A sensitive, simple and robust LC–MS/MS method for the simultaneous quantification of seven androgen- and estrogen-related steroids in postmenopausal serum

https://doi.org/10.1016/j.jsbmb.2014.08.015Get rights and content

Highlights

  • DHEA, 5-diol, 4-dione, testo, DHT, E1 and E2 are measured simultaneously.

  • The LLOQs were 500, 100, 100, 50, 10, 4 and 1 pg/mL, respectively.

  • A simple and single extraction was used for the seven steroids.

  • Importantly, the LC–MS/MS technology was validated according to FDA guidelines.

  • Specificity, precision and reliability are assured by proper validation.

Abstract

Steroids were first analyzed by immunoassay-based methods followed by gas chromatography mass spectrometry (GC–MS or GC–MS/MS) with derivatization techniques since steroids are neutral and do not ionize at a high level using the electrospray ionization technique. We now report a validated liquid chromatography tandem mass spectrometry (LC–MS/MS) method for the simultaneous quantification of seven steroidal compounds, i.e., estradiol (E2), estrone (E1), testosterone (Testo), dihydrotestosterone (DHT), androst-5-ene-3β, 17β-diol (5-diol), dehydroepiandrosterone (DHEA) and androstenedione (4-dione). The system used is a UPLC–MS/MS (Qtrap 6500) system. With this method, the sample preparation is the combination of liquid–liquid extraction and a simple selective derivatization for only E1 and E2. This assay method is simple and practically eliminates potential contamination. Low quantification limits of 1 pg/mL, 4 pg/mL, 50 pg/mL, 10 pg/mL, 100 pg/mL, 500 pg/mL and 100 pg/mL have been found, respectively for the steroids mentioned above. Without derivatization, DHT sensitivity can be as low as 4 pg/mL with S/N  5. A full validation has been performed for the seven compounds in compliance with GLP and FDA guidelines for bioanalytical method development and validation. Recovery of all seven compounds in unstripped serum is similar to that in stripped serum: 72.1–84.7% for E2, 83.6–94.5% for E1, 88.2–90.3% for Testo, 82.0–90.6% for DHT, 84.9–92.0% for 5-diol, 88.1–93.8% for DHEA and 86.2–90.3% for 4-dione, respectively. A good linearity is obtained with R > 0.99 for each compound within its calibration range. Accuracies of all levels of QC are within the range of 15% for all seven compounds. The between day variation coefficients are 6.1–8.9% for the low limits of quantification of all seven compounds with 0.7–6.1% for higher levels of QCs for all seven compounds. All results of other test parameters similarly meet the acceptance criteria of EndoCeutics SOPs and FDA guidelines. By comparison of GC–MS/MS and LC–MS/MS data for six derivatized and nonderivatized free steroids, the present data show the crucial importance to use validated assays according to the FDA guidelines to increase specificity, precision and reliability of the absolute values associated with MS/MS-based assays. This method has already been applied to series of samples from clinical trials and is ready for wide clinical use.

Introduction

Sex steroids play an important role in human health and diseases and their assay at low concentrations in the circulation after menopause is particularly challenging. However, knowledge about their precise concentration is important for the choice of therapy for the main symptoms and signs of menopause, namely hot flushes, vaginal atrophy (pain at sexual activity, vaginal dryness, irritation/itching), osteoporosis/fractures, muscle loss, loss of memory, cognition loss and type 2 diabetes [1]. The conventional technique in the clinical laboratory rests upon immunoassay-based assays. These methods are of uncertain specificity while their sensitivity does not meet the requirement for some compounds [2], [3], [4], [5].

Combining the separation power of GC and the sensitive detection from MS for volatile compounds, GC–MS and GC–MS/MS have been successfully applied for steroid assays in many laboratories with one-step or two-step derivatization [5], [6]. Selection of the derivatization reagent is determined by the ionization efficiency of the resulting derivatized steroid [6], [7]. Derivatization for GC–MS/MS for the simultaneous assay of many analytes also improves the specificity by changing their chemical and physical properties such as chemical structure, polarity and molecular weight. In fact, the chemical structure change may provide a better separation of analytes by chromatography while the molecular weight change may provide additional multiple reaction monitoring (MRM) transitions and thus eliminate interferences. GC–MS/MS, however, has constantly encountered robustness issues since the extracted samples are not sufficiently clean and the sample size becomes too large to reach the required sensitivity. Derivatization for the GC–MS/MS method implies a sample preparation procedure which is time- consuming while the instrumentation needs frequent maintenance. The derivatization of neutral steroids usually requires water free environment and solvents. In addition, the relatively large amount of by- products from the derivatization reagent, usually the hydrolysis by-products of the derivatization reagent, can cause a significant matrix effect and thus be a challenge for the robustness of GC–MS/MS assays.

On the other hand, the challenge for the LC–MS/MS quantification of free steroids is two-fold: sensitivity and specificity. Many steroids have low ionization efficiency due to their neutral nature, i.e., lack of polarity. Furthermore, the concentration of many endogenous steroids in human serum or plasma is in the low pg/mL range [5]. Consequently, a highly sensitive MS instrument is required. Since hormones are endogenous compounds, the most probable interferences are likely to come from their isomers and matrices. These two factors add to the challenge for a sensitive and accurate measurement of free steroids in human serum, plasma or blood samples with or without derivatization by the LC–MS/MS technology.

