Drug Testing and Analysis
Research article Received: 29 June 2013
Revised: 8 August 2013
Accepted: 9 August 2013
Published online in Wiley Online Library
(www.drugtestinganalysis.com) DOI 10.1002/dta.1533
Stable carbon isotope ratio profiling of illicit testosterone preparations – domestic and international seizures† Lance Brooker,* Adam Cawley, Jason Drury, Claire Edey, Nicole Hasick and Catrin Goebel Gas chromatography-combustion-isotope ratio mass spectrometry (GC-C-IRMS) is now established as a robust and mature analytical technique for the doping control of endogenous anabolic androgenic steroids in human sport. It relies on the assumption that the carbon isotope ratios of naturally produced steroids are significantly different to synthetically manufactured testosterone or testosterone prohormones used in commercial medical or dietary supplement products. Recent publications in this journal have highlighted the existence of black market testosterone preparations with carbon isotope ratios within the range reported for endogenous steroids (i.e. δ13C ≥ 25.8 ‰). In this study, we set out to profile domestic and international law enforcement seizures of illicit testosterone products to monitor the prevalence of ‘enriched’ substrates – which if administered to human subjects would be considered problematic for the use of current GC-C-IRMS methodologies for the doping control of testosterone in sport. The distribution of δ13C values for this illicit testosterone sample population (n = 283) ranged from 23.4 ‰ to 32.9 ‰ with mean and median of 28.6 ‰ – comparable to previous work. However, only 13 out of 283 testosterone samples (4.6 %) were found to display δ13C values ≥ 25.8 ‰, confirming that in the vast majority of cases of illicit testosterone administration, current GC-CIRMS doping control procedures would be capable of confirming misuse. Copyright © 2014 John Wiley & Sons, Ltd. Keywords: forensic science; doping control; anabolic androgenic steroids; testosterone; isotope ratio mass spectrometry
Introduction Gas chromatography-combustion-isotope ratio mass spectrometry (GC-C-IRMS) is now established as a robust and mature analytical technique for the doping control of endogenous anabolic androgenic steroids in human sport.[1,2] The carbon isotope ratio (CIR) test relies on the assumption that the δ13C of naturally produced steroids are significantly different to synthetically manufactured testosterone or testosterone prohormones used in commercial medical or dietary supplement products. In recent years, a multitude of GC-C-IRMS studies have been conducted by anti-doping laboratories on urine samples collected from drug-free human subjects.[3–12] Despite differences in sample preparation approaches, these reference population studies have consistently demonstrated that the normal population of δ13C of urinary steroids is found to be greater than or equal to 25.8 ‰. Likewise, studies have been undertaken to investigate the range of δ13C expected for synthetic testosterone originating from both legitimate medical and illicit black market products. Early studies demonstrated the principle that exogenous testosterone was depleted in 13C, finding that the tested materials had δ13C less than 25.9 ‰.[13–15] A comprehensive study undertaken in 2010 at the National Measurement Institute by Cawley et al. profiled legitimate pharmaceutical and veterinary formulations and illicit testosterone products seized at the border by the Australian Customs and Border Protection Service.[16] While the δ13C of legitimate preparations ranged as expected from 26.6 ‰ to 31.8 ‰, 24 out of 266 seized illicit testosterone products were found to have δ13C greater than or equal to 25.8 ‰. It was noted that many of these products had common
Drug Test. Analysis (2014)
features, for instance deliberate mislabelling of product details, seemingly to avoid border controls. Recently, Forsdahl et al. have published research pertaining to a similar case study.[17] Of 30 black market testosterone products seized in Austria and analyzed by GC-C-IRMS, more than half were within the range reported for endogenous steroids (the full data set ranged from 23.6 ‰ to 29.4 ‰). In the research presented below, further profiling of illicit testosterone preparations seized from Australia and around the world is undertaken to estimate the prevalence of ‘enriched’ substrates such as described by Cawley et al.[16] and Forsdahl et al.[17] which are considered problematic for the use of GC-C-IRMS methodologies for the doping control of testosterone in sport.
