* Probably the most challenging forms to quantify, due to their relatively low concentrations are the so called free forms. The free forms of all the hormones (shown in Figs. 1–3) are the hormones not bound to proteins, or conjugated to other molecules (see Fig. 2) and are biologically active
* In normal men and women, between 40 % and 65 % of circulating T and between 20 % and 40 % of circulating E2 are bound to SHBG. [17] Binding to SHBG prevents these hormones from diffusing out of the bloodstream, making it essential to assess the concentration of non-SHBG-bound forms, which are the bioactive forms.
* Overall, the RI of total T in serum/plasma between the ages of 20–40 years was established at 7.17–33.70 nmol/L in healthy males and at 0.33–2.05 nmol/L in healthy females.
* The significance of establishing RIs of estrogens across various life stages was emphasized by the Endocrine Society’s statement in 2013 that “The measurement of estradiol in biological fluids is important in human biology from cradle to grave.” [45]
* As shown in Table 4, the establishment of RIs of estrogens has not reached the level of success achieved with androgen (especially in T) assays [46]. Among all estrogens, only E2 was reported once in Mezzullo study [42] as a quantitative level at 80 pmol/L in males (20–70 years old), 395 pmol/L in females (21–51 years old). In other studies, E2 and other estrogen values were BQL or not included (X).
ABSTRACT
Accurate quantification of androgens and estrogens is critical for elucidating their roles in endocrine disorders nd advancing research on their functions in human biology and pathophysiology. This review highlights recent advances and ongoing challenges in liquid chromatography- mass spectrometry (LC- MS) methodology for quantifying androgens and estrogens in human serum and plasma. We summarized current approaches for analyzing the different forms of androgens and estrogens, along with their reported levels in publications from2010 to the present. These published levels pointed out the inconsistencies in reference intervals across studies.To address these issues, advances in derivatization methods and chromatographic separation techniques arere viewed. Future perspectives for improving the accuracy and consistency of hormone quantification in clinical and research settings were also proposed.
2. LC-MS approaches for analysis of different form of androgens and estrogens in human serum and/or plasma
Hormones that are circulating in blood and be found in many forms. Probably the most challenging forms to quantify, due to their relatively low concentrations are the so called free forms. The free forms of all the hormones (shown in Figs. 1–3) are the hormones not bound to proteins, or conjugated to other molecules (see Fig. 2) and are biologically active. On the other hand, protein-bound forms are those hormones bound to carrier proteins such as sex hormone-binding globulin (SHBG) or albumin. In normal men and women, between 40 % and 65 % of circulating T and between 20 % and 40 % of circulating E2 are bound to SHBG. [17] Binding to SHBG prevents these hormones from diffusing out of the bloodstream, making it essential to assess the concentration of non-SHBG-bound forms, which are the bioactive forms. Conjugated forms are the hormones that were conjugated to sulfate or glucuronide groups. These conjugates (sulfates and glucuronides) can serve as a reservoir for respective hormones because they can be taken up by cells and enzymatically hydrolyzed by steroid sulfatase and glucuronidase.[18] Therefore, quantifying the conjugates is as important as analyzing the free bioactive forms.
2.1. Analysis of androgens and estrogens in free and protein-bound forms
It is still challenging to use LC-MS assays to directly determine free forms of androgens and estrogens in human plasma or serum samples. We found that only a few papers have been published on direct measurement of free forms of T (FT) in human serum samples (Table 1), [19–22] and there appeared to be no published assays for other free androgens or estrogens. Equilibrium dialysis (ED) [20,22] and ultrafiltration [19,21] are two methods for separating FT from other forms in serum samples. A large sample volume, from 200 to 1000 μL, was necessary due to the relatively low concentration of FT. APCI source offered lower sensitivity without derivatization step versus electrosprayionization (ESI) source for detecting hydroxylamine-hydrochloride(HOA) derivatives. The lowest limit of quantification (LLOQ) of T in the publications could reach to 2.4 pmol/L [20].
Measuring bioactive forms of androgens and estrogens (non-SHBG bound forms) rather than SHBG-bound forms is recommended for evaluating hormone disorders in humans. In most applications, the concentrations of bioactive forms were calculated based on quantitative results for total levels, albumin concentration, and SHBG concentration using mathematical models [24,25]. Therefore, the accuracy of the results relies on the measurement of total levels and SHBG. Recent studies have shown that different mathematical models can produce significantly different final concentrations. [17,25] It is important to be cautious when selecting a calculation method, as the choice of model can significantly impact the results.
