undoubtedly disadvantages of using LC-MS/MS. These include the initial cost of instrument purchase, the increased expertise required and often a more laborious sample preparation. Additionally, the specificity of mass spectrometry should not be readily assumed; careful selection of multiple reaction monitoring (MRM) transitions and chromatography conditions are essential to separate isobaric steroids and other interfering compounds. However, in the context of improving the biochemical tools available to us to aid the diagnosis of ACC, the advantages of LC-MS/MS far outweigh these limitations.
Summary
In summary, LC-MS/MS serum steroid panelling offers an additional tool for the challenge that is the diagnosis of ACC. Our method combines measurement of both common and rarely measured steroids in a single sample, which we have shown provides useful data to aid the discrimination of ACC from benign adrenal tumours. Use of LC-MS/MS gives several advantages over the immunoassay and GC-MS-based methods currently used to assess steroid overproduction, but further work is required to demonstrate the full potential of its use in the diagnosis of ACC.
References
1. Fassnacht M, Kroiss M, Allolio B. Update in adrenocortical carcinoma. J Clin Endocrinol Metab 2013; 98: 4551–4564.
2. Else T, Kim AC, Sabolch A, Ramond VM, Kandathil A, Caoili EM, Jolly S, Miller BS, Giordano TJ, Hammer GD. Adrenocortical carcinoma. Endocr Rev 2014; 35: 282–326.
3. Arlt W, Biehl M, Taylor AE, Hahner S, Libé R, Hughes BA, Schneider P, Smith DJ, Stiekema H, et al. Urine steroid metabolomics as a biomarker tool for detecting malignancy in adrenal tumours. J Clin Endocrinol Metab 2011; 96: 3775–3784.
4. Fassnacht M, Terzolo M, Allolio B, Baudin E, Haak H, Berruti A, Welin S, Schade-Brittinger C, Lacroix A, et al. Combination chemotherapy in advanced adrenocortical carcinoma. N Engl J Med 2012;366:2189–2197.
5. Hamrahian AH, Ioachimescu AG, Remer EM, Motta-Ramirez G, Bogabathina H, Levin HS, Reddy S, Gill IS, Siperstein A, Bravo EL. Clinical utility of noncontrast computed tomography attenuation value (Hounsfield units) to differentiate adrenal adenomas/hyperplasias from nonadenomas: Cleveland Clinical experience. J Clin Endocrinol Metab 2005; 90: 871–877.
6. Taylor DR, Ghataore L, Couchman L, Vincent RP, Whitelaw B, Lewis D, Diaz-Cano S, Galata G, Schulte KM, et al. A 13-steroid serum panel based on LC-MS/MS: use in detection of adrenocortical carcinoma. Clin Chem 2017; 63: 1836–1846.
The authors
Victoria Treasure* MSc and Dr David Taylor PhD
Department of Clinical Biochemistry
(Viapath), King’s College Hospital NHS Foundation Trust, London, UK
*Corresponding author
E-mail: Victoria.treasure@nhs.net
The accuracy of LC-MS/MS technology with the convenience of automation
, /in Featured Articles /by 3wmediaThe new nSMOL Antibody Bioanalysis Kit
, /in Featured Articles /by 3wmediarx series – Excellence In Clinical Chemistry Testing
, /in Featured Articles /by 3wmediaMedica, Düsseldorf, Nov 12-15, 2018
, /in Featured Articles /by 3wmediaGonotec: a continued success story since 1979
, /in Featured Articles /by 3wmediaDxH 900 hematology analyzer – the right results, the first time
, /in Featured Articles /by 3wmediaMonoclonal Antibodies
, /in Featured Articles /by 3wmediaLyme disease diagnosis: waiting for the next gold standard
, /in Featured Articles /by 3wmediaLyme disease is caused by Borrelia spirochaetes: predominantly Borrelia burgdorferi in North America (but also present in Europe), and predominantly B. afzelii and B. garinii in Europe and Asia and is spread to people via infected deer ticks. Infection occurs after only a minority of tick bites, but is typified by three stages. Stage 1, early localized lyme disease is characterized by the bull’s eye rash (erythema migrans (EM)). Stage 2, early disseminated infection occurs within days to weeks after the local infection as the bacteria begin to spread through the bloodstream. Stage 3, late disseminated infection, where the infection has spread throughout the body, can occur several months later in untreated or inadequately treated patients involving chronic symptoms that can be severe and disabling. Treatment by antibiotics is effective in the early localized stage of the disease but this is often hampered by late diagnosis. Diagnosis can be delayed for a number of reasons: there is a lack of awareness in the general public (as well as GPs outside of what are thought to be the high-risk areas); approximately 25% of people do not get the typical bull’s eye rash; and symptoms can be so varied and vague that, when occurring weeks or months later, are difficult to relate back