mass spectrometric applications for diagnosing metabolic ...171886/fulltext01.pdf · endocrinology...

ACTA

UNIVERSITATIS

UPSALIENSIS

UPPSALA

2008

Digital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Science and Technology 426

Mass Spectrometric Applicationsfor Diagnosing Metabolic andEndocrine Diseases

MARK M. KUSHNIR

ISSN 1651-6214ISBN 978-91-554-7172-9urn:nbn:se:uu:diva-8658

�� !"#$� %�� &�� '�� (� %))* �� )��$ +� �,� �� +�� + -,��,�. /,� �� &�� !��,.

��

0��,�� . %))*. �� 1�� +� �� !�� . 1�� . �� %(. *) ��. ��.2 �3 4#*"4�"$$�"#�#%"4.

��"��+�� 5��6��7 �� 8�� &��,�� . /,�� ,�� + ��9��,��,� �� 5:�"� ;� 7 �� +� �� + �� . /,� �� ,�� "�� ++"�� 8�� + �,�� ,� �� +�� . 1�� ,�� &�� +�� +� �,� ��6�� &�� ,�� +,��,� �� ,��. 1�� + �,� :�"� ;� �� ,�9�� + �� + �� + �+�� ,��. 2 �,�� ,�� &� �� ,�� +� �,� �� + �,��+�� + �� ,� ��,�� ,�� . /�,�� ,��,�� + �,� :�"� ;� �� +� �,� ��6�� ,�� ,�� , &�� +� 9�� + �� +�� ,��,�� 6�. �� ,�� ,�� ,�� &� ��+�� + �,�� +�� + ,��,� &�� &�� &��, �� 5-�< 7. <�� ,� �� &�� ,�� ,� ��,&�� &�� ,��+�� +� �� + �,� ��,�,��. -�� 6�� + -�< &�� +�� ,� �,��= +��,�� &�� +�� ,�� .

� �� 6�� 9�� ,��,�� :�"� ;� �� ,�� ,��

�� ! � �� "�� #�� ! #�� !$% &''! �� ! �()*&+,- ��! ��

> ��6 �. 0��,�� %))*

2 3 �($�"(%��2 �3 4#*"4�"$$�"#�#%"4��"*($* 5,��;;��.6�.��;��?��@��"*($*7

Papers included in the thesis

I Kushnir MM, Rockwood AL, Nelson GJ. Simultaneous quantita-tive analysis of isobars by tandem mass spectrometry from unre-solved chromatographic peaks. Journal of Mass Spectrometry 2004;39:532–540.

II Kushnir MM, Rockwood AL, Nelson GJ, Yue B, Urry FM. As-sessing analytical specificity in quantitative analysis using tan-dem mass spectrometry. Clinical Biochemistry 2005;38:319–327.

III Kushnir MM, Nielson R, Roberts WL, Rockwood AL. Cortisol and cortisone analysis in serum and plasma by atmospheric pres-sure photoionization tandem mass spectrometry. Clinical Bio-chemistry 2004;37:357–362.

IV Kushnir MM, Rockwood AL, Roberts WL, Pattison EG, Bunker AM, Fitzgerald RL, Meikle AW. Performance characteristics of a novel tandem mass spectrometry assay for analysis of testos-terone in serum. Clinical Chemistry 2006;52:120–128.

V Kushnir MM, Rockwood AL, Roberts WL, Owen WE, Bunker AM, Meikle AW. Development and performance evaluation of a tandem mass spectrometry assay for four adrenal steroids. Clini-cal Chemistry 2006;52:1559–1567.

VI Kushnir MM, Rockwood AL, Bergquist J, Varshavsky M, Rob-erts WL, Yue B, Bunker AM, Meikle AW. High sensitivity tan-dem mass spectrometry assay for serum estrone and estradiol. American Journal of Clinical Pathology 2008;129:530-539.

VII Kushnir MM, Naessen T, Kirilovas D, Chaika A, Nosenko J, Mogilevkina I, Rockwood AL, Carlström K, Bergquist J. Steroid profiles in ovarian follicular fluid samples using high sensitivity tandem mass spectrometry methods. Submitted for publication.

VIII Naessen T, Kushnir MM, Chaika A, Nosenko J, Mogilevkina I, Rockwood AL, Carlström K, Kirilovas D, Bergquist J. Steroid profiles in ovarian follicular fluid in women with and without polycystic ovary syndrome. Manuscript.

All reprints were made with kind permission of the publishers.

The author’s contribution to the papers:

Paper I: Co-developed the algorithm, performed experiments, validated the method, and wrote the manuscript. Paper II: Developed and validated the methods, wrote manuscript Paper III: Performed experiments, developed the method, participated in the method validation, and wrote the manuscript. Paper IV: Performed experiments, developed and validated the method, determined reference intervals for adults and wrote the manuscript. Paper V: Performed experiments, developed and validated the method, de-termined reference intervals for adults and wrote the manuscript. Paper VI: Performed experiments, developed and validated the method, determined reference intervals for adults and children, wrote the manuscript. Paper VII: Performed experiments, performed statistical analysis and evaluation of the results and wrote the manuscript. Paper VIII: Performed experiments, participated in statistical analysis and evaluation of the results, participated in writing of the manuscript.

Papers not included in the thesis

Physical and analytical chemistry 1. Grekov SP, Kushnir MM, Kalusky AE. Dynamics of absorption by a

layer of an adsorbent accompanied by irreversible chemical reaction. Journal of Physical Chemistry (in Russian)1990;64:2572–2575.

2. Kushnir MM, Zhadan OA. Oxidation of nitrogen oxide into nitrogen dioxide with analytical purpose. Journal of Analytical Chemistry (in Russian) 1990;45:1990–1993 .

3. Lutsyk AI, Rudakov ES, Kushnir MM. Kinetics of nitrogen oxide oxida-tion by chrome oxide in aqueous solutions of acids and salts. Kinetics and Catalysis (in Russian) 1990;31:1309–1313.

4. Kushnir MM, Urry FM. Use of statistically designed experiment ap-proach to optimize propylketal derivatization of barbiturates. Journal of Chromatographic Science 2001;39:129–136.

5. Rockwood AL, Kushnir MM, Nelson GJ. Dissociation of individual isotopic peaks: predicting isotopic distributions of product ions in MSn. Journal of American Society of Mass Spectrometry 2003;14:311-322.

Industrial, forensic and clinical toxicology 6. Gladkov UA, Klasovsky NA, Kushnir MM, Zemsky BP. Nitrogen oxide

and nitrogen dioxide control in mine. Safety in Industry (in Russian) 1991;4:67–69.

7. Jennison TA, Brown PI, Crossett J, Kushnir MM, Urry FM. A rapid gas chromatographic method quantitating clozapine in human plasma or se-rum for the purpose of therapeutic monitoring. Journal of Analytical Toxicology 1995;19: 537–541.

8. Urry FM, Kushnir MM, Nelson G, McDowell M, Jennison TA. Improv-ing ion mass ratio performance at low concentrations in methampheta-mine GC-MS assay through internal standard selection. Journal of Ana-lytical Toxicology 1996;20:592–595.

9. Jennison TA, Brown PI, Crossett J, Kushnir MM, Urry FM. Potential for Overestimation of clozapine concentration. Journal of Analytical Toxi-cology 1997;21:73–74.

10. Urry FM, Komaromy-Hiller G, Staley B, Crockett DK, Kushnir MM, Nelson G, Struempler RE. Nitrite adulteration of workplace urine drug testing specimens I. Sources and associated concentrations of nitrite in urine and distinction between natural sources and adulteration. Journal of Analytical Toxicology 1998;22:89–95.

11. Kushnir MM, Crossett J, Brown PI, Urry FM. Analysis of gabapentin in serum and plasma by solid phase extraction and GC-MS for therapeutic drug monitoring. Journal of Analytical Toxicology 1999;23:1–6.

12. Kushnir MM, Crockett DK, Nelson G, Urry FM. Comparison of four derivatizing reagents for 6-acetylmorphine GC-MS analysis. Journal of Analytical Toxicology 1999;23:262–269.

13. Kushnir MM, Crossett J, Brown PI, Urry F.M. Analysis of gabapentin in serum and plasma by solid phase extraction and GC-MS. Newsletter SPEC/ANSYS 1999;4:1–3.

14. Coles R, Kushnir MM, Nelson GJ, McMillin GA, Urry FM. Simulta-neous determination of codeine, morphine, hydrocodone, hydromor-phone, oxycodone and 6-acetylmorphine in urine, serum, plasma, whole blood and meconium by LC-MS/MS. Journal of Analytical Toxicology 2007;31:1–10.

Endocrinology and metabolism 15. Kushnir MM, Komaromy-Hiller G. Optimization and performance of a

rapid GC-MS analysis for methylmalonic acid determination in serum and plasma. Journal of Chromatography B 2000;741:231–241.

16. Kushnir MM, Komaromy-Hiller G, Shushan B, Urry FM, Roberts WL. Analysis of dicarboxylic acids by tandem mass spectrometry. High throughput quantitative measurement of methylmalonic acid in serum, plasma and urine. Clinical Chemistry 2001;47:11 1993–2002.

17. Kushnir MM, Urry FM, Frank EL, Roberts WL, Shushan B. Analysis of catecholamines in urine by positive ion electrospray tandem mass spec-trometry. Clinical Chemistry 2002;48:2 323–331.

18. Kushnir MM, Shushan B, Roberts WL, Pasquali M. Acylcarnitines and vitamin B12 deficiency. Clinical Chemistry 2002;48:7 1126–1128.

19. Kushnir MM, Rockwood AL, Nelson GJ, Terry A, Meikle AW. Analy-sis of cortisol in urine by LC-MS/MS. Clinical Chemistry 2003;49:6 965–967.

20. Meikle AW, Findling J, Kushnir MM, Rockwood AL, Nelson GJ, Terry AH. Pseudocushings syndrome cause by interference of drug fenofibrate with HPLC assay for cortisol. Journal of Clinical Endocrinology and Metabolism 2003;88:3521–3524.

21. Liu A, Kushnir MM, Roberts WL, Pasquali M. Solid phase extraction procedure for urinary organic acid analysis by gas chromatography mass spectrometry. Journal of Chromatography B 2004;806:283–287.

22. Choi HH, Gray PB, Storer TW, Calof OM, Woodhouse L, Singh AB, Padero C, Kushnir MM, Rockwood AL, Meikle AW, Mac RP, Sinha-Hikim I, Shen R, Dzekov J, Dzekov C, Lee M, Hays RD, Bhasin S. Ef-fects of testosterone replacement in human immunodeficiency virus-infected women with weight loss. Journal of Clinical Endocrinology and Metabolism 2005;90:1531–1541.

23. Singh AB, Lee ML, Sinha-Hikim I, Kushnir MM, Meikle AW, Rock-wood AL, Afework S, Bhasin S. Pharmacokinetics of a testosterone gel in healthy postmenopausal women. Journal of Clinical Endocrinology and Metabolism 2006;91:136–144.

24. Miller JW, Garrod MG, Rockwood AL, Kushnir MM, Allen LH, Haan MN, Green R. Holotranscobalamin II measurement improves the predic-tive value of total plasma B12 in screening for B12 deficiency. Clinical Chemistry 2006;52: 278–285.

25. Rockwood AL, Kushnir MM, Meikle AW Low-level testosterone meas-urement by LC-tandem mass spectrometry. Clinical Laboratory Interna-tional 2006;30:20–23.

26. Anderson JL, Carlquist JF, Roberts WL, Horne BD, May HT, Schwarz EL, Pasquali M, Nielson R, Kushnir MM, Rockwood AL, Bair TL, Muhlestein JB.; for the Intermountain Heart Collaborative (IHC) Study Group. Asymmetric dimethylarginine, cortisol/cortisone ratio, and C-peptide: Markers for diabetes and cardiovascular risk? American Heart Journal 2007;153:67–73.

27. Meikle AW, Kushnir MM, Rockwood AL, Pattison EG, Terry AH, San-drock T, Bunker AM, Phanslkar AR, Owen WE, Roberts WL. Adrenal steroid concentrations in children seven to seventeen years of age. Jour-nal of Pediatric Endocrinology and Metabolism 2007;20:1281–1291.

28. Meier C, Nguyen TV, Handelsman DJ, Schindler C, Kushnir MM, Rockwood AL, Meikle AW, Center JR, Eisman JA, Seibel MJ. Endoge-nous sex hormones and incident fracture risk in older men: the Dubbo osteoporosis epidemiology study. Archives of Internal Medicine 2008;168:47–54.

29. Yue B, Rockwood AL, Sandrock T, La'ulu SL, Kushnir MM, Meikle AW. Free thyroid hormones in serum by direct equilibrium dialysis and on-line solid-phase extraction-liquid chromatography/tandem mass spec-trometry. Clinical Chemistry 2008 (in press).

Book Chapter 30. Kushnir MM, Frank EL. Analysis for urinary free catecholamines by

LC-MS/MS. In: Burtis CA, Ashwood ER eds. Tietz Textbook of Clini-cal Chemistry and Molecular Diagnostics. Philadelphia: W.B. Saunders Company (2005).

Contents

1 Introduction................................................................................................13 1.1 Biomarkers .........................................................................................13 1.2 Mass spectrometry role in clinical diagnostic laboratories ................15 1.3 Aims of the thesis ...............................................................................15

2 Sample preparation and liquid chromatography tandem mass spectrometry......................................................................................................................18

2.1 Sample preparation.............................................................................18 2.2 LC-MS/MS.........................................................................................19 2.3 Method development..........................................................................22 2.4 Quantitative analysis ..........................................................................23 2.5 Quantitation of isobars from unresolved chromatographic peaks ......24

3. Validation of mass spectrometry based tests ............................................29 3.1 Evaluation of method performance ....................................................29 3.2 Assessment of analytical specificity...................................................33 3.3 Quality management ..........................................................................34 3.4 Reference intervals .............................................................................35

4. Clinical diagnostic applications ................................................................37 4. 1 Biosynthesis of steroids and related diseases ....................................37 4.2 High sensitivity methods for endogenous steroids.............................38

4.2.1 Adrenal steroids ..........................................................................38 4.2.2 Glucocorticoids...........................................................................43 4.2.3 Androgens...................................................................................44 4.2.4 Estrogens ....................................................................................49

4.3 Ovarian steroidogenesis .....................................................................53 4.3.1 Steroids in ovarian follicles of healthy women ..........................53 4.3.2 Polycystic ovary syndrome (PCOS) and steroids in follicular fluid of PCOS patients .........................................................................56

5. Concluding remarks ..................................................................................59

6. Acknowledgements...................................................................................61

7. Swedish summary .....................................................................................62

8. References.................................................................................................65

Abbreviations

A4 Androstenedione ADF Androgen-dominant follicles AMR Analytical measurement range APCI Atmospheric pressure chemical ionization APPI Atmospheric pressure photo ionization AUC Area under curve CID Collision induced dissociation CoA Coenzyme A CV Coefficient of variation CAH Congenital adrenal hyperplasia CYP P450 enzyme DC Direct current DHEA Dehydroepiandrostenedione 11DC 11 Deoxycortisol E Cortisone E1 Estrone E2 Estradiol E3 Estriol EDF Estrogen-dominant follicles ESI Electrospray ionization F Cortisol FF Follicular fluid GC-MS Gas chromatography mass spectrometry 17OHProg 17-hydroxyprogesterone 17OHPregn 17-hydroxypregnenolone HPLC High performance liquid chromatography HSD Hydroxysteroid dehydrogenase IA Immunoassay IS Internal standard IVF In-vitro fertilization LC-MS/MS Liquid chromatography tandem mass spec-

trometry LOQ Limit of quantitation

LOD Limit of detection MMA Methylmalonic acid MRM Multiple reaction monitoring m/z Mass to charge ratio Pregn Pregnenolone PCOS Polycystic ovary syndrome RIA Radioimmunoassay RF Radio frequency RM Regularly menstruating ROC Receiver operating characteristic SA Succinic acid SD Standard deviation SHBG Sex hormone binding globulin S/N Signal to noise ratio SPE Solid phase extraction TS Tanner stage Te Testosterone QA Quality assurance QC Quality control ULOL Upper limit of linearity

13

1 Introduction

Clinical chemistry has its beginnings in XVIII and XIX centuries when a relationship between chemistry and living organisms was first realized. Many great chemists and physicians contributed to the field, and one of the first contributors was the Swedish chemist Jöns Jakob Berzelius (1779–1848), who published the first book on this subject (1, 2). The book “Lec-tures in Animal Chemistry” (1) summarized the chemical knowledge of that time related to animal tissues and body fluids. Many fundamental discover-ies and developments have been made since then, which enabled chemists to effectively assist physicians in detecting diseases, and now clinical laborato-ries have become an essential part of the practice of medicine. It has been estimated that medical laboratory testing plays a significant role in 60 to 70% of all decisions regarding establishing patients’ diagnoses, selection and monitoring of treatments (3). In part this is related to the fact that many diseases have similar clinical presentation, and testing often helps to differ-entiate between diseases; which in turn result in more efficient treatments and better outcomes.

1.1 Biomarkers Effectiveness of the diagnostic testing depends on the appropriateness of the utilized markers of diseases. Successful markers are features that could be objectively measured and evaluated as indicators of biological and patho-logical processes (4, 5). Biomarkers can be anatomic, physiologic or bio-chemical in nature and should be associated with a disease. In order to be medically useful, a biomarker should be detectable and measurable by objec-tive techniques such as physical examination, imaging or analytical meas-urement. The main emphasis of this thesis is on the development and evaluation of methods for measurement of biochemical markers of diseases. Biochemical markers are endogenous compounds, which are either not pre-sent in normal physiological state (e.g. tumor markers) or present within certain range of concentrations (e.g. intermediates and products of metabolic pathways). Biomarkers are important because accurate diagnoses and treat-ment monitoring make the foundation for successful outcomes. Biomarkers

14

may serve for early diagnostic needs, as indicators of severity of a disease, response to a treatment or to determine patient’s prognosis (Figure 1).