2.2. Analysis of androgens and estrogens in unconjugated and conjugated forms
2.2.1. Unconjugated androgens and estrogens
2.2.2. Conjugated and combined androgens and estrogens
3. Reference intervals (RIs) of androgens and estrogens determined by LC-MS/MS
Reference intervals (RIs), representing the range of values expected for a healthy population, are critical for assessing whether an individual’s androgen or estrogen levels are within normal limits or indicative of potential hormonal disorders or conditions. However,establishing accurate RIs involves studying a large, well-characterized, healthy population and considering many factors such as age, sex, and diurnal variations, etc. Among all the androgens and estrogens, only DHEA, DHEA-S, 4-AD, and T have been reported in more than three papers with over 300 participants. Table 5 shows the reference intervalranges of androgens and estrogens (if available) established by LC-MS/MS method from 2010 to the present, involving over 300 participants [39–43] (Supplementary Table 1). Table 5 lists the sample preparation method and LC-MS conditions for individual reference.
3.1. Androgen RIs in human serum and plasma
Androgen levels in healthy populations typically are at the concentration of μmol/L to nmol/L in human serum or plasma [39–43]. With sensitivity being a relatively minor concern, the high specificity and high accuracy of LC-MS/MS method are valued for improved diagnosticsin routine androgen analysis in clinical practice. To this end, the Endocrine Society proposed that “the best prospect for a gold stand (in testosterone testing) lies in extractions and chromatography followed by MS or MS/MS in which the chemical structure of the molecule measuredi s identified”. [44]
Table 5 shows that total T levels were measured in all studies, 4-AD levels were quantified in four studies, DHEA and DHEA-S were shown in three studies, and DHT levels were BQL or not available (NA) in most of the studies. The age ranges and sub-groups varied due to the limits of the studied population (Supplementary Table 1). Overall, the RI of total T in serum/plasma between the ages of 20–40 years was established at 7.17–33.70 nmol/L in healthy males and at 0.33–2.05 nmol/L in healthy females. Studies such as Damgaard-Olesen group [40] found no significant correlation between age and serum/plasma T concentrationsin men. Conversely, an age-related T decline has also been reported in some studies [41,43]. In most studies, the age-related decline in the concentrations of DHEA-S, DHEA, and 4-AD was observed [39–41,43] (Table 5 and Supplementary Table 1). Specifically, Kunz’s group [43] found a significant reduction in the upper reference limits across age groups. In males, there was a decrease from the youngest to the oldest age group by 73 % for DHEA, 44 % for DHEA-S, and 44 % for 4-AD, respectively. A similar result was obtained in females, post-menopause values showed decreases of 8 %, 12 %, and 46 % for DHEA, DHEA-S, and AD, respectively [43] (Supplementary Table 1)
3.2. Estrogen RIs in human serum and plasma
The significance of establishing RIs of estrogens across various life stages was emphasized by the Endocrine Society’s statement in 2013 that “The measurement of estradiol in biological fluids is important in human biology from cradle to grave.” [45] Today, laboratories areequi pped to measure E2 concentration at 184 pmol/L (50 pg/mL) in serum or plasma to support women undergoing fertility treatments. Moreover, the high sensitivity method becomes critical for monitoring postmenopausal women, with typical E2 levels near 55 pmol (15 pg/mL) [3,27]. For patients with breast cancer on aromatase inhibitors, an ultra-high sensitivity below 3.7 pmol/L (1 pg/mL) is required [45]. As shown in Table 4, the establishment of RIs of estrogens has not reached the level of success achieved with androgen (especially in T) assays [46]. Among all estrogens, only E2 was reported once in Mezzullo study [42] as a quantitative level at 80 pmol/L in males (20–70 years old), 395 pmol/L in females (21–51 years old). In other studies, E2 and other estrogen values were BQL or not included (X).
4. Advances in other analytical aspects of LC- MS
4.1. Derivatization
4.2. Chromatographic separation
5. Future perspective
It is essential to ensure consistency and accuracy in steroid measurements, particularly for establishing reliable RIs across different laboratories and platforms. The CDC’s Hormone Standardization Program (HoSt) and the National Institute of Standards and Technology (NIST) have aimed to unify analytical procedures. These efforts, along with Clinical and Laboratory Standards Institute (CLSI) guidelines, emphasize establishing new RIs, particularly with novel analytical methods. In addition, a standard reference material (SRM 971) was developed to address the need for improved accuracy of routine clinical assays [41].