to the time of the tick bite. Knowledge of a tick bite and an associated EM rash is sufficient for diagnosis. However, in cases where there is a clinical suspicion of Lyme disease but no EM rash, laboratory testing is advised. Testing for antibodies is done via a two-tiered approach, starting with a sensitive ELISA, which, if positive or equivocal, is followed by a more specific immunoblot. However, the overall sensitivity of the two-tiered tests is only 64% when done in the early stages of infection, which is when accurate diagnosis is most needed. Because of these diagnostic limitations, the prevalence of Lyme disease is likely to be far higher than is currently thought. With increasing incidence and geographic spread of the disease, better testing for diagnosis, particularly in the early stages of infection, is perhaps required. Research is ongoing into PCR methods as well as and for the detection of OspA antigens that are shed into urine. An LLT-MELISA (lymphocyte transformation test-memory lymphocyte immunostimulation assay) has been developed and is suggested to be a useful supportive diagnostic tool, particularly in infections acquired in Europe. In the USA, next-generation sequencing (NGS) has been used for specific pathogen identification and to guide treatment decisions. With technological advances making NGS quicker and cheaper, could this eventually become the next gold standard test for Lyme disease?
Urine ethyl glucuronide and ethyl sulphate measurement using liquid chromatography-tandem mass spectrometry
, /in Featured Articles /by 3wmediaBackground
Ethyl glucuronide (EtG) and ethyl sulphate (EtS) are minor ethanol metabolites that can be used to detect recent alcohol consumption [1, 2]. Following the ingestion of alcohol, over 95% is metabolized by alcohol dehydrogenase to acetaldehyde. Up to 5% of ethanol is excreted unchanged in breath, sweat and urine. A small amount of ethanol (<0.1%) is metabolized in the liver by conjugation of glucuronic acid or sulphate to form EtG and EtS (Fig. 1). Following alcohol consumption, ethanol itself can only be detected in breath or urine for up to 6 or 12 hours, respectively (depending on the amount of alcohol consumed) [3]. In comparison, it has been demonstrated that EtG and EtS can be detected in urine for at least 24 hours and over 48 hours with heavy alcohol consumption [4].
The ability of these markers to detect alcohol intake over a longer time period means that they can be useful to identify alcohol relapses in alcohol-dependent individuals in treatment programmes [5]. In the UK, alcohol treatment programmes rely on breath ethanol and self-reporting to detect recent alcohol intake. However, this will only detect a proportion of individuals who are continuing to drink alcohol; this has been a low as 7% in one study comparing breathalyser/self-reported alcohol intake to urine EtG measurement [6]. Therefore, EtG and EtS can be helpful to detect those in alcohol treatment who are continuing to drink alcohol but deny it and have a negative breath ethanol test [7]. This allows additional interventions in individuals who are continuing to drink, which may ultimately improve outcomes. During 2016–17, 80 454 individuals entered alcohol treatment in England; of those 61% were free of alcohol dependence following the standard 12-week programme [8]. Therefore, improved detection of continuing alcohol consumption could lead to initiation of earlier intervention and altered strategies to increase the numbers successfully completing treatment.
Measurement of ethyl glucuronide and ethyl sulphate
Liquid chromatography (LC) to separate analytes with detection using mass spectrometry (MS) is now routinely used in clinical laboratories for an increasing number of tests. It is routine practice in urine toxicology testing for results to be confirmed by either LC or gas chromatography with detection using MS and it has been recommended by the United States Substance Abuse and Mental Health Services Administration (SAMHSA) that MS confirmation should be used for the measurement of EtG and EtS [9].
In tandem MS, two mass spectrometers are arranged sequentially with a ‘collision cell’ placed between the two instruments (Fig. 2). Using selective reaction monitoring, the first mass spectrometer (MS1) selects the ion with the mass/charge (m/z) ratio of interest. The selected ion (parent ion) is fragmented into small ions that enter the second mass spectrometer (MS2) where an ion with a specific m/z ratio is selected (daughter). Detection of analytes using an m/z ratio is very specific and sensitive allowing detection of very small amounts of EtG and EtS.