30 40 50 60 70 80

0.4

2

100

Anemia caused by B12 deficiency*

(reversible condition)

Upper normal

Moderately elevated

10 Severely elevated

Life-threatening

Con

cent

ratio

n,�m

ol/L

Time, years

Mild B12 deficiency*

Neuropathy**(irreversible disease)

Natural variation

Figure 1. Example of diagnostic and prognostic use of methylmalonic acid as bio-marker of vitamin B12 deficiency (paper I). * - diagnostic marker, ** - prognostic marker.

A variety of biomarkers utilized in contemporary human diagnostics, which range from DNA and RNA, to proteins, and small molecules. One of the main tasks of genetics is to predict difference in biological systems based on genetic variations. A single gene is translated into messenger RNA, which encode multiple proteins. The proteome is more complex than the genome, as protein expression with over 400 known posttranslational modi-fications may differ between different cells, at different time and physiologi-cal conditions. Knowledge of the gene and protein expression is very useful, but alone it is unable to detect and explain all phenotypic variations and changes. As a result of protein expression, many biochemical interactions and changes occur in living organisms and produce physiological functions that lead to formation of active and inactive metabolites. These metabolites along with proteins often serve as markers of disease and physiological con-ditions. Genetic and protein analyses provide information about systemic regulation on higher levels, while metabolites and intermediates of bio-chemical pathways are representative of physiological response to the regu-lation.

15

1.2 Mass spectrometry role in clinical diagnostic laboratories Modern clinical laboratories utilize diverse techniques and instrumentation that vary in reliability and specificity. Since the introduction of tandem mass spectrometry in clinical laboratories it proved to be one of the most specific analytical techniques available for clinical diagnostics. Clinical applications of tandem mass spectrometry could be subdivided in two groups: screening and target analysis (Figure 2). Typically, screening methods are intended for detecting multiple markers of diseases, drugs, or toxins (e.g. newborn screening for metabolic disease, toxicology screening). In these applications the goal is to achieve high throughput of the testing and a low false negative rate. In target analysis the main focus is on accurate and precise quantitation and the assurance of the analyte identity. In all analytical applications accuracy of measurements is important. Er-rors encountered in the clinical diagnostics, however, are especially costly compared to other fields, because they may lead to misdiagnosis, mistreat-ment, and even to the loss of life (6–8). Therefore highly specific methods are required for clinical diagnostic testing, and strict guidelines should be followed with respect to the quality control and management of pre- and post-analytic variables. The challenges of clinical diagnostic testing are related to the complexity of the biological samples, large diversity of classes of molecules present in the samples, variability of the sample matrices between individuals, and the wide range of concentrations of the constituents in the samples. Figure 3 [based on (10)] shows some of the clinically useful diagnostic biomarkers along with their biologically relevant concentrations, which span over ten orders of magnitude. Such diversity of concentrations suggest a need for using highly sensitive and specific instruments, enabling accurate measure-ment of minor sample constituents in the presence of excessive amounts of other endogenous substances.

1.3 Aims of the thesis The aims of the work in this thesis were to develop tandem mass spectrome-try based diagnostic tests for biomarkers of endocrine and metabolic disease, evaluating performance of the tests and establishing reference intervals for the biomarkers in serum samples of healthy adults and children. Major crite-ria for the developed methods (papers III–VI) were to achieve high sensi-tivity, specificity and robustness, allowing routine use of the methods for measurement of endogenous concentrations of the biomarkers in biological fluids. Other aspects of the thesis were related to the assessment of the ana-lytical specificity in methods using tandem mass spectrometry (paper II),

16

and development of an approach for high throughput analysis of isomers (paper I). The developed methods were further applied in a study of the steroidogenesis in ovarian follicles (papers VII, VIII) of healthy women, women undergoing ovarian stimulation for in-vitro fertilization (IVF) treat-ment, and women with polycystic ovary syndrome (PCOS).

B

5.4 5.6 5.8 6.0 6.2 6.4 6.6

Analyte Deuterium labeled internal standard

Time, min

*

*

***

*

*

*

A

4 6 8 10 12Time, min

Figure 2. Example of methods for screening (A) and target analysis (B). A - GC-MS chromatogram of a urine sample from a patent with maple syrup urine disease [* represents abnormal organic acids (9)], B - chromatogram of MRM transitions of estradiol and its internal standard (paper VI).

1.0E

-06

1.0E

-04

1.0E

-02

1.0E

+00

1.0E

+02

1.0E

+04

1.0E

+06

ChlorideCarbon dioxide

CholesterolGlucose

PotassiumCalcium

PhosphateMagnesiumGlutamine

AlanineGlycineProlineLysine

PhenylalanineLeucine

FluorideArginine

CreatinineCystine

HistidineIsoleucineVitamin C

AsparagineVitamin E

MethionineCopper

ZinkAcetylcarnitine

HomocysteineDHEA sulfate (male)

DHEA sulfate (female)Vitamin A

Beta-carotenePropionylcarnitine

SerotoninVitamin B2

Cortisol, totalMethylmalonic acid

Vitamin B1Vitamin D (25)

Vitamin B6Progesterone Luteal phase

DHEA (male)Testosterone total (male)

DHEA (female)Acrylylcarnitine

FolateCortisol, free

NormetanephrineTestosterone bioavailable (male)

NickelMetanephrine

Androstenedione17hydroxyprogesterone

11-deoxycortisolNorepinephrine

DHT (male)Progesterone (male)

Estradiol midcycle (female)Testosterone total (female)

Vitamin KEstrone midcycle (female)

Testosterone free (male)DHT (female)

Normetanephrine (free)Vitamin B12

DopamineAldosterone

Metanephrine (free)Epinephrine

Estrone (male)Estradiol (male)

Testosterone bioavailable (female)Vitamin D (1,25)Thyroxine (free)

Testosterone free (female)

�mol/L

F

igur

e 3.

Dist

ribut

ion

of c

once

ntra

tions

(med

ians

and

rang

es) o

f end

ogen

ous s

mal

l mol

ecul

es in

seru

m o

f adu

lts [b

ased

on

(10)

].

18

2 Sample preparation and liquid chromatography tandem mass spectrometry

2.1 Sample preparation One of the challenges in analysis of biological samples is the complexity of sample matrices. Sample preparation allows reducing complexity of the samples and is especially important for methods intended for routine testing because it improves ruggedness and specificity of the methods. Tasks com-monly performed during sample preparation include pre-concentration of the analyte, removal of the harmful matrix components and changing the sample solvent. In some cases it is helpful to modify an analyte by making its de-rivative and analyzing the derivative instead of the original compound (pa-pers II, IV–VI). The sample preparation usually becomes more critical when lower detection limits are required. The most common technique used for separating analytes from sample matrix is liquid-liquid extraction. Principles of the liquid-liquid extraction are based on the solubility, acid-base equilibrium, and partitioning of the analyte between aqueous and organic phases (papers III–VI). Variety of organic solvents, immiscible with biological fluids (e.g. ethyl acetate, hep-tane, hexane, methyl-t-butyl ester), could be used for liquid-liquid extrac-tion. Choice of a solvent may have effect on the specificity and efficiency of the extraction. Choice of pH of solution depends on the pKa of the analyte; and should be made with regard of shifting equilibrium in solution to the non-ionized form of the analyte. Solid phase extraction (SPE) is another commonly used method of separat-ing analytes from the sample matrix. In SPE an analyte is adsorbed by a stationary phase, the impurities should be either unretained or weakly re-tained by the adsorbent and need to be removed using weak solvents. After the wash the analyte should be eluted using a strong solvent (papers IV–V). SPE is based on the principle of frontal chromatography (11) where parti-tioning of sample components between the liquid and the stationary phases take place while sample is applied onto a column packed with an adsorbent. In contrast to the traditional chromatography, where discrete sample is in-jected in a column and the separation subsequently proceeds, in frontal chromatography sample is continuously applied. Frontal separation typi-cally has relatively low efficiency, but well designed SPE methods may re-

19

move large number of impurities and harmful matrix constituents making samples more amenable for the instrumental analysis. Traditionally, SPE is performed off line that makes it time consuming and labor intensive. The same principle as in traditional SPE is utilized in the on-line separation methods (12, 13). The on-line separation was used in the method for high sensitivity analysis of estrogens (paper VI). The on-line purification combined with sample preparation performed in 96-well plates led to significant reduction in sample preparation time and labor; and im-proved the detection limits as well. Analytical derivatization is another technique that may be used during sample preparation for LC-MS analysis. The derivatization is often used for making analytes amenable for the analysis and may be useful for improving ionization efficiency (14–18, papers IV–VI), elimination of the interfer-ences (19), increasing molecular weight of a parent ion (20, 21), changing polarity of the detection (15, 19, paper I), or fragmentation pattern (19). Derivatization allowed faster analysis of methylmalonic acid (19, paper I); reducing sample volume required for testing of testosterone (paper IV), adrenal steroids (paper V) and estrogens (paper VI); detection and analysis of steroids which were otherwise undetectable at physiological concentra-tions (paper V), and analyzing minute samples (40 �L) of ovarian follicular fluid (papers VII, VIII) for fifteen endogenous steroids.

2.2 LC-MS/MS

Mass spectrometry is a technique that enables characterization of molecules according to their mass-to-charge ratios (m/z). Mass analysis can be based on different physical principles, but essential components of all types of mass spectrometers are the same: a sample introduction system, an ion source, a mass analyzer that separates ions based on their m/z, a detector, a data collection computer and data analysis software. The quadrupole mass filter was developed in the early 1950's by Paul and Steinwedel (22) and is based on the mathematical theory of Mathieu (23), which defines dimensions and operating conditions of a quadrupole. Mass separation in a quadrupole is based on achieving stable trajectory of ions in an electric field. A quadrupole mass spectrometer consists of four parallel rods located in a vacuum chamber (22, 24). The vacuum required for in-strument operation depends on the dimensions of the mass analyzer and should allow ions to travel the distance from the ion source to the detector without collisions. The DC and RF voltages are applied to the pairs of rods located opposite to each other, creating an electric field in the space between the quadrupoles where ion separation takes place. The DC voltages applied

20

to the pairs of rods are opposite in polarity and the RF voltages are shifted by 180�� A combination of different DC (VDC) and the RF voltages (VRF), and RF frequency allows only ions with certain m/z to have stable trajectory while flying through the quadrupole; all other ions do not have stable trajec-tory and thus collide with the rods. The mass resolution of a quadrupole mass analyzer depends on the number of cycles that the voltage changes while the ions are traveling through the quadrupole, and the stability region defined by the frequency of VRF and the ratio VRF/VDC. As the ions spend longer time to travel through the mass analyzer, the mass resolution in-creases but the transmission efficiency decreases. Because of this, compro-mises between the sensitivity and the mass resolution need to be made dur-ing the instrument operation. The ratio VRF/VDC is commonly set to allow ions within 1 Da window to pass through the quadrupole. Main advantages of the quadrupole mass analyzers include efficient ion transmission, high sensitivity of detection and the potential to couple multiple quadrupoles to one another or to other types of mass analyzers. Other advantages of quad-rupole mass analyzers include independence of the separation from the ini-tial spatial distribution of the ions and the initial kinetic energy. Because of the relatively simple design and relatively low cost, quadrupole mass ana-lyzers have become one of the most widely used types of mass spectrome-ters (24, 25). LC-MS is a hyphenated technique, in which samples are first separated using liquid chromatography (LC), and the effluent from the chroma-tographic column is transferred for separation into a mass spectrometer. The LC separation may be performed according to one of such properties as hy-drophobicity, size of the molecules, presence of specific functional groups, etc. Parameters that need to be considered for optimization of the chroma-tographic separation include choice of the stationary phase, particle size, dimensions of the chromatographic column, flow rate, mobile phase compo-sition, temperature, and compatibility of the mobile phase with the condi-tions required for efficient ionization of the analyte in the ion source of the mass spectrometer (26). A current trend in methods using mass spectrome-try detection is to use short columns with small internal diameter, packed with smaller particles and performing chromatographic separation under higher pressure (paper VI). The above conditions result in achieving faster analysis with flow rates optimal for the ion source operation. During LC-MS analysis a series of mass spectra are continuously acquired. The mass spectrometer detects ions, which are registered as m/z, with inten-sity of the signal proportional to the amount of analyte present in the sample. The combination of chromatography and mass spectrometry enables synergy effect by overcoming weaknesses of each individual technique with regards to the quality of the obtained information. Chromatography resolves indi-vidual components of the samples in time, while mass spectrometry selec-tively detects these components. In LC-MS analysis sample constituents are

21

characterized by two physical properties: the compound-specific retention time and the m/z ratios corresponding to individual components of a sample. The analytical methods described in this thesis (papers I–VIII), utilize detection with triple quadrupole (or tandem) mass spectrometry (MS/MS). The MS/MS by design consists of three components: two mass spectrome-ters with a collision cell placed in the middle (Figure 4). The first mass spectrometer selects ions of a specific m/z (parent ions), which then are di-rected into a collision cell where the parent ion is fragmented through colli-sions with neutral gas molecules (27,28). The generated fragments (product ions) are directed into the second mass spectrometer, where the ions get resolved based on their m/z and then detected. Data collected during the LC-MS/MS analysis contain three dimensions: retention time, mass-to-charge ratio of the parent ion, and mass-to-charge ratios of the product ions. The masses of the parent and the product ions represent fundamental properties of the molecules: their molecular weight and the structural fragments. Reli-ance on these fundamental properties is the basis for the high degree of ana-lytical specificity of LC-MS/MS as a technique.

Q0 Q1 Q2 Q3

L

D

Figure 4. Schematic diagram of triple quadrupole (tandem) mass spectrometer: Q0 - RF only quadrupole,Q1 - first mass analyzer, Q2 - collision cell, Q3 - second mass analyzer, L - exit lens, D - detector.

Tandem mass spectrometers may acquire data in various modes of opera-tion: parent ion scanning (first mass analyzer selects an ion, which is frag-mented in a collision cell, and second mass analyzer scans over m/z range of interest), precursor ion scanning (the mirror image of the product ion scan: first mass analyzer scans over the range of interest, while the second mass analyzer is fixed at a chosen m/z), neutral loss (both mass analyzers are scanning with a fixed mass difference and only the ions that lose specific mass in the collision cell get detected), or alternatively both mass analyzers could be fixed on transmission of the compound-specific parent and product ions (14, 24, 25, 29). The first three modes are typically used to detect and identify unknown compounds in a sample, or to screen samples for classes of compounds. The last mode is known as multiple reaction monitoring (MRM) and it is mainly used in the quantitative targeted analysis. The MRM mode of operation is highly selective and sensitive because the mass

22

analyzers transmit only ions characteristic of the target analyte and remove the majority of chemical noise. The combination of liquid chromatography with MS/MS operated in MRM mode represents a technique with one of the highest achievable analytical specificity and the most sensitive type of mass spectrometric detection. All analytical methods developed in this thesis used MRM mode of data acquisition.

2.3 Method development Due to the nature of the clinical diagnostic testing, acceptable performance of the methods is required before they could be used for diagnostic purposes. This translates to significant amount of time and efforts required for method development, validation, and support. Goals for the method’s performance characteristics need to be established prior to starting the method development. Method development is usually include optimization of all individual steps involved in sample preparation, chromatographic separation, ionization and mass spectrometric detection; followed by optimization of the entire process of the analysis along with selection of the laboratory equipment, reagents, standards and reference materials. Extensive method optimization is particularly required when targeted sen-sitivity is approaching limits of the instrument’s capabilities. Some of the pitfalls associated with the LC-MS/MS methods are related to isobaric inter-ferences and matrix effects caused by endogenous and exogenous constitu-ents of biological samples. When targeted sensitivity and specificity cannot be achieved using common approaches, such techniques as high efficiency chromatographic separation, enhanced mass resolution (30), selective ioniza-tion (papers I, III), chemical derivatization (14–17, papers IV–VI), multi-stage mass spectrometry (MS3), and orthogonal separations [e.g. multidi-mensional chromatography (31), ion mobility (32)] can be used. The above techniques could be useful for eliminating interferences, improving the sig-nal to noise ratio and could be used instead of more complex sample prepa-ration. Many of the fine points of the method development depend on the chemical properties and structure of the targeted molecules. Papers I–VI provide examples of the enhancement of the methods’ performance based on the analyte-specific properties. The final step of the method evaluation is its validation, which determines acceptability of the method’s performance for diagnostic use.

23

2.4 Quantitative analysis Quantitative analysis is typically performed using MRM mode of LC-MS/MS operation using quantitative calibration and stable isotope (2H, 13C or 15N) labeled internal standards. The purpose of the quantitative calibra-tion is to relate observed signal intensity to concentration of the analyte. In order to normalize the signal intensity, the same amount of internal standard is added to the calibration standards and the samples, and the calibration is established as a regression equation of the dependence between the relative concentrations of the analyte and the internal standard, and their relative peak intensities. Concentration of analyte in the samples is calculated using the calibration curve and the relative intensity of the peaks of the analyte and the internal standard. Performance of quantitative methods depends on a proper choice of the internal standard. The internal standard should be a stable non-endogenous substance of high purity and have chemical proper-ties similar to the target analyte. This allows to compensate for loses of the analyte during the sample preparation. The similarity in the chromatographic retention compensates for the ion suppression, which is one of major pitfalls of the LC-MS methods. Since chemical composition of the internal standard is identical to the target analyte (in the cases when stable isotope labeled internal standards are used), it behaves as the target analyte during the sam-ple preparation and analysis. Because of the difference in molecular weight, the mass spectrometer selectively detects signal from the analyte and the internal standard. In papers I–VIII deuterium labeled analogs of the tar-geted analytes were used as the internal standards. One of the pitfalls associated with isotopically labeled internal standards is potential interference with analysis, in cases when the internal standard was inappropriately selected (33). Isotopically labeled internal standard should have a sufficient number of isotope atoms to increase its molecular weight above the molecular weight of the naturally occurring isotopes of the target analyte (typically for small molecules composed of carbon, hydrogen, oxy-gen and nitrogen, 3 to 5 isotope atoms are sufficient for adequate separa-tion). This is to avoid potential interference cause by the overlap of the iso-topic ions of the analyte with the molecular ion of the internal standard.