Since 2000 [8], [9], [10], [11], LC–MS/MS methods have been reported for the measurement of steroids with simple or without derivatization since it is usually more rugged for biosample assay while providing high specificity. Recently, with the availability of more sensitive mass spectrometers combined with ultra performance liquid chromatography (UPLC), LC–MS/MS has become an advantageous technique for the assay of steroids. The UPLC column with more separation power potentially improves the specificity of the LC–MS/MS method. For some compounds having high ionization efficiency due to their large proton affinity such as testo, progesterone and 4-dione, the low limit of quantification (LLOQ) can be achieved at the 5 pg/mL level for Testo and 40 pg/mL for 4-dione without derivatization using LC–MS/MS [12]. For compounds such as estrone (E1) and estradiol (E2) having low ionization efficiency, the pg/mL level of the LLOQ can be best achieved with derivatization [13]. The derivatization reagent is usually categorized as a keto derivatization reagent, for instance, hydroxylamine while hydroxyl derivatization can be achieved with a dansyl chloride and picolinic acid [13], [14], [15], [16], [17]. With the derivatization using dansyl chloride or picolinic acid, E1 and E2 can be quantified at a lower level (≤1 pg/mL) in serum or blood. Without derivatization, the LLOQ of ≤1 pg/mL level for E1 and E2 in negative mode was reported by Fiers and Harwood [18], [19]. In these methods, almost all the extracted sample from 500 μL of serum was injected into the system while the mobile phase contains ammonium fluoride. Apparently, this method is pretty challenging to be validated as a GLP method for steroid hormone measurement since the sample volume could be insufficient for the reinjection needed in case of instrumentation or operation errors. Potentially, more extraction volume could bring more matrix effects.

More recently, a few research groups [20], [21], [22], [23] have reported simultaneous measurement of steroids. However, there is no method reported for the simultaneous sensitive measurement of estrogen- and androgen-related steroids with high specificity. A sensitive method reported by Keski-Rahkonen [22] only covered androgenic and progestagenic steroids while a method reported by Kulle [21] did not cover estrogens and has shown a LLOQ of 29 pg/mL for dihydrotestosterone (DHT). Sixteen steroids could be measured simultaneously by a standardized LC–MS/MS, but with a LLOQ of 30 pg/mL for E1 and 20 pg/mL for E2 [23].

In the present report, we describe a robust and sensitive LC–MS/MS method for the simultaneous quantification of seven free steroidal compounds, i.e., E2, E1, testosterone (Testo), DHT, androst-5-ene-3β, 17β-diol (5-diol), dehydroepiandrosterone (DHEA) and androstenedione (4-dione), using a relatively simple sample preparation. With this technique, E2 and DHT are the most challenging ones in terms of sensitivity, since the sensitivity for E2 should reach ≤1 pg/mL while the sensitivity requirement for DHT is ≤10 pg/mL in postmenopausal serum. The present method has been validated in full compliance with the FDA guidelines and EndoCeutics SOPs and has already been used for a large number of clinical samples.

Section snippets

Chemicals

Standards and internal standards (IS) are commercially available: E2 (95.8%), E1 (99.9%) and Testo (99.9%) from USP reference standard (Rockville, USA); 5-diol (91.6%), DHT (98.8%) and 4-dione (98.9%) from Steraloids Inc. (Newport, USA). DHEA has been obtained from Proquina (Ixtaczoquitlán, México). Internal standards of E2-d4, E1-d4, 4-dione-d7, Testo-d5, DHT-d6, 5-diol-d3 were purchased from CDN isotopes (Pointe-Claire, Canada) while DHEA-d5 was from Cerilliant (Round Rock, USA).

Analyte detection

The sample preparation combines the simple liquid–liquid extraction and derivatization of the phenol group of E1 and E2 under aqueous and basic conditions. Under the current conditions, i.e., in basic aqueous conditions at 60 °C, the 17-hydroxyl group derivatization product of E2 was not observed. Therefore, it is reasonable to conclude that the current derivatization conditions would not derivatize the hydroxyl groups of the other four compounds (DHT, DHEA, Testo and 5-diol) but only the

Summary

A simultaneously sensitive and accurate measurement of seven steroids in one assay is a challenging task for detailed and precise steroid profiling. In the present study, a selective derivatization of E1 and E2 using UPLC-Qtrap-6500 has proved to be a sensitive, highly specific and robust method for the accurate measurement of seven sex steroids, among which the LLOQs of E2 and DHT can be achieved at ≤1 pg/mL and ≤10 pg/mL, respectively. The sensitivity for DHT can be achieved as low as 4 pg/mL

Acknowledgements

The authors thank Mr. Réjean Gaudy, Ms. Geneviève Smith and Ms. Arianne Beaulieu for their skillful technical assistance during the method validation.

References (26)

  • W. Rosner et al.

    Challenges to the measurement of estradiol: an endocrine society position statement

    J. Clin. Endocrinol. Metab.

    (2013)
  • W. Rosner et al.

    Position statement: utility, limitations, and pitfalls in measuring testosterone: an Endocrine Society position statement

    J. Clin. Endocrinol. Metab.

    (2007)
  • C. Ohlsson et al.

    Comparisons of immunoassay and mass spectrometry measurements of serum estradiol levels and their influence on clinical association studies in men

    J. Clin. Endocrinol. Metab.

    (2013)
  • Cited by (0)

    View full text