A number of liquid chromatography-tandem mass spectrometry (LC-MS/MS) methods for EtG and EtS have been published and a reference method has been proposed for EtG using solid phase extraction followed by LC-MS/MS [10]. Deuterated standards (EtG-d5 and EtS-d5) are readily available to purchase for use as internal standards ensuring reproducibility and reliability; an internal standard must mimic the analyte of interest but have a different mass to allow the MS detector to differentiate between the analyte of interest and the internal standard.
Sample preparation in published methods ranges from solid phase extraction to protein precipitation to dilution of urine in mobile phase. Solid phase extraction or protein precipitation of urine samples prior to LC-MS/MS can reduce the presence of potentially interfering substances which may cause ion suppression. It may also help to increase the lifespan of the column. For chromatographic separation of EtG and EtS, the mobile phases are usually formic acid in HPLC grade water and acetonitrile. Published methods have used both isocratic and gradients of mobile phase A and B to achieve separation of EtG and EtS; this is dependent on the sample preparation, the exact composition of the mobile phases and the column chosen. A rapid sample preparation of diluting urine samples in mobile phase A and then adding internal standard has been shown to be effective with no ion suppression or enhancement at or near the retention times for EtG and EtS [11]. Our experience has been to use an increasing gradient of mobile phase B (acetonitrile) from 1% to 10% over the first 2 minutes and then 10% to 100% from 2 minutes to 2.5 minutes. The increase from 1% to 10% acetonitrile elutes EtS/EtS-d5 at 1.27 minutes and the increase from 10% to 100% elutes EtG/EtG-d5 at 2.03 minutes. Figure 3 shows an example chromatogram for a urine sample collected from an individual attending the community based alcohol treatment programme; the high EtG and EtS results demonstrate that this person was continuing to drink alcohol.
Using MS to measure EtG and EtS requires the availability of LC-MS/MS equipment within the laboratory, the technical expertise required to set up an LC-MS/MS method and a dedicated member of staff to perform the analysis. In laboratories already using LC-MS/MS for other assays, there should be no difficulty in setting up a method to measure urine EtG and EtS.
An enzyme immunoassay method is also available to measure EtG and may be adapted for use on many automated laboratory analysers. This method has been shown to compare well to an LC-MS method [12]. For routine use, an immunoassay for EtG on an automated analyser has a number of advantages including rapid turnaround times, availability of EtG analysis out of routine working hours and the same staff members performing the analyses of multiple tests at the same time. However, there is no requirement for urine EtG and EtS analysis to be performed 24/7 as they would not be required in an acute setting. Generally, clients in a community treatment programme attend weekly, so once or twice weekly analysis using LC-MS/MS should be adequate for feedback of results to clients at their next visit. Not requiring a dedicated member of staff (as would be required for LC-MS/MS) is advantageous but according to SAMHSA guidelines, immunoassay results will require confirmation using a MS method. In addition, there is currently no immunoassay method available to measure EtS. This is important as there are a number of scenarios that can cause a false positive EtG result with a negative EtS result. For example, ‘positive’ EtG results (but not EtS results) have been demonstrated after the consumption of non-alcoholic beers (alcohol content 0.5%) [13]. EtG could also be formed in subjects with glycosuria and E.coli infection. If ethanol was formed due to the fermentation of sugars in the urine, this could be converted to EtG by bacteria present in the urine [14]. EtS would not be produced so again EtS can verify whether the EtG result is a true positive. Both EtG and EtS have been detected in individuals who used ethanol-based mouthwash or hand gel; however, the mouthwash was gargled 4 times/day which is much higher than the recommended frequency of use [15]. Owing to these factors, it is advisable to measure both EtG and EtS, which is currently only possible if using LC-MS/MS.