24

2.5 Quantitation of isobars from unresolved chromatographic peaks Typical applications of tandem mass spectrometry are based on detecting unique mass transitions derived from chromatographically resolved peaks. When similarly fragmenting molecules with identical parent and product ions are present in a mixture, it is common practice to chromatographically separate them. Few methods have been developed for relative quantitation of isomers from mixtures without chromatographic separation (34–37). These methods assume equal sensitivity to all isomers and the ability to cal-culate relative concentrations of the isomers as a percent of total, however, none of the methods allows determining absolute concentration of the com-ponents of a mixture. In paper I we developed a method for absolute quan-titation of isomers from unresolved chromatographic peaks. The approach is based on an assumption that when similarly fragmented molecules are pre-sent in a mixture, the observed mass spectrum represents a linear combina-tion of the mass spectra of pure components, added in proportion corre-sponding to their relative concentrations. Product ion mass spectra of iso-mers commonly have identical characteristic product ions, but in some cases conditions can be found at which the isomers will have different relative intensities of the product ions. The difference in the relative intensity of the product ions forms basis of the proposed method. Mass spectrum of a mixture (X ) can be represented as a function

� �CSRfX s ,,� (Eq 1) where Rs represents reference mass spectra of the pure components; S repre-sents relative sensitivity; and C represents concentration of the components present in the mixture. In practice, X is observed in an experiment, Rs is determined from acquired spectra of pure components, S could be deter-mined using calibration standards of known concentration, and C is an un-known. The developed algorithm (paper I) allows determining C (Eq 1) from measured mass spectra of a mixture. The algorithm is based on the following assumptions: (i) no peaks of the same mass transitions other than originating from the isomers are present under the target peak; (ii) the total acquired signal is a linear combination of signals from the coeluting isomers; and (iii) branching ratios of the monitored mass transitions are significantly different among the isomers (38). Based on these assumptions we proposed a model (Eq 2), describing the relationship between the signal intensities of the mass ions and the observed mass spectrum of a mixture. The model is a system of linear equations:

25

aiaaA IIII �� 21

bibbB IIII �� 21

:

:

jijjJ IIII �� 21 (Eq 2)

bi

aii I

IA �

ji

bii I

IB �

:

:

ji

aii I

IJ �

where JBA III ,, represents total experimental intensities of the signal of the monitored mass transitions; jibiai III ,, represents intensity of the signal contributed from individual components of the mixture; iii JBA ,, repre-sents branching ratios of the mass transitions of the pure analytes. The system of equations (Eq 2) has a unique mathematical solution when the number of equations is equal to the number of unknowns. The number of the product ions needed for each compound should be equal to the num-ber of potentially coeluting isomers. This means that for two coeluting iso-mers two mass transitions should be used and a system of 4 equations need to be solved; for n coeluting isomers, n transitions should be used, and a system of n2 equations need to be solved. The above system of equations was solved for mixtures of two and three isomers and the formulas for the calculations were derived (paper I). The developed method was used for the quantitative analysis of isomers, methylmalonic (MMA) and succinic (SA) acids, in human serum and urine. MMA is a metabolic intermediate in the enzymatic conversion of propionic acid to SA, where vitamin B12 serves as a cofactor (Figure 5) and SA is the final product. If a person has deficiency of one of the enzymes involved in the transformation of MMA-CoA to SA-CoA, or deficiency of vitamin B12, or deficiency of one of the carrier proteins transporting vitamin B12; the pathway would be blocked (39, 40). The above conditions result in accumu-lation of MMA, which serves as a biomarker of this group of diseases. The

26

method for analyzing MMA that we earlier developed (19) is selective to dicarboxylic acids and the only compound present in biological fluids that may coelute with MMA is SA. The difficulty of analyzing MMA in the presence of SA, is related to the lack of fully unique mass transitions in the mass spectrum of MMA. The method for analyzing MMA (paper I) is based on the difference in the relative intensities of the mass transitions of MMA and SA (a1/b1 and a2/b2 in Figures 6D and 6E) and allows quantitaion of MMA and SA without chromatographic separation.

O O H

O

HO

O

HO

O

OH

Methylmalonyl CoA

Methylmalonic acid Succinic acid

Succinyl CoA Propionyl CoAMMR, MUT

Vitamin B 12

O

OH

Propionic acid

Fatty acids with odd number of

carbons

Threonine

Isoleucine Methionine , Valine

Figure 5. Pathway of degradation of fatty acids with odd number of carbons and 4 amino acids. Lack of vitamin B12 (cofactor in enzymatic conversion of Methylmalonyl-CoA to Succinyl-CoA), proteins participating in B12 transport, or enzymes MMR (D,L-methylmalonyl-CoA racemase) or MUT (methylmalonyl-CoA mutase) result in accu-mulation of methylmalonic acid in tissues and body fluids.

27

Isomersseparated

Isomers coelute

70 90 110 130 150Time, s

500

1000

Q1/Q3 transitions, amu

a2

b2


b1

a1


BA

5 15 25 35 45Time, s

1500

750

C

D

FG

E

Figure 6. Illustration of the approach for quantitation of isomers [methylmalonic (MMA) and succinic (SA) acids] from unresolved chromatographic peaks (paper I). A, B - mass transitions of mixture of the coeluted peaks of MMA and SA (G); C - chroma-togram of resolved peaks of MMA and SA; F - chromatogram of coeluted peaks of MMA and SA; a1, b1 - mass transitions of MMA (D); a2, b2 - mass transitions of SA (E). Intensity of the ions corresponding to the isomers (a1, b1, a2, b2) calculated from the intensities of the ions A and B (G) and relative intensities of the mass transitions of pure isomers (calculations performed using formulas from paper I).

Figure 6 illustrates the principle of the method for analyzing isomers from unresolved chromatographic peaks. Figures 6C and 6F show chromatograms of MMA and SA in the methods with (Figure 6C) and without (Figure 6F) chromatographic separation of the peaks. Times for analysis were 160 and 50 s, respectively. Calculations of the intensities corresponding to each iso-mer were performed using intensities of two mass transitions of the coeluted peaks (G) and relative intensities of the mass transitions of the pure stan-dards of MMA (a1/b1) and SA (a2/b2) using formulas derived in paper I. Concentration of MMA is determined utilizing the ratio of the calculated peak intensity of MMA to the intensity of the peak of the internal standard (d3-MMA) and the calibration curve. The method was compared to the con-ventional analysis utilizing chromatographic separation of MMA and SA using a set of 235 human serum samples (Figure 7). The comparison showed good agreement between the calculation-based method and the method using chromatographic separation of MMA and SA. Because this method does not require chromatographic separation of the isomers it pro-duces a 3-fold improvement in the throughput of the LC-MS/MS instrument. Vitamin B12 deficiency is a common condition, especially in elderly people

28

(prevalence ~15%); the reduction in the analysis time resulted in increased throughput of the instrument, and increased availability of the test for screening of population for this disorder.

y = 0.999x + 0.005r = 0.985

Syx = 0.024n=235

0.25

0.50

0.75

0.25 0.50 0.75

Conventional LC, �mol/L

No

LC s

epar

atio

n +

deco

nvol

utio

n, �

mol

/L

Figure 7. Comparison of the results of analysis of methylmalonic acid (MMA) with and without chromatographic separation from succinic acid (SA). Concentrations of MMA in the method without chromatographic separation were calculated from total peak intensities of MMA and SA utilizing formulas derived in paper I.

29

3. Validation of mass spectrometry based tests

3.1 Evaluation of method performance Validation experiments for mass spectrometry based diagnostic tests com-monly include evaluation of the precision, sensitivity, linearity, accuracy and analytical specificity (41, 42). A flowchart of the sequence of the experi-ments is shown in Figure 8. To avoid systematic errors during the validation it is preferable to utilize calibration and control materials from a commercial source (ideally traceable to a certified reference material) and to prepare standards and controls in amounts sufficient for the length of the entire validation. As shown in Fig-ure 1, physiological concentrations and the degree of the biological variation vary extensively between the markers. Thus, targeted values for sensitivity, accuracy, precision and the analytical measurement range should be based on the clinical needs of each individual marker (43, 44). Greater precision and accuracy are usually required for the methods intended for measurement of the markers with a narrow distribution of physiological concentrations (e.g. free T4, sodium). Commonly used requirements for LC-MS/MS meth-ods intended for use in clinical diagnostics are listed in Table 1.

Precision (often referred as “imprecision”) is one of the main characteris-tics of a method and represents the extent to which the results of replicate sample analysis agree. Evaluation of imprecision is performed using repli-cate analysis within run and between runs. The total imprecision of the measurements is then calculated based on these values. Imprecision is usu-ally evaluated using at least three samples prepared at concentrations around, below and above the clinical decision level. The samples should be prepared in a matrix similar to the targeted clinical samples and analyzed in replicates of three to five on at least five separate occasions. Figure 9 shows an exam-ple of the results of the evaluation of imprecision of an assay (paper VI).

Valid

atio

n su

mm

ary

Prep

are

stand

ards

an

d co

ntro

ls

With

in - run

impr

ecisi

on

Betw

een - r

un

impr

ecisi

on

Met

hod

impr

ecisi

on

Analy

tical

sens

itivi

tyLi

near

ityM

etho

d ac

cura

cy

Analy

tical

spec

ificit

y

Inte

rferin

g su

bsta

nces

Mat

rix

effe

cts

Ion

supp

ress

ion

Spec

imen

re

quire

men

ts

Type

of

spec

imen

Sam

ple c

ollec

tion

cont

ainer

,pr

eser

vativ

es Stor

age

stabi

lity

Com

pare

to

refe

renc

e m

etho

ds

Com

pare

to

othe

r com

mon

te

chniq

ues

Evalu

atio

n of

ca

rry-o

ver

Esta

blish

refe

renc

eint

erva

ls

Clini

cal

sens

itivi

ty/

spec

ificit

y Met

hod -

spec

ific

requ

irem

ents

Initi

al

prot

ocol

Certi

fied

refe

renc

e m

ater

ials

Tota

l im

prec

ision

Analy

tical

mea

sure

men

t ra

nge

F

igur

e 8.

Seq

uenc

e of

the

exp

erim

ents

for

valid

atio

n of

mas

s sp

ectro

met

ry b

ased

met

hods

int

ende

d fo

r us

e in

clin

ical

dia

gnos

tics.

31

Table 1. Commonly used criteria for the performance characteristics of LC-MS/MS methods intended for use in clinical diagnostics.

Imprecision, %

Accuracy, %

Qualitative identification

Comments

Precision �10 na yes na

LOQ �15 �80 yes At least ½ of the concentration corresponding to the clinical deci-sion level, s/n � 5

LOD na na yes Quantitation may not be accurate, s/n � 3

ULOL �10 �85 yes Highest concen-tration at which accuracy and imprecision are acceptable

AMR �10 �85 yes Should overlap clinically relevant range of concen-trations

Carry-over

na na na Less than LOD

0

20

40

60

80

100

0 2 4 6 8 10 12 14 16

Level I

Level II

Level III

Mean

+/- 2SD

Measurement

Con

cent

ratio

n, p

g/m

L

Figure 9. Evaluation of the imprecision of LC-MS/MS assay for estrone (paper VI).

32

The lowest concentration that could be reliably detected by a method is the limit of detection (LOD). At the LOD an analyte needs to be qualitatively identified and the signal-to-noise (s/n) ratio should be greater than 3. The limit of quantitation (LOQ) is the lowest concentration at which an analyte could be qualitatively identified and the quantitation is accurate. The upper limit of linearity (ULOL) is the highest concentration, below which quantita-tion is accurate. The analytical measurement range (AMR) of a method represents concentrations between the LOQ and the ULOL of the method. Quantitation should not be performed outside of the AMR. Samples for evaluation of sensitivity are often prepared from a patient sample, containing measurable concentration of the analyte. The sample should be serially diluted using an analyte-free matrix and analyzed in repli-cate. Potential problems with this approach could be encountered while ana-lyzing endogenous compounds since it may be difficult or impossible to obtain analyte-free samples in the physiological matrix. When this occurs, one may resort to selectively removing the analyte from the matrix or using an artificial matrix. Samples for the evaluation of the linearity are often prepared using two patient samples, one with low concentration of the target analyte and another one with highly elevated concentration. The samples should be mixed in different proportions to obtain at least five samples covering the concentra-tion range of interest and analyzed in replicate. The upper limit of linearity is the highest measured concentration at which the accuracy is within 15% of the expected value. Before implementing a mass spectrometry based assay in diagnostic use, it should be compared to other available assays by analyzing patient sam-ples. Concentrations of the analyte in the samples should be evenly distrib-uted over the measurement range, should overlap with the physiologically relevant range and the clinical decision levels. Use of patient samples (over analyte-supplemented matrix) is preferred because it allows identifying po-tential problems of the assay. The agreement between the methods should be assessed using Deming regression (45) and is considered acceptable when the slope of the regression line is statistically not different from 1.0, the in-tercept is statistically not different from zero and the correlation coefficient is greater than 0.95. Constant or proportional biases, poor precision or inter-fering substances affecting one or both methods may cause disagreement between the methods. In cases when the methods do not agree, the same samples could be retested by the evaluated method (to reduce the impreci-sion of the measurements); then the mean of two measurements can be com-pared to the comparative method. If the original comparative method is suspected of causing the disagreement, an alternative comparative method could be selected. To further investigate the problem, discrepant samples may be retested using more efficient chromatographic separation in order to

33

evaluate if the interfering peaks were present under the peaks of the analyte or the internal standard. In some cases agreement between methods cannot be reached. For exam-ple, it is not unusual for mass spectrometric methods to produce lower val-ues compared to immunoassay-based methods; often as a result of cross-reactivity in the immunoassays (46). When working with biological samples handling and storage conditions are important aspects of the analysis. Important factors that should be deter-mined during the method development are (i) type of collection tubes that should be used for the samples, (ii) sample types suitable for the test (e.g. serum, plasma, urine), (iii) acceptable anticoagulants and preservatives, and (iv) stability of the analyte during storage. Samples that were improperly collected or stored may be compromised and this may lead to erroneous results of the testing. Papers III–VI include data on the performance evaluation and validation of methods for analysis of endogenous steroids. These methods were exten-sively validated to justify their use for clinical diagnostics and were com-pared to commonly used methods of other clinical laboratories.

3.2 Assessment of analytical specificity Well-developed tandem mass spectrometry based methods are more specific than other analytical techniques. Additional advantages of tandem mass spectrometry based methods are related to the ability of simultaneous analy-sis of multiple analytes and the ability to assess specificity of analysis in every sample. Assessment of the specificity could be accomplished through monitoring multiple mass transitions of analytes (papers I–VI), or evalua-tion of full product ion mass spectra of the parent ions. In paper II we pro-posed various approaches and the acceptability criteria for the assessment of specificity in methods using tandem mass spectrometry detection. Compounds potentially interfering with tandem mass spectrometry detec-tion could be isobars of the target analyte; or isotopic analogs and adducts of the impurities, which become isobaric to the analyte or the internal standard. A substance could interfere with an analyte of interest if its peak coelutes with the target analyte and it has the same characteristic parent and product ions (19, 47, 48). A list of potentially interfering substances for newly de-veloped methods could be compiled by searching for isobars of the target analytes in mass spectral databases. The experiments for evaluation of the interference potential should include the identified isobars of the target ana-lyte, compounds structurally related to the analytes, common endogenous sample constituents, drugs, that might be administered at the targeted physiological condition and the drug metabolites (papers I–VI, 48). Mass transitions used in a method should be extensively evaluated to assure that

34

only the target analyte is measured and that the method does not suffer from interferences. As part of the evaluation of the specificity and robustness of the methods large number of patient samples should be analyzed and results should be evaluated for signs of interference. Other types of commonly recognized interferences that affect LC-MS methods, are impurities introduced with solvents and reagents, and ion sup-pression (49–53). Some of the impurities present in the solvents and re-agents may cause changes in the ionization efficiency, cause chemical noise, loss of sensitivity and even degradation of the analytes (54, 55). Therefore, high purity reagents and solvents are preferred for the methods using mass spectrometric detection. Several methods could be used to determine if sample matrix affects performance of a method (49–52). Common ap-proaches for the evaluation of ion suppression include post-column infusion of the target analyte in the effluent of the chromatographic column while analyzing patient samples (52, 53). Negative peaks eluting on chromatogram at the retention time of the analyte of interest are signs of the ion suppres-sion. Another approach for evaluation of ion suppression is based on direct comparison of signal intensity of the analyte measured in samples without matrix and in different matrices (51–53).

3.3 Quality management Maintaining high quality diagnostic testing requires monitoring the entire process of the analysis including the pre-analytical, analytical, and post-analytical stages. This is achieved by implementing quality assurance (QA) and quality control (QC) programs (56). Goals of the QA are to establish policies and practices necessary for reliable performance of the entire proc-ess of testing. The QC represents a set of techniques and procedures, aimed at monitoring assay performance and recognizing potential problems related to the analytical and post-analytical phases of testing. Major tasks for a QC program are to provide information on methods’ performance and to assure acceptable quality of the results. In order to assess performance of a method, QC samples with known concentration of the analyte are analyzed with every set of patient samples. The acceptability of the results of QC samples should always be evaluated prior to accepting results of the clinical samples.