A promising direction in analytical method development is enhancing the breadth and depth of steroid hormone profiling considering the diversity of steroid metabolism. Such advances are expected to generate complex data sets that are well-suit to metabolomics and steroidomics [61]. In hybrid instruments (e.g. quadrupole/time-of-flight (TOF) or quadrupole/Orbitrap), precursors were selected in quadrupole, and MS/MS scan of product ions can be performed in HR-MS,which could increases the specificity of the analysis. Normally the latest generation of triple-quadrupoles instrument would ensure the maximum sensitivity, but due to the unit resolution, it is possible that in some particular cases a PRM acquired on a HRMS actually offers more specificity, even at a minor loss of sensitivity. An approach has proposed the development of the high-resolution multiple reaction monitoring (MRMHR) method on a TripleTOF 5600 (AB Sciex) for quantification of circulating steroids in human and mice samples with confidence, precision, and accuracy [62]. Groessl’s group [63] reported using a Vanquish LC (Thermo) coupled with a QExactive Plus (Thermo) for the absolute quantification of 51 steroids for clinical analysis in human serum and peritoneal fluid for the first time. [63] Data acquisition on HRMS can be set under both parallel reaction monitoring (PRM) and fullscan (FS) mode, which allows simultaneously targeted and untargeted analysis. This approach opens novel possibilities for the post-analysis of clinical samples (ss shown by Snyder group in 2015 for the Girard derivatives of androgens) as the data can be examined for virtually any steroid even after data acquisition. This flexibility ensures that absolute quantification can be seamlessly implemented when reference standards are obtainable, positioning this technique as a valuable tool in clinical analytics.
Chromatographic development plays a crucial role in the analysis of steroids due to the complexity of steroid isomers, stero-isomers and isobaric compounds. Recent chromatographic advances have also seen the incorporation of multi-dimensional chromatography techniques. Multi-dimensional chromatography techniques, such as ion mobility spectrometry coupled with mass spectrometry (IM-MS) [64], have improved peak capacity and specificity in steroid analysis by providing additional separation. The recently introduced traveling wave ionmobility (TWIM) technique further enhances ion propulsion and separation. [65] For instance, ESI-TWIM-MS [66,67] has been successfully used to analyze steroid isomers and glucuronides of T and epi-T. Additionally, cryogenic IR spectroscopy offers structured IR fingerprints for clear molecular identification. A combination of HR-IMS with messenger-tagged IR spectrometry [68] has been applied to quickly identify steroid isobars and isomers, eliminating the need for recurrent analytical standards.
* In normal men and women, between 40 % and 65 % of circulating T and between 20 % and 40 % of circulating E2 are bound to SHBG. [17] Binding to SHBG prevents these hormones from diffusing out of the bloodstream, making it essential to assess the concentration of non-SHBG-bound forms, which are the bioactive forms.
* Overall, the RI of total T in serum/plasma between the ages of 20–40 years was established at 7.17–33.70 nmol/L in healthy males and at 0.33–2.05 nmol/L in healthy females.
* The significance of establishing RIs of estrogens across various life stages was emphasized by the Endocrine Society’s statement in 2013 that “The measurement of estradiol in biological fluids is important in human biology from cradle to grave.” [45]
* As shown in Table 4, the establishment of RIs of estrogens has not reached the level of success achieved with androgen (especially in T) assays [46]. Among all estrogens, only E2 was reported once in Mezzullo study [42] as a quantitative level at 80 pmol/L in males (20–70 years old), 395 pmol/L in females (21–51 years old). In other studies, E2 and other estrogen values were BQL or not included (X).
ABSTRACT
Accurate quantification of androgens and estrogens is critical for elucidating their roles in endocrine disorders nd advancing research on their functions in human biology and pathophysiology. This review highlights recent advances and ongoing challenges in liquid chromatography- mass spectrometry (LC- MS) methodology for quantifying androgens and estrogens in human serum and plasma. We summarized current approaches for analyzing the different forms of androgens and estrogens, along with their reported levels in publications from2010 to the present. These published levels pointed out the inconsistencies in reference intervals across studies.To address these issues, advances in derivatization methods and chromatographic separation techniques arere viewed. Future perspectives for improving the accuracy and consistency of hormone quantification in clinical and research settings were also proposed.
