Development of a highly sensitive method for the quantification of estrone and estradiol in serum by liquid chromatography tandem mass spectrometry without derivatization

https://doi.org/10.1016/j.jchromb.2012.02.034Get rights and content

Abstract

Measurement of estrone (E1) and estradiol (E2) values <1 pg/mL (3.7 pmol/L) is necessary for postmenopausal, pediatric and male serum samples. Until now this was rarely reached and only through derivatization which can present problems for estradiol. A very sensitive LC–MS/MS method was developed avoiding derivatization, convenient for large-scale studies. The desired sensitivity and specificity were achieved using ESI negative mode, LLE and a 2D chromatography consisting of a trapping column and a second dimension reverse-phase C8 analytical column. A mixture of an aqueous solution of ammonium fluoride at 0.2 mM and methanol was used on the analytical column to further increase the sensitivity. Serum LOQ was <0.5 pg/mL (1.9 pmol/L) for E2 and E1 and recoveries ranged from 95 to 105%. No carry-over was detectable. Inter assay CV's were 4.0% at 21 pg/mL (77 pmol/L) for E2, 7.6% at 25 pg/mL (93 pmol/L) for E1. Comparison with commercial direct estrogen assays (Roche Diagnostics E170 for E2, Bioline RIA for E1) exposed analytical unsuitability (due to a combined lack of sensitivity and specificity) for the assay of male, postmenopausal or pediatric samples.

Highlights

► Analysis of estradiol and estrone by LC–MS/MS without derisvatization. ► LOQ below 0.5 pg/mL for estradiol and estrone. ► Suitable for analysis in children and post-menopausal women.

Introduction

Both for clinical research and potential clinical applications, there is considerable interest for reliable and practical measurement of low estrogen serum concentrations of estrone (E1) and estradiol (E2). Particularly in postmenopausal women (<30 pg/mL or 110 pmol/L), in amenorrheic premenopausal women (e.g. in anorexia nervosa and other forms of hypothalamic amenorrhea), in prepubertal children or in patients treated with inhibitors of aromatase [1], [2], [3], [4]. Postmenopausal E2 serum levels have been associated with breast cancer, osteoporosis and fracture risk [1], [5], [6]. Estrogens also play an important physiological role in men, who present at adult age with moderately low serum E2 levels (mean levels around 18 pg/mL or 67 pmol/L) [7]. Male serum E2 levels have been associated with parameters of skeletal health such as rate of bone loss and fracture risk in the elderly [8], [9].

Unfortunately the most widely used techniques for E2 measurement, direct serum immunoassays, do not perform well in the lower range such as seen in postmenopausal women due to lack of sensitivity and specificity. For concentrations below 20 pg/mL (74 pmol/L) correlation with concentrations measured with a reference GC–MS/MS method is very poor [6], [10], [11]. Although indirect immunoassays with extraction and chromatographic separation steps (e.g. on LH20 gel column) perform better, they have largely been abandoned because they require the use of radioactive isotopes and large amounts of serum (up to 2 mL).

Highly sensitive bioassays for E2 have been proposed [1], [2], [6], but their use is limited to research settings in a limited number of specialized labs. Presently, mass spectroscopy-based methods (GC–MS/MS; LC–MS/MS) are the methods of choice for steroid hormone assays where GC based methods suffer from requiring large sample volumes and long run times limiting practical usage [12].

To increase sensitivity for estrogen measurement on LC–MS/MS, reported methods often resorted to derivatization [13], [14], [15], [16], [17]. However derivatization based methods are less preferable for estradiol measurement because PH and temperature changes can potentially influence hydrolysis of the conjugated estrogens resulting in falsely high measurements [18], [19]. In addition specificity can be compromised [24], [25] and the more lengthy and delicate sample preparation is less suitable for large-scale studies. Some attempts have been made to design methods without derivatization [19], [20], [21], [22] but these were until now not sensitive enough for the analysis of the very low serum concentrations typically seen in children or postmenopausal women [23]. We describe here a very sensitive assay with LOQ well below 0.5 pg/mL (1.9 pmol/L) for E2 and E1 with serum extraction followed by direct measurement on 2D-LC–MS/MS without need for derivatization.

Section snippets

Materials and methods

17beta estradiol (E2) and estrone (E1) were obtained from Sigma–Aldrich, 17beta estradiol-d4 (d4-E2) and estrone-d4 (d4-E1) were purchased from CDN Isotopes, Inc. All standards and internal standards were dissolved in methanol. Methanol, water and acetonitrile were LC–MS grade from BioSolve BV (Varkenswaard, The Netherlands).

As for comparison with routine assays, 17β-estradiol was measured by electrochemiluminescence immunoassay ‘ECLIA’ on a Modular E170 immunoassay analyzer (Roche Diagnostics,

Chromatography

One of the challenges in measuring most of the steroids by LC–MS is their inherent poor sensitivity accountable to a low polarity of these molecules. In this frame, estradiol has demonstrated to provide better sensitivity when measured in negative ion mode with the ESI source, and APCI ionization has shown to give a lower yield. For willing to pursue the approach in skipping any derivatization, all the actions must be taken in order both to improve the achievable sensitivity and to minimize any

Conclusions

The primary goal of this endeavour was to devise a robust, accurate and ultrasensitive determination of E2 and E1, convenient for large-scale studies and avoiding unwanted derivatization steps as current commercial direct estrogen assays are unusable for these samples. The finalized setup, although employing sophisticated plumbing, has proven that measuring serum values well below 0.5 pg/mL (1.9 pmol/L) with excellent CV's, needed for pediatric, pre-pubertal and postmenopausal studies can be very

References (25)

  • R. Santen et al.

    Steroids

    (2008)
  • K. Yamashita et al.

    Steroids

    (2007)
  • W.C. Yang et al.

    J. Chromatogr. B: Anal. Technol. Biomed. Life Sci.

    (2008)
  • T. Guo et al.

    Clin. Biochem.

    (2008)
  • U. Ceglarek et al.

    Clin. Chim. Acta

    (2009)
  • D. Caruso et al.

    Neurochem. Int.

    (2008)
  • D.T. Harwood et al.

    Clin. Chim. Acta.

    (2009)
  • I.A. Blair

    Steroids

    (2010)
  • L. Xu et al.

    Anal. Biochem.

    (2008)
  • F. Paris et al.

    J. Clin. Endocrinol. Metab.

    (2002)
  • M. Misra et al.

    J. Clin. Endocrinol. Metab.

    (2008)
  • S.A. Burnett-Bowie et al.

    J. Clin. Endocrinol. Metab.

    (2009)
  • Cited by (116)

    View all citing articles on Scopus
    View full text