35

3.4 Reference intervals Results of diagnostic testing are not clinically useful unless they are related to appropriate reference values. The reference values could be biomarker concentrations observed in samples from the general population of healthy people, people with specific disease, or earlier test results of the same indi-vidual. The reference values help establishing the basis for clinical interpre-tation of the test results (57–60). Population-based reference values gener-ally obtained from a well-defined group of individuals, which resemble the targeted population in all aspects except presence of the disease or condition for detection of which the testing is performed. The conditions under which samples for establishing reference values are obtained, collected and proc-essed should be analogous to the targeted clinical application. All the testing should be performed using standardized methods and appropriate QC. In cases when the distribution of the concentrations is Gaussian, reference in-tervals usually based on estimates of the population parameters, the mean and the standard deviation. If the reference distribution is not Gaussian and cannot be transformed to the Gaussian (as often the case for endogenous biomarkers), nonparametric methods can be used (61). Validated and well-characterized mass spectrometry based methods are often more specific compared to other commonly used techniques. Because of this, implementing these methods in diagnostic practice requires estab-lishing new reference intervals, which in many cases are different compared to the reference intervals used with older techniques (46, 62–64, papers IV–VI). Reference intervals could be specific to age, gender, ethnic origin, physiological state, collection time and need to be established prior to the clinical diagnostic use of a method. Other important values that should be determined prior to the diagnostic use of a method are the clinical decision levels. The clinical decision levels are concentrations of an analyte at which the diagnosis could be established or the management of the patients needs to be changed. The clinical decision levels are usually established based on the reference intervals, clinical information and epidemiological studies of patients having disease targeted by the test. In papers III–VI reference intervals for concentrations of endogenous steroid hormones in blood were established for women, men and children of different ages and Tanner stages (TS, stage of sexual development). In papers VII and VIII concentrations of endogenous steroids were determined in ovarian follicular fluid samples of healthy women and women with polycystic ovary syndrome. Usefulness of diagnostic biomarkers depends on the ability to distinguish samples of people having disease from samples of healthy individuals. Some of the factors, effecting usefulness of biomarkers are natural variation of the concentrations within and between individuals, and the magnitude of the difference of concentrations in population with and without the disease. The diagnostic tests have better clinical value for the markers for which

36

there is no (or minimal) overlap in the reference intervals between healthy individuals and affected patients. In cases when the distributions are par-tially overlapped, other factors (clinical symptoms, complementary diagnos-tic testing, medical risk related to a false positive result relative to the risk of a false negative result, etc.) should be taken into consideration.

37

4. Clinical diagnostic applications

4. 1 Biosynthesis of steroids and related diseases Steroid hormones are synthesized from cholesterol (Figure 10) through a series of enzyme-controlled reactions (65, 66). The rate-limiting step in the biosynthesis is conversion of cholesterol to pregnenolone (Pregn). Pregne-nolone is formed on the inner membrane of mitochondria and then trans-ferred back and forth between the mitochondria and the endoplasmic reticu-lum for further enzymatic conversions to other steroid hormones (66). Regu-latory enzymes in the pathway belong to two classes, the P450 (membrane-bound proteins CYP11, CYP17, CYP19, CYP21, etc); and the hydroxyster-oid dehydrogenases (HSD, short-chain alcohol dehydrogenase-reductases, 3HSD, 11HSD, 17HSD, etc). The difference between P450 enzymes and HSD’s is that all the P450 enzymes are products of a single gene, while the HSDs are products of separate genes (66). The isoforms and isoenzymes vary in tissue distribution, sub-cellular localization, catalytic activity, sub-strate and cofactor specificity. Because of the difference in enzyme distribu-tion between tissues, steroids have different primary sites of production, resulting in the ability of biosynthesis of many of the steroids in more than one tissue.

Cholesterol

Pregnenolone

Aldosterone

Corticosterone

11-deoxycorticosterone

Progesterone

17-OH-pregnenolone

17-OH-progesterone Androstenedione

Testosterone

Dehydroepiandrosterone

11-deoxycortisol

Cortisol

Estrone

Estradiol

Cortisone

Estriol

Androstanedione

Dihydrotestosterone

CYP17

CYP17CYP17

CYP17

3HSD 3HSD 3HSD

CYP21 CYP21

CYP11b2 CYP11b1

11HSD2 11HSD1

5a Reductase

CYP19

CYP19

17HSD17HSD17HSD

Bile acidsLiver

Various tissues

HO

H

H

H

O

O OH

OHHO

H

H

H

O

OH

O

HO

H

H

H

O

OH

H

H

H

HO

OH

H

H

H

Figure 10. Biosynthesis of steroid hormones in the cholesterol pathway.

38

The physiologic effect of steroid hormones is initiated by their binding to steroid receptors on the surface of target cells. Typically these are trans-membrane proteins, which when coupled to G proteins stimulate or inhibit intracellular signaling through binding to response elements on DNA. De-pending on the type of receptor to which they bind, steroid hormones are classified into five groups: mineralocorticoids, progestines, glucocorticoids, androgens and estrogens. Major fraction of steroids after secretion from the cells binds to carrier plasma proteins. Some of these proteins, like albumin, have a low binding affinity (for testosterone Kd=2.5*10-5 M), while other carrier proteins, such as sex hormone binding globulin (SHBG); have high binding affinity (for testosterone Kd=1*10-9 M). Binding affinity affects the steroids’ half-life, availability for the target tissues and the elimination rate. Free and non-specifically bound steroids are physiologically available, while steroids bound to the binding globulins are physiologically unavailable. In order to assess the physiologically available concentration of steroids it is important to be able to determine not only the total concentration, but also free and nonspecifically bound fractions. Determination of the free fraction of steroids can be accomplished using dialysis or ultrafiltration of the sam-ples (67–69), followed by instrumental analysis; alternatively it could be calculated using methods based on the dissociation constants and the law of mass action (70, 71).

4.2 High sensitivity methods for endogenous steroids

4.2.1 Adrenal steroids The adrenal cortex is the main site for biosynthesis of a major fraction of the steroid hormones, of which especially important are glucocorticoids and mineralocorticoids. The major sites of production of sex hormones are ova-ries in women and testes in men. Normally the adrenal cortex produces rela-tively small amounts of sex steroids, but this pathway becomes active in a group of diseases known as congenital adrenal hyperplasia (CAH). Depend-ing on the specific type of mutation causing the disease, symptoms of CAH could range from life-threatening conditions like salt wasting crisis (condi-tion that result in sudden death if not treated timely) and adrenal insuffi-ciency, to hermaphroditism, precocious puberty, infertility, hypo- and hyper-tension. CAH is caused by a deficiency in one of four enzymes (Figure 10) re-quired for the biosynthesis of glucocorticoids, mineralocorticoids and sex hormones (72–76). Of all forms of CAH the most common type is 21-

39

hydroxylase deficiency, which is the result of mutations in gene CYP21. The other cases of CAH are caused by deficiency of one of the enzymes, 11-hydroxylase, 17-hydroxylase or 3-hydroxysteroid dehydrogenase. Evaluation of concentration of the steroids intermediates of the pathway, [pregnenolone, 17-hydroxyprogesterone (17OHP), 17-hydroxypregnenolone (17OHPregn) and 11-deoxycortisol (11DC)] allows detecting these four enzyme defects. If a person is deficient in one of these enzymes, this causes an accumulation of the precursors of the deficient enzyme and decreased concentration of the products below the blockage. Excess precursors lead to overproduction of other steroids in the adjacent branches of the pathway (Figure 10). Testing for these steroids also allows differentiating late onset CAH and PCOS, conditions that have many of the symptoms in common (77). Because these steroids normally are present in blood at low concentra-tions in a complex sample matrix, the analytical methods for their measure-ment should be sensitive and specific. Prior to development of our LC-MS/MS method (paper V), determination of 11DC, Pregn and 17OHPregn was exclusively based on immunoassays (IA), which were commercially performed only in a few clinical laborato-ries. While developing this method, it was discovered that existing immu-noassays for 17OHPregn and especially for Pregn have poor performance, which is likely related to the cross-reactivity of the antibodies’ used in these immunoassays with other endogenous steroids. Similar observations related to poor negative and positive predictive values of these tests were made ear-lier (78, 79). Difficulties in analyzing 11DC, Pregn and 17OHPregn using LC-MS/MS methods in the past were related to poor ionizability of these molecules and their nonspecific collision induced dissociation (CID). The approach that we used for enhancing the sensitivity of the mass spectrometric detection and improving the fragmentation patterns of these steroids was through incorpo-ration of a functional group in the structure promoting ionization (80, paper V). The reaction of choice for ketosteroids in this method was oximation with hydroxylamine (Figure 11 A). Hydroxylamine reacts with keto-groups of steroids and forms oxime derivatives. The derivatization result in an in-hancement of the electrospray ionization of the keto steroids through im-provement of the solvation of the molecules and introduction in the structure of a tertiary amine moiety. This reaction has quantitative yield and is suit-able for analytical purposes. Enhancement in the sensitivity was especially significant for 17OHPregn and Pregn, likely related the position of double bond in structure of these molecules.

40

H2N OH

HO

O

H

H

H

HO

N

OH

H

H

H

+

A

HO

OH

H

H

H

SN

O

O

Cl

O

OH

SN

O

OH

H

H+

B

Figure 11. Derivatization reactions used in the methods for analysis of ketosteroids (A, papers IV, V); and estrogens (B, paper VI).

Chromatograms of Pregn, 17OHPregn, 17OHP, 11DC extracted from hu-man serum sample are shown in Figure 12 and the method’s performance characteristics are listed in Table 2. The method was compared to radioim-munoassays (RIA) for 11DC, Pregn, 17OHPregn, and the LC-MS/MS method and RIA for 17OHP (Table 3). In order to diagnose abnormalities in biosynthesis of adrenal steroids, ac-curate reference intervals for children and adults are required. Reference intervals were determined (Figure 13, paper V, 81) for healthy adults (n=100) and children (n=919) of different TS. Age and gender-specific changes in concentrations of adrenal steroids were observed in the reference intervals for children of different TS. Con-centrations of Pregn were rising with age in both genders and were higher in girls between TS 2 and 5 compared to boys. Concentrations of 17OHPregn increased in boys and girls with age and were significantly higher in boys. Concentrations of 11DC in girls were declining rapidly with age and were lower compared to boys. Concentrations of 17OHP in females were higher than in males and were dependent on the stage of the menstrual cycle (range of concentrations was 3 fold greater during the luteal stage of the cycle, compared to the follicular stage). In adults, median observed concentrations of Pregn, and 17OHP were comparable between men and women; concen-trations of 17OHPregn and 11DC were up to twice as high in men compared to women.

41

1 2 5 6 7

450

3 4Time, min

900

1 2 5 6 7

450

3 4Time, min

900

1.0 1.6 2.2 2.8 3.4 3.81.0 1.6 2.2 2.8 3.4 3.8Time, min

3000

6000

0.2 0.8 1.4 2.0 2.6 3.2Time, min

0.2 0.8 1.4 2.0 2.6 3.2Time, min

1.0e4

1.9e4

0.2 0.8 1.4 2.0 2.6 3.2Time, min

0.2 0.8 1.4 2.0 2.6 3.2Time, min

5500

1.10e4

1.0 1.6 2.2 2.8 3.4 4.01.0 1.6 2.2 2.8 3.4 4.0

Time, min

700

1500

4.5 5.0 5.5 6.0Time, min

4.5 5.0 5.5 6.0Time, min

4000

8000

2.00 2.25 2.50 2.75 2.95

Time, min

2.00 2.25 2.50 2.75 2.95

Time, min

1200

2400

5.0 5.5 6.0 6.55.0 5.5 6.0 6.5

Time, min

1400

2800

3.5 4.0 4.5 5.0

600

1200

3.5 4.0 4.5 5.0

Time, min

A

J

IH

5.0 5.5 6.0 6.50 7.05.0 5.5 6.0 6.5

Time, min

1600

3400

7.0

G

FED

CB

Inte

nsi ty

, cp s

Inte

nsity

, cps

Inte

nsity

, cps

Inte

nsity

, cps

Figure 12. MRM chromatograms (two characteristic mass transitions per com-pound) of steroids extracted from human serum samples: A - cortisol, B - cortisone, C – 11 deoxycortisol, D - pregnenolone, E - 17 hydroxypregnenolone, F - 17 hy-droxyprogesterone, G - estrone, H - estradiol, I - estriol, J - testosterone (papers III–VI).

Tabl

e 2.

Sum

mar

y of

the

met

hods

’ per

form

ance

cha

ract

eris

tics.

Impr

ecis

ion

Com

poun

d

Eval

uate

d co

ncen

tra-

tions

, ng/

mL

With

in-

run,

%

Betw

een-

run,

%

Tota

l, %

LOD

, ng

/mL

LOQ

, ng

/mL

ULO

L,

ng/m

L Re

cov-

ery,

%

Refe

renc

e

Preg

neno

lone

0.

8, 4

.6, 9

, 18

3.4

5.0

6.1

0.02

5 0.

05

100

96.5

Pa

per

IV

17O

H p

regn

enol

one

0.8,

4.6

, 9, 1

8 4.

4 7.

4 8.

6 0.

1 0.

25

40

99.3

Pa

per

IV

17O

H p

roge

ster

one

0.8,

4.6

, 9, 1

8 3.

8 4.

3 5.

9 0.

025

0.05

10

0 97

.2

Pape

r IV

11 d

eoxy

corti

sol

0.8,

4.6

, 9, 1

8 3.

1 3.

7 4.

9 0.

025

0.05

10

0 91

.0

Pape

r IV

Cor

tisol

6,

30,

50

5.1

8.7

10.0

1

5 80

00

97.5

Pa

per

III

Cor

tison

e 2,

5, 1

0 6.

2 8.

2 10

.4

5 5

1600

96

.4

Pape

r II

I

Test

oste

rone

0.

04, 0

.8, 4

.6, 9

, 18

6.2

7.9

9.9

0.00

5 0.

01

100

98.0

Pa

per

V

Estro

ne

0.00

8, 0

.018

, 0.0

80,

0.21

0 6.

1 6.

9 9.

4 0.

0005

0.

001

60

95.3

Pa

per

VI

Estra

diol

0.

008,

0.0

18, 0

.080

, 0.

210

6.7

6.0

9.2

0.00

05

0.00

1 10

0 94

.9

Pape

r V

I

Estri

ol

0.00

8, 0

.018

, 0.0

80,

0.21

0 5.

1 5.

5 8.

0 0.

0005

0.

001

20

96.5

(8

2)

43

0 2000 4000 6000

Tanner 1 Tanner 2 Tanner 3

Tanner 4/5 Adults

MalesFemales

pg/mL

0 500 1000 1500 2000 2500

Tanner 1 Tanner 2 Tanner 3

Tanner 4/5 Adults

MalesFemales

pg/mL

0 300 600 900 1200 1500

Tanner 1

Tanner 2

Tanner 3

Tanner 4/5

Adults

Males Females

D

B

pg/mL

A

C

0 500 1000 1500 2000 2500

Tanner 1

Tanner 2

Tanner 3

Tanner 4/5

Adults

Follicular phase Luteal phase

MalesFemales

pg/mL

Figure 13. Central 95% of the distribution and median concentrations of adrenal steroids in children (by Tanner stage) and in adults. A - pregnenolone, B - 17 hy-droxyprogesterone, C - 17 hydroxypregnenolone, D - 11 deoxycortisol. Solid lines - median concentrations in males, dashed lines - median concentrations in females.

4.2.2 Glucocorticoids Cortisol plays an important role in human physiology and is commonly ana-lyzed to diagnose the Cushing's disease and the adrenal insufficiency. Ab-normal metabolism of cortisol may also be related to insulin resistance, obe-sity, hypertension, glucose intolerance, type 2 diabetes mellitus and apparent mineralocorticoid excess syndrome (82–86). Cortisol is produced and se-creted by the adrenal gland and its concentration in tissues is controlled by the relative activity of 11HSD type 1 and type 2, which are responsible for inter-conversion between cortisol and cortisone (Figure 10) and maintaining the physiologically available cortisol (86–87). The relative activity of the 11HSD type 1 and 2 enzymes can be evaluated through simultaneous measurement of cortisol and cortisone in blood. The objective of paper III was to develop a simple and rapid assay for simultaneous measurement of cortisol and cortisone in blood, and to estab-lish reference intervals for cortisol, cortisone and their ratio. The developed

44

method uses a soft ionization technique, atmospheric pressure photo-ionization (APPI). In the APPI ion source, UV light initiates a cascade of gas-phase reactions, which transfer charge from easily ionized intermediates to non-polar molecules (88, 89). An advantage of the APPI compared to other types of ion sources is in the more selective ionization that results in improved signal-to-noise ratios. Prior to the development of this method, the majority of published LC-MS/MS methods for analysis of cortisol used atmospheric pressure chemical ionization (APCI). In our method, use of the APPI ion source led to an im-provement of the signal to noise ratio for cortisol and cortisone by a factor of 3 compared to the APCI. The chromatograms of cortisol and cortisone ex-tracted from human serum sample analyzed by this method are shown in Figure 12 and method’s performance characteristics are listed in Table 2. We established reference intervals for cortisone, cortisol and the ratio in serum of adults. Nonparametric reference intervals determined as central 95th percent were 33–246 ng/mL for cortisol, 8–27 ng/mL for cortisone, and the interval for cortisone/cortisol ratio was 0.08–0.30. No statistically signifi-cant difference in the reference intervals was observed between men and women. Cortisone is an inactive metabolite of cortisol and its measurement by it-self currently does not have clinical utility, but the ratio of cortisone/cortisol is representative of the activity of the enzymes 11HSD type 1 and 2. Defi-ciency of the type I enzyme causes inability to convert cortisone to cortisol. This enzyme is also required for the conversion of the synthetic steroid prednisone to active glucocorticoid prednisolone; because of this, people with 11HSD type I deficiency do not respond to the treatment with predni-sone. The deficiency of the type II enzyme is the cause of a congenital form of hypertension (apparent mineralocorticoid excess syndrome), resulting from inability of converting cortisol to cortisone. The ratio below 0.08 is representative of 11HSD type II deficiency and the ratio above 0.30 is rep-resentative of 11HSD type I deficiency. This method (paper III) was used in a study of the relationship between cortisol/cortisone ratio and cardiovascular disease (90). In the study it was found that the cortisone/cortisol ratio moderately correlated (p=0.01) with the outcome of the myocardial infarction in patients with type II diabetes.