2. LC-MS approaches for analysis of different form of androgens and estrogens in human serum and/or plasma
Hormones that are circulating in blood and be found in many forms. Probably the most challenging forms to quantify, due to their relatively low concentrations are the so called free forms. The free forms of all the hormones (shown in Figs. 1–3) are the hormones not bound to proteins, or conjugated to other molecules (see Fig. 2) and are biologically active. On the other hand, protein-bound forms are those hormones bound to carrier proteins such as sex hormone-binding globulin (SHBG) or albumin. In normal men and women, between 40 % and 65 % of circulating T and between 20 % and 40 % of circulating E2 are bound to SHBG. [17] Binding to SHBG prevents these hormones from diffusing out of the bloodstream, making it essential to assess the concentration of non-SHBG-bound forms, which are the bioactive forms. Conjugated forms are the hormones that were conjugated to sulfate or glucuronide groups. These conjugates (sulfates and glucuronides) can serve as a reservoir for respective hormones because they can be taken up by cells and enzymatically hydrolyzed by steroid sulfatase and glucuronidase.[18] Therefore, quantifying the conjugates is as important as analyzing the free bioactive forms.
2.1. Analysis of androgens and estrogens in free and protein-bound forms
It is still challenging to use LC-MS assays to directly determine free forms of androgens and estrogens in human plasma or serum samples. We found that only a few papers have been published on direct measurement of free forms of T (FT) in human serum samples (Table 1), [19–22] and there appeared to be no published assays for other free androgens or estrogens. Equilibrium dialysis (ED) [20,22] and ultrafiltration [19,21] are two methods for separating FT from other forms in serum samples. A large sample volume, from 200 to 1000 μL, was necessary due to the relatively low concentration of FT. APCI source offered lower sensitivity without derivatization step versus electrosprayionization (ESI) source for detecting hydroxylamine-hydrochloride(HOA) derivatives. The lowest limit of quantification (LLOQ) of T in the publications could reach to 2.4 pmol/L [20].
Measuring bioactive forms of androgens and estrogens (non-SHBG bound forms) rather than SHBG-bound forms is recommended for evaluating hormone disorders in humans. In most applications, the concentrations of bioactive forms were calculated based on quantitative results for total levels, albumin concentration, and SHBG concentration using mathematical models [24,25]. Therefore, the accuracy of the results relies on the measurement of total levels and SHBG. Recent studies have shown that different mathematical models can produce significantly different final concentrations. [17,25] It is important to be cautious when selecting a calculation method, as the choice of model can significantly impact the results.
2.2. Analysis of androgens and estrogens in unconjugated and conjugated forms
2.2.1. Unconjugated androgens and estrogens
2.2.2. Conjugated and combined androgens and estrogens
3. Reference intervals (RIs) of androgens and estrogens determined by LC-MS/MS
Reference intervals (RIs), representing the range of values expected for a healthy population, are critical for assessing whether an individual’s androgen or estrogen levels are within normal limits or indicative of potential hormonal disorders or conditions. However,establishing accurate RIs involves studying a large, well-characterized, healthy population and considering many factors such as age, sex, and diurnal variations, etc. Among all the androgens and estrogens, only DHEA, DHEA-S, 4-AD, and T have been reported in more than three papers with over 300 participants. Table 5 shows the reference intervalranges of androgens and estrogens (if available) established by LC-MS/MS method from 2010 to the present, involving over 300 participants [39–43] (Supplementary Table 1). Table 5 lists the sample preparation method and LC-MS conditions for individual reference.
3.1. Androgen RIs in human serum and plasma
Androgen levels in healthy populations typically are at the concentration of μmol/L to nmol/L in human serum or plasma [39–43]. With sensitivity being a relatively minor concern, the high specificity and high accuracy of LC-MS/MS method are valued for improved diagnosticsin routine androgen analysis in clinical practice. To this end, the Endocrine Society proposed that “the best prospect for a gold stand (in testosterone testing) lies in extractions and chromatography followed by MS or MS/MS in which the chemical structure of the molecule measuredi s identified”. [44]
Table 5 shows that total T levels were measured in all studies, 4-AD levels were quantified in four studies, DHEA and DHEA-S were shown in three studies, and DHT levels were BQL or not available (NA) in most of the studies. The age ranges and sub-groups varied due to the limits of the studied population (Supplementary Table 1). Overall, the RI of total T in serum/plasma between the ages of 20–40 years was established at 7.17–33.70 nmol/L in healthy males and at 0.33–2.05 nmol/L in healthy females. Studies such as Damgaard-Olesen group [40] found no significant correlation between age and serum/plasma T concentrationsin men. Conversely, an age-related T decline has also been reported in some studies [41,43]. In most studies, the age-related decline in the concentrations of DHEA-S, DHEA, and 4-AD was observed [39–41,43] (Table 5 and Supplementary Table 1). Specifically, Kunz’s group [43] found a significant reduction in the upper reference limits across age groups. In males, there was a decrease from the youngest to the oldest age group by 73 % for DHEA, 44 % for DHEA-S, and 44 % for 4-AD, respectively. A similar result was obtained in females, post-menopause values showed decreases of 8 %, 12 %, and 46 % for DHEA, DHEA-S, and AD, respectively [43] (Supplementary Table 1)