Cut-off values for EtG and EtS
There has been a lot of debate in the literature about suitable cut-off values to use for EtG and EtS. Some authors have suggested using the lower limit of detection (LLOD) or lower limit of quantitation (LLOQ) for the method so that any detectable EtG and EtS is a ‘positive’ result. However, the LLOD and LLOQ in LC-MS/MS methods will be variable between laboratories depending on a number of factors including sample preparation, column choice, chromatography and the tandem MS optimization. For EtG and EtS, the published LLOQs range from 0.05–0.20 mg/L and 0.04–0.10 mg/L respectively. New Clinical & Laboratory Standards Institute (CLSI) guidelines were published in 2016 and these should help to improve standardization between LC-MS/MS methods [16]. Alternatively, cut-off values could be defined by measuring EtG and EtS in a non-drinking population and incorporating measurement uncertainty (0.26 mg/L and 0.22 mg/L for EtG and EtS respectively) [11]. For EtG, a cut-off of 0.50 mg/L has been proposed to reduce the risk of false positive results. The disadvantage of a higher EtG cut-off is a reduction in sensitivity. Jatlow et al. demonstrated that using a 0.50 mg/L cut-off would only detect the intake of a low dose of alcohol 12 hours earlier (estimated blood alcohol 20 mg/dL) in 50% of participants. However, all participants had results above 0.10 mg/L and 0.20 mg/L after the same low alcohol dose 12 hours earlier [4]. SAMHSA have suggested separating EtG results into ‘high’ positive (>1.00 mg/L), ‘low’ positive (0.50–1.00 mg/L) and ‘very low’ positive (0.10–0.50 mg/L). They suggest that a ‘very low’ positive result may indicate previous heavy drinking (1–3 days ago), previous light drinking (12–36 hours ago) or ‘extraneous’ exposure [9].
Another consideration for urine EtG and EtS analysis is the dilution of urine samples; in urine toxicology testing, it is standard practice to measure creatinine to check the validity of a urine sample. There is limited data on the utility of EtG and EtS creatinine ratios. However, it is good practice to measure creatinine and question the validity of the EtG and EtS results if the creatinine is ≤2.0 mmol/L [17].
Conclusion
Urine EtG and EtS are valuable additional tools to detect recent alcohol intake in individuals undergoing treatment for alcohol dependence to ensure continued abstinence. Owing to the risk of false positive EtG results from unintentional exposure (e.g. non-alcoholic beer, urine infection with glycosuria, ethanol-based hand gel/mouthwash), the measurement of EtS in addition to EtG is recommended. An immunoassay is available for EtG but only MS allows the detection of both EtG and EtS to confidently confirm recent alcohol intake. There are a number of published methods for LC-MS/MS for EtG and EtS which are applicable for routine use in a clinical laboratory.
References
1. Dahl H, Stephanson N, Beck O, Helander A. Comparison of urinary excretion characteristics of ethanol and ethyl glucuronide. J Anal Toxicol 2002; 26: 201–204.
2. Helander A, Beck O. Ethyl Sulphate – a metabolite of ethanol in humans and a potential biomarker of acute alcohol intake. J Anal Toxicol 2005; 29: 270–274.
3. Helander A, Beck O, Jones W. Laboratory testing for recent alcohol consumption: comparison of ethanol, methanol and 5-hydroxytryptophol. Clin Chem 1996; 42: 618–624.
4. Jatlow P, Agro A, Wu R, Nadim H, Toll BA, Ralevski E, Nogueira C, Shi J, Dziura JD, et al. Ethylglucuronide and ethyl sulfate assays in clinical trials, interpretation and limitations: results of a dose ranging alcohol challenge study and two clinical trials. Alcohol Clin Exp Res. 2014; 38: 2056–2065.
5. Dahl H, Voltaire Carlsson A, Hillgren K, Helander A. Urinary ethyl glucuronide and ethyl sulphate for detection of recent drinking in an outpatient treatment program for alcohol and drug dependence. Alcohol Alcohol 2011; 46: 278–282.
6. Wetterling T, Dibbelt L, Wetterling G, Göder R, Wurst F, Margraf M, Junghanns K. Ethyl glucuronide (EtG): better than breathalyser or self-reports to detect covert short-term relapses into drinking. Alcohol Alcohol 2014; 49: 51–54.
7. Armer J, Gunawardana L, Allcock R. The performance of alcohol markers including ethyl glucuronide and ethyl sulphate to detect alcohol use in clients in a community alcohol treatment programme. Alcohol Alcohol 2017; 52: 29–34.
8. Knight J, Brand P, Willey P, van der Merwe J. Adult substance misuse statistics from the National Drug Treatment Monitoring System (NDTMS): 01 April 2016 – 31 March 2017. Public Health England 2017
(https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/658056/Adult-statistics-from-the-national-drug-treatment-monitoring-system-2016-2017.pdf).