4.2.3 Androgens Testosterone (Te) is the major androgen in males (Figure 10), which is in-volved in the development and maintenance of male phenotype. It is also the

predominant bioactive androgen in women, with circulating concentrations 90 to 95% lower than in men. Te is important for many non-gender-specific

45

functions such as growth, muscle development, bone metabolism and protein biosynthesis. Accurate measurement of low concentrations of Te in women and children is important for diagnosing and following up steroid hormone-related diseases. In women, Te is measured as part of the investigation of alopecia, acne, hirsutism, osteoporosis, disorders of libido, detection of an-drogen-secreting tumors, late onset CAH, PCOS and other endocrine and reproductive diseases (91). In children, testosterone is analyzed for gender assignment of infants with ambiguous genitalia, follow-up of children with precocious or delayed puberty and CAH (72–74). In men measurement of low concentrations of Te is needed during the treatment of hypogonadism and monitoring of the androgen suppression therapy during prostate cancer treatment. Until approximately two years ago, immunoassays were the predominant methodology for analyzing testosterone in samples from all groups of popu-lation. Immunoassays for testosterone have acceptable performance for con-centrations characteristic of healthy men, but suffer from a lack of specific-ity at concentrations characteristic of women and children (62–64, 92–94). Taieb et al. (46) evaluated ten commercially available automated immunoas-says for testosterone and reported that none of the assays had adequate specificity for measuring testosterone in serum of women. Because of poor agreement among the results between different immunoassays, the follow-up of patients over time or between laboratories was impossible. The studies (46, 62–64) revealed the need for the development of high sensitivity/ high specificity methods for measurement of testosterone in samples from women and children. As with the adrenal steroids (paper V), we found that oxime derivative (Figure 11) enhanced the sensitivity for detection of Te and allowed measur-ing endogenous concentrations of Te in blood of women and children. A chromatogram of testosterone extracted from serum of a healthy woman is shown in Figure 12 and method’s performance characteristics are listed in Table 2. Considering the analytical sensitivity and small volume of sample used (100 �L), this method is one of the most sensitive of the published LC-MS/MS methods for measurement of Te (95–98). This method compared well with LC-MS/MS and GC-MS methods of other clinical laboratories (Figure 14 and Table 3). Comparison with immu-noassay (Vitros ECi, Ortho-Clinical Diagnostics, Inc.) showed substantial disagreement between the methods at concentrations characteristic of women (Table 3). This was consistent with earlier observations of Fitzger-ald and Herold (63, 64) and Taieb (46) that immunoassays do not perform adequately for samples with low concentrations of Te. To address the issue of poor performance of commercial immunoassays for Te it was suggested to use the isotope-dilution LC-MS/MS method for routine measurement of testosterone in samples from women and children (94). During last two years majority of the testing for testosterone in samples from women and

46

children in commercial clinical laboratories in developed countries was per-formed using LC-MS/MS methods based on this and other principles (paper IV, 92–98).

Table 3. Summary of the methods’ comparison (n - number of samples, r - correla-tion coefficient, Sy/x - standard error).

Deming regres-sion equation Reference

Com-pound

Compara-tive

method n

Slope y-inter-cept

r Sy/x, ng/mL

Cortisol

LC-MS/MS, commercial laboratory

20 0.98 1.6 0.998 1.97 Paper III

Cortisol

RIA, Diag-nostic Products Corpora-tion, Inc.

71 0.74 - 0.70 0.965 32.7 Paper III

11 de-oxycor-tisol

RIA of commercial laboratory

10 0.59 - 0.08 0.985 0.09 Paper V

17 OH pregne-nolone


35 0.93 0.09 0.955 0.76 Paper V

17 OH proges-terone

Coat-a-count, Diagnostic Products Corpora-tion, Inc.

87 1.76 0.42 0.986 0.95 Paper V

17 OH proges-terone


10 1.01 0.009 1.000 0.14 Paper V

Pregne-nolone


58 1.53 - 0.48 0.310 0.54 Paper V

47

Testos-terone (women)

Vitros ECi, Ortho-Clinical Diagnostics

150 0.47 0.093 0.637 0.078 Paper IV

Testos-terone (men and women)

Vitros ECi, Ortho-Clinical Diagnostics

216 1.15 0.11 0.976 0.05 Paper IV

Testos-terone

LC-MS/MS and GC-MS, com-mercial laboratories

30 1.01 0.0009 0.953 0.029 Paper IV

Estrone


20 0.98 0.0045 0.959 0.006 Paper VI

Estrone RIA, com-mercial laboratory

20 0.66 - 0.018 0.961 0.006 Paper VI

Estrone

Diagnostic Systems Laborato-ries

67 0.97 - 0.018 0.59 0.011 Paper VI

Estradiol


20 1.06 0.002 0.995 0.004 Paper VI

Estradiol RIA, com-mercial laboratory

20 0.79 - 0.014 0.985 0.007 Paper VI

Estradiol

Vitros ECi; Ortho-Clinical Diagnos-tics, Inc.

41 0.65 - 0.0113 0.743 0.018 Paper VI

48

200

400

600

800

200 400 600 800

LC-MS/MS

GC-MS

Comparative methods, pg/mL

Eval

uate

d m

eth o

d, p

g/m

L

Figure 14. Comparison of LC-MSMS method for testosterone (paper IV) with LC-MS/MS and GC-MS (62) methods of clinical diagnostic laboratories.

In order to enable using this method in clinical diagnostic practice we es-tablished reference intervals (paper IV, Figure 15) for total Te (free and bound to carrier proteins), bio-available Te (free and bound to albumin) and free Te in serum of women, men and children (n=1063).

This method (paper IV) was used in epidemiological studies of Te re-placement therapy in HIV infected women (99) and healthy post-menopausal women (100). It has been reported (101–103) that raising Te concentrations into the upper end of the normal (for female) range was asso-ciated with improvements in the muscle strength and well being of women. The objective of the first study (99) was to assess whether testosterone re-placement therapy increases weight and muscle strength in HIV-infected women with androgen deficiency. At the doses of testosterone used in the study, testosterone concentration in blood of the HIV infected women was raised to upper normal (for females) levels, but this did not significantly increase the patients’ body weight or muscle strength. The objective of the second study (100) was to determine the time course profile of serum testos-terone concentrations in postmenopausal women during treatment with dif-ferent doses of testosterone gel and to assess whether Te treatment affects endogenous concentrations of estradiol (E2). The Te supplementation raised the concentrations of Te in blood to the targeted concentrations but did not affect biosynthesis of E2 (although Te is precursor of E2).

49

BA

C D

0 50 100 150

Tanner 1Tanner 2Tanner 3

Tanner 4/5Adult men

Before menarcheAfter menarche

PM womenEarly follicular phaseLate follicular phase

Luteal phase

MalesFemales

pg/mL

0 250 500 750

Tanner 1

Tanner 2

Tanner 3

Tanner 4

Tanner 5

Women age 18-51

PM women

pg/mL

0 50 100 150 200 250 300

Tanner 1Tanner 2Tanner 3

Tanner 4/5Adult men

Before menarcheAfter menarche

PM womenEarly follicular phaseLate follicular phase

Luteal phase

MalesFemales

pg/mL

0 2500 5000 7500 10000

Tanner 1

Tanner 2

Tanner 3

Tanner 4

Tanner 5

Men age 18-52

pg/mL

Figure 15. Central 95% of the distribution and median concentrations of sex ster-oids in children (by Tanner stage) and in adults. A - estrone, B - estradiol, C - testos-terone in females, D - testosterone in males. Solid lines - median concentrations in males, dashed lines - median concentrations in females. PM women - postmenopausal women.

4.2.4 Estrogens Estrogens have their highest biologic activity in the 17-hydroxy configura-tion and reductive function of 17-hydroxysteroid dehydrogenase (17HSD) is essential for their biosynthesis (Figure 10). Eleven 17HSD iso-enzymes were identified, which participate in inter-conversion between steroids in various tissues. The 17HSDs differ in tissue distribution, specificity, sub-cellular localization, and mechanism of regulation (66). Inter-conversion of estrogens in different tissues is responsible for a tissue-specific action of estrogens and contributes to total concentration of estrogens present in circu-lating blood.

50

In females estrogens are responsible for the development and maintenance of secondary gender characteristics and reproductive function. Low concen-trations of estrogens in women are associated with disturbed puberty, oligo-amenorrhea, estrogen deficiency and menopause (72, 91, 104–106). Recent studies suggest that low concentrations of estrogens in both genders corre-late with osteoporosis, cardiovascular, cognitive and neurological diseases (107–114). An accurate measurement of low concentrations of estrogens could assist in the diagnoses of the above conditions. Challenges in measurement of estrogens in blood of postmenopausal women, men and children are related to their low physiologic concentrations and presence in the samples of interfering substances. Analysis of estrogens in biological samples is commonly performed using immunoassays (115–117). Recently it was shown that commercial immunoassays for E2 are inac-curate when utilized for measuring low endogenous concentrations (117). Earlier published LC-MS/MS methods for estrogens (118, 119) also lack specificity needed for measurement of low concentrations. In paper VI our goal was to develop a high sensitivity/ high specificity method suitable for measurement of endogenous concentrations of estrogens in postmenopausal women, men and children. To enhance sensitivity of detection, estrogens were derivatized using dan-syl chloride (118), an amine-containing sulfonyl halide (Figure 11B). Ad-vantages of analyzing estrogens as dansyl derivatives include significant gain in sensitivity (through introduction of tertiary amine in the structure), mild reaction conditions and quantitative yield of the products. The yield of the dansylation reaction (Figure 11B) was found to be highly dependent on the conditions under which the reaction was performed. Optimization of the reaction using experimental design methods (120) demonstrated that concen-tration of dansyl chloride; pH of solution, incubation temperature and time had a strong impact on the reaction recovery (Figure 16).

51

-1.25

-0.75

-0.25

0.25

0.75

1.25

Conc

DancCl

Conc

DancCl

* Conc

DancCl

Incub

time,

min

Incub

Temp, C

Conc

DancCl

* Incub

Temp

Incub

time *

Incub

time

Conc

DancCl

* Incub

time

Incub

Temp *

Incub

Temp

Incub

Temp *

Incub

time

Estrone

Estradiol

EstriolRatio

E/E

max

Figure 16. Paretto diagram (120) with relative effect of the reaction conditions on the recovery of dansyl derivatives of estrogens.

As sensitivity improved through the derivatization, specificity of the analysis was inadequate because large number of potentially interfering peaks was present in the chromatograms. An approach that we found suit-able for removing the interfering peaks and reducing the background noise was through the use of a two-dimensional (2D) chromatographic separation. The best selectivity for the 2D separation was observed using a C1 column for the 1st dimension separation and a phenyl column (retention based on -�interactions) as the 2nd dimension. The derivatization and the 2D separation when applied simultaneously in this method produced synergetic improvement and enabled enhancing the sensitivity and the specificity of the method (paper VI). The above ap-proach resulted in a method, which is more sensitive and specific compared to earlier published LC-MS/MS methods (118, 119). Chromatograms of estrone (E1), E2 and estriol (E3) extracted from human serum sample are shown on Figure 12 and method’s performance characteristics are listed in Table 2. Method comparisons for estrone and estradiol showed good agreement with LC-MS/MS assays (Table 3) performed at another clinical laboratory (Table 4.2) while the sensitivity of our method was better. Com-parison with immunoassays for E1 and E2 showed discrepancy between the methods (Table 3); which was likely related to the cross-reactivity of the antibodies used in the immunoassays. Because sensitive and specific methods for measurement of estrogens were not available in the past, limited information was published on pediatric reference intervals of estrogens (121). Our method was used for establishing the reference intervals of estrogens in children (n=760) and adults (n=259, paper VI). A plot of median concentrations and central 95% reference intervals for E1 and E2 by TS in females and males are shown in Figure 15.

52

Concentrations of estrogens (median values) in girls reached adult levels by TS 3 and in boys at TS 4–5. Median value of the E2/E1 ratio (indicator of reproductive health in women) was 1.49 (1.30, 1.56, 1.48 during early fol-licular, late follicular and luteal stages of the menstrual cycle, respectively). The median E2/E1 ratio in girls was 1.85 (ages 12 to 17 years) and in boys it was 1.08 (age 12 to 17 years). In postmenopausal women the median ratio of E2/E1 was 0.55; in 18 to 60 year old men the median ratio was 0.98. Based on the results of the reference interval studies in boys and girls, peak concentrations (median values) of testosterone (paper IV) and estra-diol (paper VI) occur in girls at the TS 3 and in boys at the TS 4–5 (Figure 15). This suggests that potentially there is a link between the sex hormone concentrations and the behavioral and social changes occurring in teenagers of corresponding ages. The methods for estrogens (paper VI) and testosterone (paper IV) were used in an epidemiological study (124) of the relationship between sex ster-oids and bone health in elderly men. In the study we found that in men over the age of 60 low concentrations of Te in serum were associated with an increased risk of osteoporotic fractures. These results are in agreement with observation of Mellström et al. (125) but contrary to earlier studies (111, 126), where concentration of E2 were linked to the osteoporotic fractures. Difference in the findings is likely related to greater specificity of the LC-MS/MS methods used in this study, compared to the immunoassays, used in all earlier studies. These observations suggest that measurement of testos-terone in serum might provide useful clinical information for the assessment of the fracture risk in elderly men. Considering the low concentrations of testosterone and estrogens in elderly adults, high sensitivity/ high specificity methods are needed in the diagnostics and for the clinical studies involving measurements of sex steroids in these groups of population. The need in standardization of measurements of testosterone and estradiol has been emphasized in a recently published statement of The Endocrine Society (94). In response to this, the Center for Disease Control and Preven-tion (USA) in 2008 started an initiative to standardize the Te and E2 tests among clinical laboratories (122, 123). The LC-MS/MS tests developed in this thesis (papers IV, VI) are among the most sensitive and specific diag-nostic tests for testosterone and estradiol routinely used in clinical laborato-ries (127). Methods for analyzing endogenous steroids that were developed in this thesis have high sensitivity and specificity. In part this is related to the in-herent specificity of tandem mass spectrometry, but it also enhanced through is exploitation of rather unique chemical properties of the measured analytes. The APPI ionization was used as an approach for the enhancement of the sensitivity for cortisol and cortisone (paper III). Methods for keto steroids (papers IV, V) and estrogens (paper VI) utilized chemical derivatization with tertiary amine-containing functional groups as a way of enhancing the

53

sensitivity. Various modes of extraction were used for the enhancement of the specificity. In all methods methyl t-butyl ester (MTBE) was used as an extraction solvent (papers III–VI) exploiting good solubility of steroids and poor solubility of phospholipids and other impurities in this solvent. Sample preparation in the method for adrenal steroids (paper V) used SPE on poly-meric adsorbent that was critical for performance of the method; and multi-dimensional separation was the key to achieve high sensitivity and specific-ity for measurement of estrogens (paper VI). Despite the high specificity of MS/MS detection, this technique is not im-mune from interferences. In papers II–VIII two mass transitions were used for the steroids and the internal standards, and ratios of the intensities of the mass transitions were evaluated as a means of the assessment of the specific-ity.

4.3 Ovarian steroidogenesis

4.3.1 Steroids in ovarian follicles of healthy women In women of fertile age ovarian follicles are the site of biosynthesis for a major fraction of the estrogens and androgens present in circulation. Fol-licular steroids are secreted by granulosa and theca cells under control of gonadotropins, and the hormonal microenvironment affects development of the follicles and the oocytes viability (128). In normal ovulatory cycles, higher concentration of estradiol in follicular fluid (FF) is associated with healthy follicles containing oocytes capable of meiosis, and higher concen-trations of androgens are indicative of atretic changes (degeneration and subsequent resorption) (128–132). Majority of earlier studies of steroids in ovarian follicles (133, 134) were undertaken to obtain prognostic parameters for the likelihood of a successful implantation during in-vitro fertilization (IVF), but the relationship between steroid hormones and follicular devel-opment in regularly menstruating (RM) healthy women was not well stud-ied. In part this is related to the very small sample of FF that may be ob-tained from follicles of healthy women during follicular stage of menstrual cycle and absence of sensitive and specific methods for simultaneous quanti-tation of multiple steroids in small samples in the past. The aims in paper VII were to determine concentrations of endogenous steroids in FF using isotope dilution LC-MS/MS and to describe patterns of distribution of the steroids in FF during early follicular phase of the men-strual cycle and after ovarian stimulation for IVF treatment. We also com-

54

pared concentrations of steroids observed in this study with previously pub-lished concentrations of steroids obtained by immunoassays. Difference of this study from earlier publications (135–139) was in the use of more sensitive and specific methods that allowed simultaneous measure-ment of large number of bioactive steroids and intermediates of their biosyn-thesis. Majority of earlier studies of ovarian steroidogenesis utilized immu-noassays, and each study focused on the effect of one or few of the steroids in the follicular development. Analysis of a large number of steroids in FF during early stages of follicular development was not possible in the past because immunoassay-based methods would require at least a few milliliters of FF for such measurements, sample size unrealistic for the follicles during early developmental stages. One of the pitfalls associated with use of im-munoassays for analyzing FF samples is related to the significantly higher concentrations of sex steroids in FF compared to serum. The difference in the concentrations may cause cross-reactivity that is not observed in serum samples (for which immunoassays are validated). The above problem is not relevant to mass spectrometry based methods (papers III–VI, 140–142). The use of highly sensitive methods developed in papers III–VI and simul-taneous measurement of multiple steroids allowed analyzing 40 �L of FF sample for fifteen steroids and quantitating thirteen steroids. FF samples from 21 RM women and five women on IVF treatment were used in this study. Concentrations of androgens, 17OHP, and estrogens in the ovarian FF of RM women were 200 to 1000 fold greater than in serum (papers III–VI). In FF of RM women androgens were the most abundant class of steroids followed by progestins, pregnenolones, estrogens and glu-cocorticoids. Compared to RM women, women undergoing ovarian stimu-lation had significantly higher concentrations of E2, Pregn, 17OHP and cor-tisol; significantly higher ratios of cortisol/cortisone, E2/E1 and E2/Te; and significantly lower concentrations of 17OHPregn, 11DC, cortisone, DHEA, androstenedione (A4) and Te. The pie diagram with distribution of concen-trations of the steroids in FF of RM women is shown in Figure 17 and paper VII contains central 90% distribution of the concentrations. In RM women, A4 was the dominating steroid (46.8 %) followed by 17OHP, DHEA and Pregn; in women undergoing ovarian stimulation 17OHP was the dominat-ing steroid (54.0 %) followed by E2 and Pregn. Figure 18 shows median concentrations of the steroids in FF samples from RM women, and women after ovarian stimulation for IVF treatment. Bioactive sex steroids in FF are E2 and Te; depending on their relative con-tent, follicles can be classified (131) as estrogen dominant (EDF, E2/Te>4) or androgen dominant (ADF, E2/Te<4). In EDF estrogens and androgens constituted 27.7% and 41.1% of total concentration of the steroids; in the ADF estrogens and androgens constituted 4.4% and 68.3% respectively. The major estrogen in ADF was estrone and the major estrogen in EDF was estradiol. Compared to ADF, EDF had significantly higher concentration of

55

E2, significantly higher E2/E1 ratio and significantly lower concentrations of A4 and Te. In ADF, A4 was the dominating steroid (56.4 %) followed by 17OHP and DHEA; in EDF A4 was also the dominating steroid (30.8 %) followed by 17OHP and E2.