3.2. Estrogen RIs in human serum and plasma
The significance of establishing RIs of estrogens across various life stages was emphasized by the Endocrine Society’s statement in 2013 that “The measurement of estradiol in biological fluids is important in human biology from cradle to grave.” [45] Today, laboratories areequi pped to measure E2 concentration at 184 pmol/L (50 pg/mL) in serum or plasma to support women undergoing fertility treatments. Moreover, the high sensitivity method becomes critical for monitoring postmenopausal women, with typical E2 levels near 55 pmol (15 pg/mL) [3,27]. For patients with breast cancer on aromatase inhibitors, an ultra-high sensitivity below 3.7 pmol/L (1 pg/mL) is required [45]. As shown in Table 4, the establishment of RIs of estrogens has not reached the level of success achieved with androgen (especially in T) assays [46]. Among all estrogens, only E2 was reported once in Mezzullo study [42] as a quantitative level at 80 pmol/L in males (20–70 years old), 395 pmol/L in females (21–51 years old). In other studies, E2 and other estrogen values were BQL or not included (X).
4. Advances in other analytical aspects of LC- MS
4.1. Derivatization
4.2. Chromatographic separation
5. Future perspective
It is essential to ensure consistency and accuracy in steroid measurements, particularly for establishing reliable RIs across different laboratories and platforms. The CDC’s Hormone Standardization Program (HoSt) and the National Institute of Standards and Technology (NIST) have aimed to unify analytical procedures. These efforts, along with Clinical and Laboratory Standards Institute (CLSI) guidelines, emphasize establishing new RIs, particularly with novel analytical methods. In addition, a standard reference material (SRM 971) was developed to address the need for improved accuracy of routine clinical assays [41].
A promising direction in analytical method development is enhancing the breadth and depth of steroid hormone profiling considering the diversity of steroid metabolism. Such advances are expected to generate complex data sets that are well-suit to metabolomics and steroidomics [61]. In hybrid instruments (e.g. quadrupole/time-of-flight (TOF) or quadrupole/Orbitrap), precursors were selected in quadrupole, and MS/MS scan of product ions can be performed in HR-MS,which could increases the specificity of the analysis. Normally the latest generation of triple-quadrupoles instrument would ensure the maximum sensitivity, but due to the unit resolution, it is possible that in some particular cases a PRM acquired on a HRMS actually offers more specificity, even at a minor loss of sensitivity. An approach has proposed the development of the high-resolution multiple reaction monitoring (MRMHR) method on a TripleTOF 5600 (AB Sciex) for quantification of circulating steroids in human and mice samples with confidence, precision, and accuracy [62]. Groessl’s group [63] reported using a Vanquish LC (Thermo) coupled with a QExactive Plus (Thermo) for the absolute quantification of 51 steroids for clinical analysis in human serum and peritoneal fluid for the first time. [63] Data acquisition on HRMS can be set under both parallel reaction monitoring (PRM) and fullscan (FS) mode, which allows simultaneously targeted and untargeted analysis. This approach opens novel possibilities for the post-analysis of clinical samples (ss shown by Snyder group in 2015 for the Girard derivatives of androgens) as the data can be examined for virtually any steroid even after data acquisition. This flexibility ensures that absolute quantification can be seamlessly implemented when reference standards are obtainable, positioning this technique as a valuable tool in clinical analytics.
Chromatographic development plays a crucial role in the analysis of steroids due to the complexity of steroid isomers, stero-isomers and isobaric compounds. Recent chromatographic advances have also seen the incorporation of multi-dimensional chromatography techniques. Multi-dimensional chromatography techniques, such as ion mobility spectrometry coupled with mass spectrometry (IM-MS) [64], have improved peak capacity and specificity in steroid analysis by providing additional separation. The recently introduced traveling wave ionmobility (TWIM) technique further enhances ion propulsion and separation. [65] For instance, ESI-TWIM-MS [66,67] has been successfully used to analyze steroid isomers and glucuronides of T and epi-T. Additionally, cryogenic IR spectroscopy offers structured IR fingerprints for clear molecular identification. A combination of HR-IMS with messenger-tagged IR spectrometry [68] has been applied to quickly identify steroid isobars and isomers, eliminating the need for recurrent analytical standards.