9. The role of biomarkers in the treatment of alcohol use disorders. Substance Abuse and Mental Health Services Administration (SAMHSA) Advisory 2012; 11(2) (https://store.samhsa.gov/shin/content/SMA12-4686/SMA12-4686.pdf).
10. Helander A, Kenan N, Beck O. Comparison of analytical approaches for liquid chromatography/mass spectrometric determination of the alcohol biomarker ethyl glucuronide in urine. Rapid Commun Mass Spectrom 2010: 24: 1737–1743.
11. Armer J, Allcock R. Urine ethyl glucuronide and ethyl sulphate using liquid chromatography-tandem mass spectrometry in a routine clinical laboratory. Ann Clin Biochem 2017; 54: 60–68.
12. Bottcher M, Beck O, Helander A. Evaluation of a new immunoassay for urine ethyl glucuronide testing. Alcohol Alcohol 2008; 43: 46–48.
13. Thierauf A, Gnann H, Wohlfarth A, Auwärter V, Perdekamp MG, Buttler KJ, Wurst FM, Weinmann W. Urine tested positive for ethyl glucuronide and ethyl sulphate after the consumption of “non-alcoholic” beer. Forensic Sci Int 2010; 202: 82–85.
14. Helander A, Ollson I, Dahl H. Postcollection synthesis of ethyl glucuronide by bacteria in urine may cause false identification of alcohol consumption. Clin Chem 2007; 53: 1855–1857.
15. Reisfield G, Goldberger B, Pesce A, Crews BO, Wilson GR, Teitelbaum SA, Bertholf RL. Ethyl glucuronide, ethyl sulfate, and ethanol in urine after intensive exposure to high ethanol content mouthwash. J Anal Toxicol 2011; 35: 264–268.
16. Lynch K. CLSI C62-A: a new standard for clinical mass spectrometry. Clin Chem 2016; 62(1): 24–29.
17. European guidelines for workplace drug testing in urine. European Workplace Drug Testing Society 2015 (http://www.ewdts.org/data/uploads/documents/ewdts-urine-guideline-2015-11-01-v2.0.pdf).
The authors
Jane Armer*1 BA MSc FRCPath and Rebecca Allcock2 BSc MSc FRCPath
1Department of Blood Sciences, East Lancashire Hospitals NHS Trust, Blackburn, UK
2Department of Clinical Biochemistry, Lancashire Teaching Hospitals NHS Foundation Trust, Preston, UK
*Corresponding author
E-mail: jane.oakey@elht.nhs.uk
Use of an LC-MS/MS 13-steroid serum panel in the diagnosis of adrenocortical carcinoma
, /in Featured Articles /by 3wmediaLiquid chromatography-tandem mass spectrometry (LC-MS/MS) is increasingly becoming the method of choice in the clinical laboratory for the measurement of low molecular weight analytes. The major advantage that LC-MS/MS possesses relative to conventional laboratory techniques such as immunoassay is its higher specificity (and often sensitivity, although this is compound specific) and its ability to measure multiple compounds in a single run (multiplexing). LC-MS/MS thus provides the opportunity for more accurate and precise biochemical diagnosis and monitoring of human disease. One example of the increasing adoption of LC-MS/MS by clinical laboratories is the measurement of steroid hormones in various matrices (serum, saliva, urine).
Steroid metabolism
All steroids share a cyclopentanoperhydrophenanthrene nucleus, with individual species varying according to the presence of different functional groups attached to this four-ring structure, as well as by the oxidation state of the rings. Cortisol structure is given as an example in Figure 1. In humans, the major sites of steroid hormone production are the adrenal gland and the gonads. Steroids are synthesized from cholesterol via a series of enzyme-catalysed steps (Fig. 2), which are under tight regulation in healthy individuals by feedback mechanisms involving the hypothalamus and anterior pituitary. Steroids have a wide range of physiological functions which are summarized in Table 1.