Estrone3.8%

Estradiol3.4%

Estriol0.1%

Androstanedione0.2%

Pregnenolone5.7%

17 OH Pregnenolone

3.5%

17 OH Progesterone

19.3%

11deoxycortisol0.5%

Cortisol1.9%

Cortisone3.5%

DHEA9.5%

Testosterone2.0%

Androstenedione46.8%

A

17 OH Progesterone

15.0%

17 OH Pregnenolone

4.7%

Pregnenolone4.1%

DHEA11.2%

Cortisone2.9%

Cortisol1.7%

11deoxycortisol0.4%

Estradiol0.8%

Androstanedione0.3%

Estrone0.8%

Estriol0.0%

Testosterone1.9%

Androstenedione56.1%

B

Figure 17. Pie diagrams of the distribution of median concentrations of steroids in FF of RM (A) and women diagnosed with PCOS (B).

Paper VII provides the first mass spectrometry-based data on concentra-tions and patterns of distribution of thirteen endogenous steroids in ovarian FF during early follicular phase of the menstrual cycle. These data will be useful for better understanding of normal ovarian physiology and the effects of ovarian stimulation on steoridogenesis in FF.

56

300

600

900

Pregne

nolon

e

17 O

Hpre

gnen

olone

17 O

H prog

estero

ne

11de

oxyco

rtisol

Cortiso

l

Cortiso

neDHEA

Andros

tened

ione

Testost

erone

Andros

taned

ione

Estron

e

Estrad

iol

Estriol

PCOSControlsIVF

ng/m

L

Figure 18. Median concentrations of endogenous steroids in samples from RM women (control group), women diagnosed with PCOS and women after ovarian stimulation for IVF treatment.

4.3.2 Polycystic ovary syndrome (PCOS) and steroids in follicular fluid of PCOS patients PCOS is one of the most common endocrine disorders that affect 4–7% of reproductive age women (143–147). In PCOS patients, chronic absence of ovulations result in accumulation in the ovaries of large number of atretic follicles, which produce excess of androgens. In addition to reproductive abnormalities and hyperandrogenism, symptoms characteristic of PCOS may include obesity, hyperinsulinemia, type II diabetes, dyslipidemia, higher incidence of cardiovascular disease, endometrial and breast cancers (146–149). The etiology and the mechanisms underlying the PCOS are not fully understood, but it is known that disbalance of insulin (144–146), abnormali-ties in the enzymes involved in biotransformation between steroid hormones (150–153) and genetic predisposition (154, 155) play role in PCOS. Accumulation in the ovaries of large number of follicles, and androgens excess are characteristic, but not specific markers of PCOS (145–147). Be-cause of this, PCOS is considered a diagnosis of exclusion, meaning that the diagnosis requires exclusion of other possible diseases causing similar symp-toms. It is common practice to base diagnosis of PCOS on patients’ history, physical examination and semi-specific laboratory tests (e.g. LH/FSH ratio, free and total androgens). The testing is usually performed for excluding other diseases, which cause symptoms similar to PCOS (156–159). The aims in paper VIII were to investigate steroid profiles in ovarian FF of women with PCOS, to compare steroid concentrations with concentrations observed in FF of RM women and to attempt to identify specific biomarkers of PCOS.

57

Follicular fluid samples from 27 women with PCOS and 21 women with-out PCOS (control group) were used in this study. The diagnosis of PCOS was based on amenorrhea or oligomenorrhea, ultrasound examination of the ovaries and elevated concentration of androgens in blood. The FF samples were analyzed using methods described in papers III–VI. Figure 17 shows pie diagram of the distribution of median concentrations of the steroids in FF samples of the PCOS patients and the control group and Figure 18 shows median concentrations of the steroids in the groups. An-drogens were the dominant class of steroids in FF of the PCOS patients; progestines, pregnenolones, and glucocorticoids were present at lower con-centrations, and estrogens were the least abundant class. Median concentra-tions of all measured androgens in the PCOS group were 30–50% greater and median concentrations of the estrogens were 40–70% lower than in the control group (Figure 18). Significant difference between the PCOS and the control groups were also in the ratios of the concentrations of 17OHPregn/Pregn (p<0.0001) and total estrogens /total androgens (p=0.028). The identified differences (if confirmed in further studies) may potentially serve as biomarkers of PCOS in FF samples. Utility of biomarkers could be evaluated using Receiver Operating Charac-teristic (ROC) curves (160–162). The ROC curves are based on the clinical sensitivity and the clinical specificity of biomarkers. The ROC is a plot of sensitivity versus (1–specificity). Clinical sensitivity is the probability that a marker correctly predicts the presence of a condition and clinical specificity is the probability that the measured value correctly predicts that the condi-tion does not exist in healthy individuals. The area under the ROC curve is an index that represents probability that a test would correctly distinguish between the affected and non-affected individuals. A biomarker with no predictive ability produces a curve that follows the diagonal of the grid (AUC=0.5). For a biomarker with perfect sensitivity and specificity the ROC curve would pass through the point (0,1) on the graph and the area under the curve (AUC) would be 1.0. The closer the ROC curve comes to this ideal point, the better is the discriminating ability of a biomarker. Figure 19 shows ROC curves for potential biomarkers identified in the study (only markers with AUC>0.75 are shown). The greatest sensitivity and specificity out of the identified potential biomarkers had the ratio 17OHPregn/Pregn followed by concentrations of DHEA, 17OHPregn, an-drostanedione, ratio total estrogens/ total androgens and concentration of estrone. Predictive ability of the biomarkers would improve if they are used in combination. However, further studies will be needed to confirm the clinical utility of the identified markers.

58

17OHPregnAUC=0.850

DHEAAUC=0.852

AndrostanedioneAUC=0.828

EstroneAUC=0.771

Total ESTR/total ANDRAUC=0.803

17OHPregn/PregnAUC=0.947

1-Specificity

0.2 0.4 0.6 0.8 1.0 0.2 0.4 0.6 0.8 1.0 0.2 0.4 0.6 0.8 1.0

1.0

0.8

0.6

0.4

0.2

1.0

0.8

0.6

0.4

0.2

Sens

itivi

ty

Figure 19. ROC curves for identified potential biomarkers of PCOS in FF samples.

59

5. Concluding remarks

This thesis describes the development, evaluation and use of novel tandem mass spectrometry based tests for clinical diagnostics. Effectiveness of di-agnostic testing depends on the appropriateness of the utilized biomarkers, reliability of the methods and availability of reference values for the bio-markers. All methods developed in this thesis were extensively validated to assure their accuracy and the diagnostic utility, and were used for establish-ing reference intervals for the biomarkers in serum samples of adults and children. In diagnostic testing, high complexity of biological samples sets rigorous requirements for the specificity of measurements. The issue of specificity of analysis was addressed in paper II. To assure high quality of the results in methods using tandem mass spectrometry, we proposed various strategies for assessment of the specificity of analysis. We also developed an approach for simultaneous quantitation of identically fragmenting iso-mers from unresolved chromatographic peaks (paper I). The method al-lowed increasing speed of analysis of isomers and was adopted for high throughput LC-MS/MS analysis of methylmalonic acid, a marker of vitamin B12 deficiency. Methods for analyzing endogenous steroids developed in this thesis (pa-pers III–VI) are more sensitive and specific compared to earlier published methods. Greater sensitivity and specificity of these methods opened new possibilities for their use in early diagnostics and monitoring of such diverse group of endocrine diseases and conditions as disorders of puberty, repro-ductive diseases, polycystic ovary syndrome, congenital adrenal hyperplasia, adrenal insufficiency, Cushing's disease, apparent mineralocorticoid excess syndrome, alopecia, hirsutism, menopausal status, endocrine cancers, pre-disposition to osteoporosis, cardiovascular and neurological disease. All methods developed in this thesis use small sample aliquot (100–200 �L) and help reducing volume of blood needed from patients for the testing. The methods (papers I, III–VI) were used in epidemiological and clinical stud-ies (48, 90, 99, 100, 124, 163, 164) and routinely used for diagnostic testing in a large clinical laboratory. The methods developed in papers III–VI were used in a study of the ster-oidogenesis in ovarian follicles of RM women, women on IVF treatment and women diagnosed with polycystic ovary syndrome (papers VII, VIII). The data on the steroid concentrations and association between the metabolites within the pathway could be useful for better understanding of the ovarian

60

physiology, and may potentially lead to validating new biomarkers and new diagnostic testing. Great progress has already been made in clinical mass spectrometry and many new developments will come in the future. During the last fifteen years mass spectrometry based tests revolutionized approaches for diagnos-ing congenital metabolic disorders in newborns (14); current and future de-velopments in mass spectrometry will likely have major impact on the diag-nostic practices in endocrinology and other fields of medicine.

61

6. Acknowledgements

Many people have been involved in my projects and without their help I would not be able to accomplish the work. This thesis, my knowledge in mass spectrometry and my future professional development are a result of the guidance that I received from my scientific advisors Dr Jonas Bergquist and Dr Alan Rockwood. They were always encouraging, supportive and ready to help. Words cannot express my gratitude. Many thanks to Dr Francis Urry for encouragement, support in research and for introducing me to the fields of mass spectrometry and toxicology. I appreciate the help of Dr Wayne Meikle and Dr William Roberts, for providing guidance, advices and sharing knowledge in clinical chemistry and endocrinology. Many thanks to Dr Tord Naessen for introducing me to the field of gynecological and reproductive endocrinology, help, advice and scientific guidance during our collaborative research. I am very grateful to ARUP® Institute for Clinical and Experimental Pa-thology for financial support of the projects that became basis of this thesis and appreciate the support of Mark Astill and Dr Harry Hill for encouraging my continued education. Special thanks to all my friends at the department of Analytical Chemistry of Uppsala University, ARUP Institute and ARUP Laboratories. I appreci-ate all my coauthors for productive collaborations and hope to work with you in future. I appreciate all the organizational help of Barbro Nelson dur-ing the time of my graduate studies. Very warm and special thanks to my parents and all my family. Finally, special thanks to Mila, without your support and encouragement this would never be possible.

62

7. Swedish summary

Kliniska laboratorier har blivit en nödvändig del av den medicinska veten-skapen med de huvudsakliga målsättningarna att diagnostisera sjukdomar, följa en behandling och kunna ge patienterna en prognos. Sedan masspekt-rometri introducerats i klinisk diagnostik har dessa instrumentella metoder visat sig vara några av de mest noggranna analytiska tekniker som idag finns tillgängliga i kliniska laboratorier. De analytiska metoder som beskrivs i denna avhandling använder tan-demmasspektrometri (MS/MS). MS/MS består av tre komponenter; två masspektrometrar med en kollisionscell emellan (Figur 20).

Q0 Q1 Q2 Q3

L

D

Figur 20. Schematiskt bild av en triplequadrupol (tandem) masspektrometer: Q0 - quadrupol,med enbart radiofrekvent fält Q1 - första massanalysator, Q2 - kollisions-cell, Q3 - andra massanalysator, L - extraktionslins, D - detektor.

Den första masspektrometern väljer en jon med en specifik massa per laddning (moderjon), som sedan sänds in i kollisionscellen där jonen frag-menteras genom kollisioner med neutrala gasmolekyler. De joner som gene-reras (fragmentjoner) sänds vidare in i den andra masspektrometern där de-ras individuella massor detekteras. Moder- och fragmentjonernas massor representerar molekylernas kemiska struktur och specifika massa. Den hög-gradiga analytiska specificiteten hos vätskekromatografi (LC) och tandem-masspektrometri baseras således på dessa fundamentala egenskaper hos ana-lyterna. Kombinationen av kromatografi och masspektrometri åstadkommer en synergieffekt genom att svagheterna hos respektive metod övervinns när det gäller kvalitén på erhållna data. Kromatografisk separation särskiljer individuella komponenter i komplexa prover, medan masspektrometri otve-tydigt identifierar dessa komponenter. En fördel med LC-MS/MS som ana-

63

lytisk mätteknik är förmågan att utvärdera specificiteten i analysen i varje enskilt patientprov. Denna förmåga är unik för masspektrometri jämfört med andra analytiska metoder och försäkrar en hög kvalitet på de diagnostiska testerna. I artikel II föreslog vi ett flertal sätt att utvärdera specificiteten i metoderna som använder MS/MS-detektion. Denna avhandlings huvudsakliga fokus ligger på användningen av tan-demmasspektrometri i diagnostik av endokrina sjukdomar. Människans normala tillväxt och utveckling är beroende av den komplexa integrerade funktionen av hormoner. Steroidhormoner reglerar ett stort antal fysiologis-ka funktioner och många sjukdomar är relaterade till felreglering i biosynte-sen av steroidhormoner. De metoder för att analysera steroider som inklude-rats i denna avhandling (glukokortikoider (artikel III), testosteron (artikel IV) binjuresteroider (artikel V) och östrogener (artikel VI)) är känsligare och mer specifika jämfört med tidigare publicerade LC-MS/MS-metoder. För att det diagnostiska testet ska vara meningsfullt och tillåta klinisk tolkning av resultaten måste mätresultaten från individuella patientprover relateras till lämpliga referensvärden. Genom att använda de presenterade metoderna har vi etablerat referensvärden för tio steroidhormoner i serum från barn och vuxna. Alla metoder som utvecklats i denna avhandling använder ytterst små provvolymer och är specifikt designade för att reducera mängden prov som måste tas från patienten. De beskrivna metoderna hjälper till med diagnos och behandling av patien-ter med så skilda sjukdomar som kongenital binjurehyperplasi, binjureinsuf-ficiens, infertilitet, för tidig pubertet, polycystiskt ovariesyndrom (PCOS), hypo- och hypertension, Cushings syndrom, insulinresistens, fetma, typ 2-diabetes, mineralocorticoid insufficiens, alopeci, hirsutism, endokrin cancer, osteoporos, hjärt-kärlsjukdomar och en rad neurologiska sjukdomar. Ett av målen med den moderna medicinska vetenskapen är att identifiera sjukdomar i dess tidiga och behandlingsbara stadier med hjälp av sk scree-ning-tekniker. För att systematiskt undersöka ett stort antal patienter måste proverna kunna analyseras snabbt och effektivt. Detta är ett svårt problem för masspektrometriska metoder, i synnerhet när det gäller biomarkörer med endogena isomerer. I artikel I har vi utvecklat en teknik för att samtidigt kvantifiera flera isomerer från icke upplösta kromatografitoppar. Denna metod möjliggör ökad hastighet i analysen av dessa isomerer. Metoden har anpassats för LC-MS/MS-analys med hög genomströmningstakt av metyl-malonsyra, en markör för vitamin B12-brist. Artikel VII och VIII beskriver en jämförande studie av steroidogenesen i äggfolliklar hos kvinnor med regelbunden ägglossning och kvinnor med PCOS. I artikel VII bestämdes de masspektrometriska referensvärdena för koncentrationen av tretton steroidhormoner i follikelvätska. De erhållna resultaten av steroidkoncentrationerna och förhållanden mellan steroidmeta-boliter inom syntesvägarna kan användas för en ökad förståelse av äggstock-arnas fysiologi och PCOS. I artikel VIII identifierade vi potentiella biomar-

64

körer för PCOS i form av markörer för dysreglering i steroidogenesen och ytterligare studier pågår för att bekräfta den kliniska användbarheten av des-sa fynd. Man kan konstatera att masspektrometri har stort inflytande inom klinisk diagnostisk testning. Under de senaste tio åren har masspektrometriska tester revolutionerat diagnosen av bland annat medfödda metabola sjukdomar hos nyfödda. Nuvarande och framtida utveckling inom masspektrometri kommer sannolikt att ha stort inflytande på diagnostisk praxis inom många medicins-ka områden.

65

8. References

1. Berzelius JJ. Föreläsningar i djurkemien, 2 vols. Stockholm: Delen,

1806–1808. 2. Berzelius JJ. General views of the composition of animal fluids.