Adrenocortical carcinoma – a diagnostic challenge
There are many endocrine disorders that result in the improper synthesis of steroids, and one of the rarest and most severe is adrenocortical carcinoma (ACC). ACC is a malignancy of the adrenal cortex with an annual incidence of 1 or 2 cases per million [1]. The majority of ACC cases are sporadic and occur in the fifth or sixth decade of life and more commonly in women; although ACC can be associated with several familial syndromes including Li-Fraumeni, Beckwith-Wiedemann, Lynch syndrome and multiple endocrine neoplasia type 1 [2]. Functional steroid hormone-producing tumours occur in around two-thirds of cases [3], presenting with varied signs and symptoms of steroid overproduction, most commonly Cushing’s syndrome (cortisol excess) and hyperandrogenism. ACC can progress rapidly in some patients, therefore it is vital that it is distinguished from benign adrenal adenomas, as ACC has a 5-year survival rate of <50% [2]. A surgical cure is only possible if the carcinoma is detected in its localized stage, otherwise the median survival period is <15 months [4].
The diagnosis of ACC is challenging as there is no single diagnostic tool that is able to distinguish ACC from other adrenal masses, including benign adenomas with glucocorticoid or mineralocorticoid excess, phaeochromocytoma and non-functioning adenomas. Imaging alone is insufficient for diagnosis, as although patients with ACC almost always present with tumours ≥4 cm, the presence of a large mass only has a clinical specificity of 61% [5]. Additionally, whereas up to two-thirds of tumours are functional, less than half of ACC cases present with clinical signs of steroid overproduction [3], with a further proportion presenting with other symptoms including abdominal pain. However, a significant proportion are discovered incidentally [2].
The European Network for the Study of Adrenal Tumours (ENSAT) currently recommends that the initial biochemical work-up for suspected ACC includes measurement of serum cortisol (both basal and assessment of suppression after dexamethasone), dehydroepiandrostenedione sulphate (DHEAS), androstenedione, testosterone, 17-hydroxyprogesterone, estradiol and aldosterone (if the patient is hypokalemic or hypertensive). An alternative approach is to measure steroid metabolites in urine using gas chromatography-mass spectrometry (GC-MS); increases in the excretion of metabolites of the steroid precursors 11-deoxycortisol, 17-hydroxypregnenolone and pregnenolone have been shown to provide particularly high diagnostic utility in ACC. Unfortunately, urine steroid profiling is not commonly available in clinical laboratories owing to lengthy sample preparation and complex result interpretation. Further, serum 11-deoxycortisol, 17-hydroxypregnenolone or pregnenolone measurements are rarely performed either because of lack of demand, or specificity of the available immunoassays which may be subject to significant levels of cross-reactivity.
As a result of these limitations, the use of LC-MS/MS is increasingly being adopted to provide more specific steroid hormone measurements. An approach we have taken in our laboratory is to develop and fully evaluate a multiplexed LC-MS/MS method panelling 13 steroids in serum [6] to include many of the steroid synthetic pathway intermediates currently not available for ACC work-up.
Use of a serum steroid panel
The steroids included in our serum panel are highlighted in Figure 2 and are as follows:
Samples are prepared for analysis by an initial protein precipitation step to remove steroids from their binding proteins, followed by liquid-liquid extraction in order to cleanly extract the steroids from remaining matrix components. Prepared extracts are then analysed by LC-MS/MS in which steroids are first resolved on a reverse phase C18 column by gradient elution followed by MS/MS detection using positive atmospheric pressure chemical ionization (APCI) operated in multiple reaction monitoring mode. Chromatographic separation of several isobaric (same mass to charge ratio) steroids is essential, as is the use of deuterated internal standards for all steroids in the method.
When we applied our method to adrenal tumour samples [6], we were able to show that between 4 and 7 steroids were elevated in all ACC cases in comparison to non-ACC adrenal tumours where a maximum of 1–2 steroids were abnormal. The cortisol precursor 11-deoxycortisol was most useful in the discrimination between ACC and non-ACC adrenal lesions, whereas other steroids markedly elevated in ACC included 17-hydroxypregnenolone and pregnenolone. Indeed, all steroids except testosterone in males and corticosterone and cortisone in both sexes were of use in discriminating ACC. This validates the use of a panelling approach when investigating adrenal masses.
Our findings compare well with urine steroid profiling studies. Although urine steroid profiling using 24-hour collections may offer greater clinical sensitivity compared to a single blood measurement owing to diurnal rhythms of steroid production, urine measurements rely on accurately timed collections that are often performed incorrectly and are inconvenient to the patient. Advantages of our LC-MS/MS serum panel compared to urine steroid profiling by GC-MS include a less labour intensive sample preparation, as well as less expertise required for the interpretation of complex profiles, as the serum method only targets selected steroids rather than the large number of their metabolites in urine.
Use of our LC-MS/MS serum steroid panel in ACC patients has further demonstrated the limitations of assessing serum steroids by immunoassay. We observed evidence of notable interference in ACC patients in the cortisol, progesterone, 17-hydroxyprogesterone and androstenedione immunoassays, inferred to be due to elevated concentrations of structurally related steroid precursors.
Future work
Currently, our 13-steroid serum panel has been used to study a relatively small number of ACC patients (because of the rarity of the disease), and clearly larger prospective studies are required to more fully determine the diagnostic utility of our panel in ACC. Further work is also required to clarify the effects of age, sex and diurnal variation on serum steroid panelling; nonetheless the most useful markers of ACC are markedly elevated above variation attributable to these biological factors. In addition to the complexity of interpreting biomarker panels, it is not only important to consider specific reference ranges, but to also consider the patterns in results which require an omics-based analysis approach to interpretation. The challenge surrounding this, as well as the requirement for clear presentation and reporting of results to clinicians requires close involvement of clinical colleagues for the development and introduction of such testing strategies.
The analysis of steroid panels by LC-MS/MS can also undoubtedly be used in other conditions including inborn errors of steroid metabolism such as congenital adrenal hyperplasia (CAH) and polycystic ovarian syndrome (PCOS).
Although we have demonstrated the advantages of our LC-MS/MS steroid panel compared to routine immunoassays, there are undoubtedly disadvantages of using LC-MS/MS. These include the initial cost of instrument purchase, the increased expertise required and often a more laborious sample preparation. Additionally, the specificity of mass spectrometry should not be readily assumed; careful selection of multiple reaction monitoring (MRM) transitions and chromatography conditions are essential to separate isobaric steroids and other interfering compounds. However, in the context of improving the biochemical tools available to us to aid the diagnosis of ACC, the advantages of LC-MS/MS far outweigh these limitations.
Summary
In summary, LC-MS/MS serum steroid panelling offers an additional tool for the challenge that is the diagnosis of ACC. Our method combines measurement of both common and rarely measured steroids in a single sample, which we have shown provides useful data to aid the discrimination of ACC from benign adrenal tumours. Use of LC-MS/MS gives several advantages over the immunoassay and GC-MS-based methods currently used to assess steroid overproduction, but further work is required to demonstrate the full potential of its use in the diagnosis of ACC.
References
1. Fassnacht M, Kroiss M, Allolio B. Update in adrenocortical carcinoma. J Clin Endocrinol Metab 2013; 98: 4551–4564.
2. Else T, Kim AC, Sabolch A, Ramond VM, Kandathil A, Caoili EM, Jolly S, Miller BS, Giordano TJ, Hammer GD. Adrenocortical carcinoma. Endocr Rev 2014; 35: 282–326.
3. Arlt W, Biehl M, Taylor AE, Hahner S, Libé R, Hughes BA, Schneider P, Smith DJ, Stiekema H, et al. Urine steroid metabolomics as a biomarker tool for detecting malignancy in adrenal tumours. J Clin Endocrinol Metab 2011; 96: 3775–3784.
4. Fassnacht M, Terzolo M, Allolio B, Baudin E, Haak H, Berruti A, Welin S, Schade-Brittinger C, Lacroix A, et al. Combination chemotherapy in advanced adrenocortical carcinoma. N Engl J Med 2012;366:2189–2197.
5. Hamrahian AH, Ioachimescu AG, Remer EM, Motta-Ramirez G, Bogabathina H, Levin HS, Reddy S, Gill IS, Siperstein A, Bravo EL. Clinical utility of noncontrast computed tomography attenuation value (Hounsfield units) to differentiate adrenal adenomas/hyperplasias from nonadenomas: Cleveland Clinical experience. J Clin Endocrinol Metab 2005; 90: 871–877.
6. Taylor DR, Ghataore L, Couchman L, Vincent RP, Whitelaw B, Lewis D, Diaz-Cano S, Galata G, Schulte KM, et al. A 13-steroid serum panel based on LC-MS/MS: use in detection of adrenocortical carcinoma. Clin Chem 2017; 63: 1836–1846.
The authors
Victoria Treasure* MSc and Dr David Taylor PhD
Department of Clinical Biochemistry
(Viapath), King’s College Hospital NHS Foundation Trust, London, UK
*Corresponding author
E-mail: Victoria.treasure@nhs.net