Med Chir Trans 1812;3:198–276. 3. Laboratory Sciences Career Overview http://www.mayo.edu/

mshs/lab-career.html (accessed February 5, 2008). 4. Sahab ZJ, Semaan SM, Sang QX. Methodology and applications of

disease biomarker identification in human serum. Biomarker In-sights 2007;2:21–43.

5. Biomarkers definitions working group: biomarkers and surrogate endpoints: preferred definitions and conceptual framework. Clin Pharmacol Ther 2001;69:89–95.

6. Witte DL, VanNess SA, Angstadt DS, Pennell BJ. Errors, mistakes, blunders, outliers, or unacceptable results: how many? Clin Chem 1997;43:1352–1356.

7. Plebani M, Carraro P. Mistakes in a stat laboratory: types and fre-quency. Clin Chem 1997;43:1348–1351.

8. Plebani M. Errors in clinical laboratories or errors in laboratory medicine? Clin Chem Lab Med 2006;44:750–759.

9. Kushnir MM, Liu A, Crockett DK, Urry FM, Roberts WL. A GC-MS method for organic acids screening in urine. Paper 1906. Pre-sented at PITTCON New Orleans, 2001.

10. Roberts WL, McMillin GA, Burtis CA. Reference information for clinical laboratory. In: Tietz Textbook of Clinical Chemistry and Molecular Diagnostics. Ashwood ER, Burtis CA, Bruns DE eds. 4th edition, New York: Saunders; 2005:2251–2318.

11. Rhoads RA, Friedberg F. Affinity chromatography of enzymes. In: Kline T. Handbook of Affinity Chromatography. New York: Marcel Dekker, Inc.1993:79–111.

12. Barcelo D, Hennion MC. On-line sample strategies In: Trace deter-mination of pesticides and their degradation products in water, El-sevier, Amsterdam, NL, 1997:357–422.

13. Lacorte S, Vreuls JJ, Salau JS, Ventura F, Barcelo D. Monitoring of pesticides in river water using fully automated on-line SPE and LC with DAD with a novel filtration device. J Chromatogr A

66

1998;795:71–82. 14. Chace DH, Millington DS, Terada N, Kahler SG, Roe CR, Hofman

LF, Rapid diagnosis of phenylketonuria by quantitative analysis for phenylalanine and tyrosine in neonatal blood spots by tandem mass spectrometry. Clin Chem 1993;39:66–71.

15. Johnson DW. Dimethylaminoethyl esters for trace, rapid analysis of fatty acids by electrospray tandem mass spectrometry. Rapid Com-mun Mass Spectrom 1999;13:2388–2393.

16. Johnson DW. Analysis of amino acids as formamidene butyl esters by electrospray ionization tandem mass spectrometry. Rapid Com-mun Mass Spectrom 2001;15:2198–2205.

17. Johnson DW. A modified Girard derivatizing reagent for universal profiling and trace analysis of aldehydes and ketones by electros-pray ionization tandem mass spectrometry. Rapid Commun Mass Spectrom 2007;21:2926–2932.

18. Gao S, Zhang ZP, Karnes HT. Sensitivity enhancement in liquid chromatography/atmospheric pressure ionization mass spectrometry using derivatization and mobile phase additives. J Chromatogr B 2005;825:98–110.

19. Kushnir MM, Komaromy-Hiller G, Shushan B, Urry FM, Roberts WL. Analysis of dicarboxylic acids by tandem mass spectrometry. high throughput quantitative measurement of methylmalonic acid in serum, plasma and urine. Clin Chem 2001, 47:1993–2002.

20. Kushnir MM, Crossett J, Brown PI, Urry FM. Analysis of gabapen-tin in serum and plasma by solid phase extraction and GC-MS for therapeutic drug monitoring. J Anal Tox 1999;23:1–6.

21. Kushnir MM, Crockett DK, Nelson G, Urry FM. Comparison of four derivatizing reagents for 6-acetylmorphine GC-MS analysis. J Anal Tox 1999;23:262–269.

22. Paul W, Steinwedel H. A new mass spectrometer without a mag-netic field. Z. Naturforsch 1953;8:448–450.

23. Mathieu EM. Le Mouvement Vibratoire d’une Membrane de forme Elliptique. J Math Pure Appl 1868;137–203.

24. Busch KL, Glish GL, McLuckey SA. Mass spectrometry/mass spec-trometry: techniques and applications of tandem mass spectrometry, VCH Publishers, Inc., New York, 1988.

25. Dawson PH, Dawson MC. Quadrupole mass spectrometry and its applications. Amer Inst of Physics, 1995.

26. Bruins AP, Covey TR, Henion JD. Ion spray interface for combined liquid chromatography/atmospheric pressure ionization mass spec-trometry. Anal Chem 1987;59:2642–2646.

27. McLuckey SA. Principles of collisional activation in analytical mass spectrometry. J Am Soc Mass Spectrom 1992;3:599–614.

28. Cooks RG. Collision-induced dissociation: readings and commen-tary. J Mass Spectrom 1995;30:1215–1221.

67

29. Hunt DF, Giordani AB, Rhodes G, Herold DA. Mixture analysis by triple-quadrupole mass spectrometry: metabolic profiling of urinary carboxylic acids. Clin Chem 1982;28:2387–2392.

30. Xu X, Veals J, Korfmacher WA. Comparison of conventional and enhanced mass resolution triple-quadrupole mass spectrometers for discovery bioanalytical applications. Rapid Commun Mass Spec-trom 2003;77:832–837.

31. Giddings JC. Concepts and comparisons in multidimensional sepa-ration. J High Resol Chromatogr Commun 1987;10:319–323.

32. Kanu AB, Dwivedi P, Tam M, Matz L, Hill HH Jr. Ion mobility-mass spectrometry. J Mass Spectrom 2008;43:1–22.

33. Urry FM, Kushnir MM, Nelson G, McDowell M, Jennison TA. Im-proving ion mass ratio performance at low concentrations in methamphetamine GC-MS assay through internal standard selec-tion. J Anal Tox 1996;20:592–595.

34. Seymour JL, Turechek F. Distinction and quantitation of leucine–isoleucine isomers and lysine– glutamine isobars by electrospray ionization tandem mass spectrometry of copper(II)-diimine com-plexes. J Mass Spectrom 2000;35:566–571.

35. Krishna P. Prabhakar S. Vairamani M. Differentiation of derivatized leucine and isoleucine by tandem mass spectrometry under liquid secondary ion mass spectral conditions. Rapid Commuin Mass Spectrom 1998;12:1429–1434.

36. Desaire H, Leary JA. Multicomponent quantification of di-astereomeric hexosamine monosaccharides using ion trap tandem mass spectrometry. Anal Chem 1999;71:4142–4147.

37. Bennett KH, Cook KD, Seebach GL. Simultaneous analysis of bu-tene isomer mixtures using process mass spectrometry. J Am Soc Mass Spectrom 2000;11:1079–1084.

38. Wan KX, Vidavsky I, Gross ML. Comparing similar spectra: from similarity index to spectral contrast angle. J Am Soc Mass Spectrom 2002;13:85–88.

39. Fenton WA, Rosenberg LE. Disorders of propionate and methyl-malonate metabolism. In: Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The metabolic and molecular bases of inherited disease, 7th ed. New York: McGraw-Hill, 1995:1423–1449.

40. Watkins D, Matiaszuk N, Rosenblatt DS. Complementation studies in the cblA class of inborn error of cobalamin metabolism: evidence for inter-allelic complementation and for a new complementation class. J Med Genet 2000;37:510–513.

41. Dharan M. Total Quality Control in the Clinical Laboratory, The C.V. Mosby Company, St. Louis, 1977.

42. Shultz EK, Aliferis C., Aronsky D. Clinical evaluation of methods. In: Tietz Textbook of clinical chemistry and molecular diagnostics. Ashwood ER, Burtis CA, Bruns DE eds. 4th edition, New York:

68

Saunders; 2005:409–424. 43. Ricos C, Alvarez V, Cava F, Garcia-Lario JV, Hernandez A,

Jimenez CV, Minchinela J, Perich C, Simon M. Current databases on biologic variation: pros, cons and progress. Scand J Clin Lab In-vest 1999;59:491–500.

44. Desirable specifications for total error, imprecision, and bias, de-rived from biologic variation. http://www.westgard.com/ biodata-base1.htm (accessed February 5, 2008).

45. Cornbleet PJ, Gochman N. Incorrect least-squares regression coeffi-cients in method-comparison analysis. Clin Chem 1979;25:432–8.

46. Taieb J, Mathian B, Millot F, Patricot MC, Mathieu E, Queyrel N, Lacroix I, Somma-Delpero C, Boudou P. Testosterone measured by 10 immunoassays and by isotope-dilution gas chromatography–mass spectrometry in sera from 116 men, women, and children. Clin Chem 2003;49:1381–1395.

47. Kushnir MM, Rockwood AL, Nelson GJ, Terry A, Meikle AW. Analysis of cortisol in urine by LC-MS/MS. Clin Chem 2003;49:6 965–967.

48. Meikle AW, Findling J, Kushnir MM, Rockwood AL, Nelson GJ, Terry AH. Pseudocushings syndrome cause by interference of drug fenofibrate with hplc assay for cortisol. J Clin Endocrinol Me-tab 2003;88:3521–3524.

49. King R, Bonfiglio R, Fernandez-Metzler C, Miller-Stein C, Olah T. Mechanistic investigation of ionization suppression in electrospray ionization. J Amer Soc Mass Spectrom 2000;11:942–950.

50. Bonfiglio R, King RC, Olah TV, Merkle K. The effects of sample preparation methods on the variability of the electrospray ionization response for model drug compounds. Rapid Commun Mass Spec-trom 1999;13:1175–1185.

51. Matuszewski BK, Constanzer ML, Chavez-Eng CM. Strategies for the assessment of matrix effect in quantitative bioanalytical methods based on HPLC-MS/MS. Anal Chem 2003;75:3019–3030.

52. Kaufmann A, Butcher P. Segmented post-column analyte addition; a concept for continuous response control of liquid chromatogra-phy/mass spectrometry peaks affected by signal suppres-sion/enhancement. Rapid Commun Mass Spectrom 2005:19:611–617.

53. Annesley TM. Ion suppression in mass spectrometry. Clin Chem 2003;49:1041–1044.

54. Annesley TM. Methanol-associated matrix effects in electrospray ionization tandem mass spectrometry. Clin Chem 2007;53:1827–1834.

55. Napoli KL. Organic solvents compromise performance of internal standard (Ascomycin) in proficiency testing of mass spectrometry-based assays for Tacrolimus. Clin Chem 2006;52:765–766.

69

56. Westgard JO, Klee GG. Quality management. In: Tietz textbook of clinical chemistry and molecular diagnostics. Ashwood ER, Burtis CA, Bruns DE eds. 4th edition, New York: Saunders; 2005:485–532.

57. Alström T, Gräsbeck R, Lindblad B, Solberg HE, Winkel P, Vi-inikka L. Establishing reference values from adults: recommenda-tion on procedures for the preparation of individuals, collection of blood, and handling and storage of specimens. Committee on refer-ence values of the scandinavian society for clinical chemistry. Scand J Clin Lab Invest 1993;5:649–652.

58. Ritchie RF, Palomaki G. Selection of clinically relevant populations for reference ranges. Clin Chem Lab Med 2004;42:702–709.

59. Solberg HE. The IFCC Recommendation on estimation of reference intervals. Clin Chem Lab Med 2004;42:710–4.

60. Gräsbeck R. The evolution of the reference value concept. Clin Chem Lab Med 2004;42:692–7.

61. Solberg HE. Establishment and use of reference values. In: Tietz textbook of clinical chemistry and molecular diagnostics. Ashwood ER, Burtis CA, Bruns DE eds. 4th edition, New York: Saunders; 2005:425–448.

62. Fitzgerald R, Herold DA. Serum total testosterone: immunoassay compared with negative chemical ionization gas chromatography–mass spectrometry. Clin Chem 1996;42:749–755.

63. Herold DA, Fitzgerald RL. Immunoassays for testosterone in women: better than a guess? Clin Chem 2003;49:1250–1251.

64. Wang C, Catlin D, Demers L, Starcevic B, Swerdloff RS. Meas-urement of total serum testosterone in adult men: Comparison of current laboratory methods versus liquid chromatography tandem mass spectrometry. J Clin Endocrinol Metab 2004;89:534–543.

65. Miller WL. Molecular biology of steroid hormone synthesis. En-docr Rev 1988;9:295–318.

66. Payne AH, Hales DB. Overview of steroidogenic enzymes in the pathway from cholesterol to active steroid hormones. Endocr Rev 2004;25:947–970.

67. Van Uytfanghe K, Stöckl D, Kaufman JM, Fiers T, Ross HA, De Leenheer AP, Thienpont LM. Evaluation of a candidate reference measurement procedure for serum free testosterone based on ultrafil-tration and isotope dilution-gas chromatography-mass spectrometry. Clin Chem 2004;50:2101–2110.

68. Van Uytfanghe K, Stöckl D, Ross HA, Thienpont LM. Use of frozen sera for FT4 standardization: investigation by equilibrium dialysis combined with isotope dilution-mass spectrometry and immunoas-say. Clin Chem 2006;52:1817–1821.

69. Sinha-Hikim I, Arver S, Beall G, Shen R, Guerrero M, Sattler F, Shikuma C, Nelson JC, Landgren BM, Mazer NA, Bhasin S. The use of a sensitive equilibrium dialysis method for the measurement

70

of free testosterone levels in healthy, cycling women and in human immunodeficiency virus-infected women. J Clin Endocrinol Metab 1998;83:1312–1318.

70. Vermeulen A, Verdonck L, Kaufman JM. A Critical evaluation of simple methods for the estimation of free testosterone in serum. J Clin Endocrinol Metab 1999;84:3666–3672.

71. Ly LP, Handelsman DJ. Empirical estimation of free testosterone from testosterone and sex hormone-binding globulin immunoassays. Eur J Endocrinol 2005;152:471–478.

72. Grumbach MM, Hughes I, Conte FA. Disorders of sexual differen-tiation. In: Larsen PR, Kronenberg HM, Melmed S, Polonsky KS, eds. Williams textbook of endocrinology. 10th ed. New York: Saun-ders 2002:842–1002.

73. De Peretti E, Forest MG. Pitfalls in the etiological diagnosis of con-genital adrenal hyperplasia in the early neonatal period. Horm Res 1982;16:10–22.

74. Summers RH, Herold DA, Seely BL. Hormonal and genetic analy-sis of a patient with congenital adrenal hyperplasia. Clin Chem 1996;42:1483–1487.

75. Pang S. Congenital adrenal hyperplasia owing to 3 beta-hydroxysteroid dehydrogenase deficiency. Endocrinol Metab Clin North Am 2001;30:81–99.

76. Moran C, Potter HD, Reyna R. Prevalence of 3beta-hydroxysteroid dehydrogenase-deficient nonclassic adrenal hyperplasia in hyperan-drogenic women with adrenal androgen excess. Am J Obstet Gyne-col 1999;181:596–600.

77. Asuncion M, Calvo RM, San Millan JL, Sancho J, Avila S, Escobar-Morreale HF. A prospective study of the prevalence of the polycystic ovary syndrome in unselected Caucasian women from Spain. J Clin Endocrinol Metab 2000;85:2434–2438.

78. Wong T, Shackleton CH, Covey TR, Ellis G. Identification of the steroids in neonatal plasma that interfere with 17 alpha-hydroxyprogesterone radioimmunoassays. Clin Chem 1992;38: 1830–1837.

79. Speiser PW. Improving neonatal screening for congenital adrenal hyperplasia. J Clin Endocrinol Metab 2004;89: 3685–3686.

80. Higashi T, Shimada K. Derivatization of neutral steroids to enhance their detection characteristics in liquid chromatography-spectrometry. Anal Bioanal Chem 2004;378:875–882.

81. Meikle AW, Kushnir MM, Rockwood AL, Pattison EG, Terry AH, Sandrock T, Bunker AM, Phanslkar AR, Owen WE, Roberts WL. Adrenal steroid concentrations in children seven to seventeen years of age. J Pediatr Endocrinol Metab 2007;20:1281–1291.

71

82. Kushnir MM, Rockwood AL, Roberts WL, Bergquist J, Varshavsky M, Yue B, Bunker AM, Meikle AW. High sensitivity tandem mass spectrometry test for serum estrogens. Clin Chem 2007;53:A183.

83. Stewart PM, Boulton A, Kumar S, Clark PM, Shackleton CHL. Cor-tisol metabolism in human obesity: impaired cortisone to cortisol conversion in subjects with central adiposity. J Clin Endocrinol Me-tab 1999;84:1022–1027.

84. Reinke M. Subclinical Cushing’s syndrome. Endocrinol Metab Clin North Am 2000;29:43–56.

85. Andrews RC, Walker BR. Glucocorticoids and insulin resistance: old hormones, new targets. Clin Sci 1999;96:513–523.

86. Edwards CRW, Walker BR, Benediktsson R, Seckl JR. Congenital and acquired syndrome of apparent mineralocorticoid excess. J Ster-oid Biochem Mol Biol 1983;45:1–5.

87. Stewart PM. 11-hydroxysteroid dehydrogenase: implications for clinical medicine. Clin Endocrinol (Oxf) 1996;44:493–499.

88. Robb DB, Covey TR, Bruins AP. Atmospheric pressure photoioni-zation: an ionization method for liquid chromatography-mass spec-trometry. Anal Chem 2000;72:3653–3659.

89. Hanold KA, Fischer SM, Cormia PH, Miller CE, Syage JA. Atmos-pheric pressure photoionization. 1. General properties for LC/MS. Anal Chem 2004;76:2842–2851.

90. Anderson JL, Carlquist JF, Roberts WL, Horne BD, May HT, Schwarz EL, Pasquali M, Nielson R, Kushnir MM, Rockwood AL, Bair TL, Muhlestein JB for the Intermountain Heart Collaborative (IHC) study group. Asymmetric dimethylarginine, cortisol/cortisone ratio, and C-peptide: Markers for diabetes and cardiovascular risk? Amer Heart J 2007;153:67–73.

91. Haymond S, Granovsky AM. Reproductive related disease. In: Tietz Textbook of clinical chemistry and molecular diagnostics. Ashwood ER, Burtis CA, Bruns DE eds. 4th edition, New York: Saunders; 2005:2097–2152.

92. Taieb J, Benattar C, Birr AS, Lindenbaum A. Limitations of steroid determination by direct immunoassay. Clin Chem 2002;48:583–585.

93. Matsumoto AM, Bremner WJ. Serum testosterone assays – accuracy matters. J Clin Endocrinol Metab 2004;89:520–4.

94. Rosner W, Auchus RJ, Azziz R, Sluss PM, Raff H. Utility, limita-tions, and pitfalls in measuring testosterone: an endocrine society position statement. J Clin Endocrinol Metab 2007;92:405–413.

95. Cawood ML, Field HP, Ford CG, Gillingwater S, Kicman A, Cowan D, Barth JH. Testosterone measurement by isotope-dilution liquid chromatography-tandem mass spectrometry: validation of a method for routine clinical practice. Clin Chem 2005;51:1472–1479.

96. Vicente FB, Smith FA, Sierra R, Wang S. Measurement of serum

72

testosterone using high-performance liquid chromatography/tandem mass spectrometry. Clin Chem Lab Med 2006;44:70–75.

97. Borrey D, Moerman E, Cockx A, Engelrelst V, Langlois MR. Col-umn-switching LC-MS/MS analysis for quantitative determination of testosterone in human serum. Clin Chim Acta 2007;382:134–137.

98. Turpeinen U, Linko S, Itkonen O, Hamalainen E. Determination of testosterone in serum by liquid chromatography-tandem mass spec-trometry. Scand J Clin Lab Invest 2007 Jun 24:1–10.

99. Choi HH, Gray PB, Storer TW, Calof OM, Woodhouse L, Singh AB, Padero C, Kushnir MM, Rockwood AL, Meikle AW, Mac RP, Sinha-Hikim I, Shen R, Dzekov J, Dzekov C, Lee M, Hays RD, Bhasin S. Effects of testosterone replacement in human immunode-ficiency virus-infected women with weight loss. J Clin Endocrinol Metab 2005;90:1531–1541.

100. Singh AB, Lee ML, Sinha-Hikim I, Kushnir MM, Meikle AW, Rockwood AL, Afework S, Bhasin S. Pharmacokinetics of a testos-terone gel in healthy postmenopausal women. J Clin Endocrinol Me-tab 2006;91:136–144.

101. Grinspoon S, Corcoran C, Miller K, Biller BM, Askari H, Wang E, HubbardJ, Anderson EJ, Basgoz N, Heller HM, Klibanski A Body composition and endocrine function in women with acquired immu-nodeficiency syndrome wasting. J Clin Endocrinol Metab 1997;82:1332–1337.

102. Coodley GO, Coodley MK. A trial of testosterone therapy for HIV associated weight loss. AIDS 1997;11:1347–1352.

103. Kong A, Edmonds P. Testosterone therapy in HIV wasting syn-drome: systematic review and meta-analysis. Lancet Infect Dis 2002;2:692–699.

104. Kol S. Hormonal therapy of the infertile women. In: Endocrine re-placement therapy in clinical practice Meikle AW ed., New Jersey: Humana; 2003:525–537.

105. Naessen T, Rodriguez-Macias K. Menopausal estrogen therapy counteracts normal aging effects on intima thickness, media thick-ness and intima/media ratio in carotid and femoral arteries. An in-vestigation using noninvasive high-frequency ultrasound. Athero-sclerosis 2006;189:387–392.

106. Iughetti L, Predieri B, Ferrari M, Gallo C, Livio L, Milioli S, Forese S, Bernasconi S. Diagnosis of central precocious puberty: endocrine assessment. J Pediatr Endocrinol Metab 2000;13:709–715

107. Singh M, Dykens JA, Simpkins JW. Novel mechanisms for estro-gen-induced neuroprotection. Exp Biol Med 2006;231:514–521.

108. Napoli N, Donepudi, S, Sheikh, S, Rini GB, Armamento-Villareal R. Increased 2-hydroxylation of estrogen in women with a family

73

history of osteoporosis. J Clin Endocrinol Metab 2005;90:2035–2041.

109. Green PS, Gordon K, Simpkins JW. Phenolic A ring requirement for the neuroprotective effects of steroids. J Steroid Biochem Mol Biol 1997;63:229–235.

110. Green PS, Simpkins JW. Neuroprotective effects of estrogens: po-tential mechanisms of action. Int J Dev Neurosci 2000;18:347–358

111. Amin S, Zhang Y, Felson DT, Sawin CT, Hannan MT, Wilson PW, Kiel DP. Estradiol, testosterone, and the risk for hip fractures in eld-erly men from the Framingham study. Am J Med 2006;119:426–433.

112. Dubey RK, Jackson EK. Genome and Hormones: Gender Differ-ences in Physiology: Invited Review: Cardiovascular protective ef-fects of 17beta-estradiol metabolites. J Appl Physiol 2001;91:1868–1883.

113. Naessen T, Persson I, Ljunghall S, Bergstrom R. Women with cli-macteric symptoms: a target group for prevention of rapid bone loss and osteoporosis. Osteoporos Int 1992;2:225–231.

114. Writing group for the women’s health initiative investigators. risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the women’s health initiative ran-domized controlled trial. JAMA 2002;288:321–333.

115. Leung YS, Dees K, Cyr R, Schloegel I, Kao PC. Falsely increased serum estradiol results reported in direct estradiol assays. Clin Chem 1997;43:1250–1251.

116. Sinicco A, Raiteri R, Rossati A, Savarino A, Di Perri G. Efavirenz Interference in estradiol ELISA assay. Clin Chem 2000;46:734–735.

117. Lee JS, Ettinger B, Stanczyk FZ, Vittinghoff E, Hanes V, Cauley JA, Chandler W, Settlage J, Beattie MS, Folkerd E, Dowsett M, Grady D, Cummings SR. Comparison of methods to measure low serum estradiol levels in postmenopausal women. J Clin Endocrinol Metab 2006;91:3791–3797.

118. Anari MR, Bakhtiar R, Zhu B, Huskey S, Franklin RB, Evans DC. Derivatization of ethinylestradiol with dansyl chloride to enhance electrospray ionization: application in trace analysis of ethinylestra-diol in rhesus monkey plasma. Anal Chem 2002;74:4136–4144.

119. Nelson RE, Grebe SK, O’Kane DJ, Singh RJ. Liquid chromatogra-phy-tandem mass spectrometry assay for simultaneous measurement of estradiol and estrone in human plasma. Clin Chem 2004;50:373–384.

120. Taguchi G. System of experimental design, Vol. 1 and 2. Kraus In-ternational, New York, 1987.

121. Klein KO, Baron J, Colli MJ, McDonnell DP, Cutler GB, Jr. Estro-gen levels in childhood determined by an ultrasensitive recombinant cell bioassay. J Clin Invest 1994;94:2475–2480.

74

122. The endocrine society and CDC collaborate to standardize testos-terone assays for clinical practice. http://www.endo-society.org/news/press/ (accessed: December 3, 2007).

123. Stanczyk FZ, Lee JS, Santen RJ. Standardization of steroid hor-mone assays: why, how, and when? Cancer Epidemiol Biomarkers Prev 2007;16:1713–1719.

124. Meier C, Nguyen TV, Handelsman DJ, Schindler C, Kushnir MM, Rockwood AL, Meikle AW, Jacqueline R, Center, JA Eisman, MJ Seibel. Endogenous sex hormones and incident fracture risk in older men: the Dubbo osteoporosis epidemiology study. Arch Intern Med 2008;168:47–54.

125. Mellström D, Johnell O, Ljunggren O, Eriksson AL, Lorentzon M, Mallmin H, Holmberg A, Redlund-Johnell I, Orwoll E, Ohlsson C. Free testosterone is an independent predictor of BMD and prevalent fractures in elderly men. J Bone Miner Res 2006;21:529–535.

126. Goderie-Plomp HW, van der Klift M, de Ronde W, Hofman A, de Jong FH, Pols HA. Endogenous sex hormones, sex hormone-binding globulin, and the risk of incident vertebral fractures in elderly men and women: the Rotterdam Study. J Clin Endocrinol Metab 2004;89:3261–3269.

127. Cover story: New testosterone assays may become gold standard. Endocrine Today 2008;6(2):1, 12-14.

128. Speroff L, Glass RH, Kase NG. Chapter 6: Regulation of the men-strual cycle. In: Mitchell C, editor. Clinical gynecologic endocrinol-ogy and infertility. 6th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 1999:202–246

129. Greenwald GS, Roy SK. Follicular development and its control. In: Knobil E, Neill JD, eds. The physiology of reproduction. 2nd ed. New York: Raven Press; 1994:629–724.

130. Walters KA, Allan CM, Handelsman DJ. Androgen Actions and the Ovary. Biol Reprod 2008;78:380-389.

131. Klein NA, Battaglia DE., Miller PB, Branigan EF, Giudice LC, Soules MR. Ovarian follicular development and the follicular fluid hormones and growth factors in normal women of advanced repro-ductive age. J Clin Endocrinol Metab 1996;81:1946–1951.

132. Brailly S, Gougeon A, Milgrom E, Bomsel-Helmreich, Papiernik E Androgens and progestins in the human ovarian follicle: differences in the evolution of preovulatory, healthy nonovulatory, and atretic follicles J. Clin. Endocrinol. Metab 1981;53:128–134.

133. Itskovitz J, Rubattu S, Rosenwaks Z, Liu HC, Sealey JE. Relation-ship of follicular fluid prorenin to oocyte maturation, steroid levels, and outcome of in vitro fertilization. J Clin Endocrinol Metab 1991;72:165–171.

134. Elting MW, Kwee J, Schats R, Rekers-Mombarg LTM, Shoemaker J. The rise of estradiol and inhibin B after acute stimulation with

75

follicle-stimulating hormone predict the follicle cohort size in women with polycystic ovary syndrome, regularly menstruating women with polycystic ovaries, and regularly menstruating women with normal ovaries. J Clin Endocrinol Metab 2001;86:1589–1595.

135. De Sutter P, Dhont M, Vanluchene E, Vandekerckhove E. Correla-tion between follicular fluid steroid analysis and maturity and cyto-genetic analysis of human oocytes that remained unfertilized after in vitro fertilization. Fertil Steril 1991;55:958–963.

136. Andersen CY. Characteristics of human follicular fluid associated with successful conception after in vitro fertilization. J Clin Endo-crinol Metab 1993;77:1227–1234.

137. Bergh C, Carlström K, Selleskog U, Hillensjö T. Effect of growth hormone on follicular fluid androgen levels in patients treated with gonadotropins before in vitro fertilization. Eur J Endocrinol 1996;134:190–196.

138. Smitz J, Andersen AN, Devroey P, Arce J-C, and the MERIT group. Endocrine profile in serum and follicular fluid differs after ovarian stimulation with HP-hMG or recombinant FSH in IVF pa-tients. Hum Reprod 2007;22:676–687.

139. McNatty KP, Smith DM, Makris A, Osathanondh R, Ryan KJ. The microenvironment of the human antral follicle: interrelationships among the steroid levels in antral fluid, the population of granulosa cells, and the status of the oocyte in vivo and in vitro. J Clin Endo-crinol Metab 1979;49(6):851–860.

140. Guo T, Taylor RL, Singh RJ, Soldin SJ. Simultaneous determina-tion of 12 steroids by isotope dilution liquid chromatography-photospray ionization tandem mass spectrometry. Clin Chim Acta 2006;372:76–82.

141. Xu X, Keefer LK, Waterhouse DJ, Saavedra JE, Veenstra TD, Ziegler RG. Measuring seven endogenous ketoic estrogens simulta-neously in human urine by high-performance liquid chromatogra-phy-mass spectrometry. Anal Chem 2004;76:5829–5836.

142. Xu X, Veenstra TD, Fox SD, Roman JM, Issaq HJ, Falk R, Saavedra JE, Keefer LK, Ziegler RG. Measuring fifteen endogenous estrogens simultaneously in human urine by high-performance liq-uid chromatography-mass spectrometry. Anal Chem 2005;77:6646 –6654.

143. Stein IF, Leventhal ML. Amenorrhea associated with bilateral polycystic ovaries. Am J Obstet Gynecol 1935;29:181.

144. Franks S. Polycystic ovary syndrome: a changing perspective. Clin Endocrinol (Oxf) 1989;31(1):87–120.

145. Polson DW, Adams J, Wadsworth J, Franks S. Polycystic ovaries- a common finding in normal women. Lancet 1988;1:870–872.

146. Clayton RN, Ogden V, Hodgkinson J, Worswick L, Rodin DA, Dyer S, Meade TW. How common are polycystic ovaries in normal

76

women and what is their significance for the fertility of the popula-tion? Clin Endocrinol 1992;37:127–134.

147. Franks S. Polycystic ovary syndrome. N Engl J Med 1995;333:853–861.

148. Talbott E, Guzick D, Clerici A, Berga S, Detre K, Weimer K, Kuller L. Coronary heart disease risk factors in women with poly-cystic ovary syndrome. Arterioscler Thromb Vasc Biol 1995;15:821–826.

149. Dahlgren E, Janson PO, Johansson S, Lapidus L, Oden A. Polycys-tic ovary syndrome and risk for myocardial infarction: Evaluated from a risk factor model based on a prospective population study of women. Acta Obstet Gynecol Scand 1992;71:599–604.

150. Rodin A, Thakkar H, Taylor N, Clayton R. Hyperandrogenism in polycystic ovary syndrome: Evidence of dysregulation of 11beta-hydroxysteroid dehydrogenase. N Engl J Med 1994;330:460–465.

151. Stewart P, Beastall G, Shackleton C, Edwards CR. 5-Alpha-reductase activity in polycystic ovary syndrome. Lancet 1990;335:431–433.

152. Yilmaz S, Fahrettin KT. The frequency of late-onset 21-hydroxylase and 11-hydroxylase deficiency in women with poly-cystic ovary syndrome. Eur J Endocr 1997;137:670–674.

153. Kirilovas D, Chaika A, Bergström M, Bergström-Petterman E, Carlström K, Nosenko J, Korniyenko S, Yakovets A, Mogilevkina I, Naessen T. Granulosa cell aromatase enzyme activity: effects of follicular fluid from patients with polycystic ovary syndrome, using aromatase conversion and [11C]vorozole-binding assays. Gynecol Endocrinol 2006;22:685–691.

154. Legro RS, Driscoll D, Strauss JF III, Fox J, Dunaif A. Evidence for a genetic basis for hyperandrogenemia in polycystic ovary syn-drome. Proc Natl Acad Sci USA 1998;95:14956–14960.

155. Urbanek M, Legro RS, Driscoll DA, Azziz R, Ehrmann DA, Nor-man RJ, Strauss JF 3rd, Spielman RS, Dunaif A. Thirty-seven can-didate genes for polycystic ovary syndrome: strongest evidence for linkage is with follistatin. Proc Natl Acad Sci USA 1999;96:8573–8578.

156. The Rotterdam ESHRE/ASRM-sponsored PCOS consensus work-shop group. Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome. Human Reprod 2004;19:41–47.

157. Koskinen P, Penttila TA, Anttila L, Erkkola R, Irjala K. Optimal use of hormone determinations in the biochemical diagnosis of the polycystic ovary syndrome. Fertil Steril 1996;65:517–522.

158. Turhan NO, Toppare MF, Seckin NC, Dilmen G. The predictive power of endocrine tests for the diagnosis of polycystic ovaries in women with oligoamenorrhea. Gynec Obst Invest 1999;48:183–186.

77

159. Luppa P, Müller B, Jacob K, Kimmig R, Strowitzki T, Höss C, Weber MM, Engelhardt D, Lobo RA. Variations of steroid hormone metabolities in serum and urine in polycystic ovary syndrome after nafarelin stimulation: Evidence for an altered corticoid excretion. J Clin Endocrinol Metab 1995;80:280–288.

160. Beck JR, Shultz EK. The use of receiver operating characteristic (ROC) curves in test performance evaluation. Arch Path Lab Med 1986;110:13–20.

161. Lasko TA, Bhagwat JG, Zou KH, Ohno-Machado L. The use of re-ceiver operating characteristic curves in biomedical informatics. J Biomed Inform 2005;38:404–415.

162. Cook NR. Statistical evaluation of prognostic versus diagnostic models: beyond the ROC curve. Clin Chem 2008;54:17–23.

163. Kushnir MM, Shushan B, Roberts WL, Pasquali M. Acylcarnitines and vitamin B12 deficiency. Clin Chem 2002;48:7 1126–1128.

164. Miller JW, Garrod MG, Rockwood AL, Kushnir MM, Allen LH, Haan MN, Green R. Holotranscobalamin II measurement improves the predictive value of total plasma B12 in screening for B12 defi-ciency. Clin Chem 2006;52:278–285.

Acta Universitatis UpsaliensisDigital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Science and Technology 426

Editor: The Dean of the Faculty of Science and Technology

A doctoral dissertation from the Faculty of Science andTechnology, Uppsala University, is usually a summary of anumber of papers. A few copies of the complete dissertationare kept at major Swedish research libraries, while thesummary alone is distributed internationally through theseries Digital Comprehensive Summaries of UppsalaDissertations from the Faculty of Science and Technology.(Prior to January, 2005, the series was published under thetitle “Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Science and Technology”.)

Distribution: publications.uu.seurn:nbn:se:uu:diva-8658

ACTA

UNIVERSITATIS

UPSALIENSIS

UPPSALA

2008

mass spectrometric applications for diagnosing metabolic ...171886/fulltext01.pdf · endocrinology...

Documents