Experimental Reference materials Certified reference materials of testosterone, testosterone cypionate, testosterone decanoate, testosterone enanthate, and testosterone propionate were obtained from Chemical Reference Materials, National Measurement Institute (North Ryde, Australia).
* Correspondence to: Lance Brooker, Australian Sports Drug Testing Laboratory, National Measurement Institute, 105 Delhi Road, North Ryde, NSW 2113, Australia. E-mail: [email protected] †
Presented as part of the special issue; Advances in sports drug testing, 2013. Australian Sports Drug Testing Laboratory (ASDTL), National Measurement Institute, North Ryde, Australia
Copyright © 2014 John Wiley & Sons, Ltd.
Drug Testing and Analysis
L. Brooker et al.
Other steroid standards were obtained from Steraloids (Newport, RI, USA) and Sigma Chemical Co. (St Louis, MO, USA). 5αAndrostan-3β-ol-acetate, androsterone acetate, cholestane and 11-oxoetiocholanolone acetate (CU/USADA 33–1) were gifted from the Division of Nutritional Sciences, Cornell University (Ithaca, NY, USA).[18] C3 alkane mix and 5α-androstane were purchased from the Department of Geological Sciences, Indiana University (Bloomington, IN, USA). Chemicals and reagents All general laboratory chemicals and reagents were of analytical grade. All organic solvents were of HPLC grade from Merck (Darmstadt, Germany). Water was obtained using a Millipore filtration system (Bedford, MA, USA). Trimethylchlorosilane (TMCS) was purchased from Grace (Deerfield, PA, USA). Carbon dioxide (Food Fresh CO2), ultra high purity helium (> 99.999 %) and high purity oxygen (> 99.5 %) were purchased from BOC gases (Sydney, NSW, Australia). Testosterone preparations Sustanon 250 from Organon (Oss, Netherlands) [Batch #612603, Expiry 09–2008] containing a mix of testosterone isocaproate, propionate, decanoate, and phenylpropionate esters was analyzed with every batch as a quality control sample. Ten commercial pharmaceutical testosterone preparations sourced from Australia, Belgium and China were analyzed (Table 1). 283 illicit testosterone preparations that were seized by relevant authorities in the countries specified were analyzed (Table 2). They comprised of both low purity oils (injectables) and high purity powders (bulk chemicals). Sample preparation Initial purity estimation and ester identification was carried out by HPLC-UV analysis while subsequent hydrolysis of testosterone esters for GC-C-IRMS analysis was undertaken as previously described.[16] In brief, 0.5 mg of testosterone ester was subjected to acidic hydrolysis with trimethylchlorosilane in methanol. The reaction was stopped by addition of carbonate buffer and the liberated testosterone was extracted with hexane. Evaporation of the organic extract afforded a residue that was reconstituted in cyclohexane for GC-MS and GC-C-IRMS analysis. A blank steroid free sample and a Sustanon 250 were included as quality controls in every batch of 20 samples prepared.
Table 2. Summary of the country-of-seizure and form of illicit testosterone preparations analyzed Country
Total
Oil
Powder
Other
Australia Belgium Germany Switzerland USA Total
131 19 71 33 29 283
99 7 49 33 22 210
31 12 21
1 1
7 71
2
Gas chromatography–mass spectrometry An Agilent 6890 GC coupled to an Agilent 5973 MSD (Santa Clara, CA, USA) was used. The carrier gas was helium with a flow rate of 1.5 ml/min for an initial inlet pressure of 14.1 psi (constant flow). The injection volume was 1 μl (9:1 split) at 280°C. The GC column (0.25 mm I.D x 30 m) was a J&W DB17MS cross-linked 50% phenyl-methyl siloxane (0.25 μm film thickness). The column temperature was programmed from 70°C (1.5 min) to 250°C at 35°C/min, to 275°C at 2°C/min, then to 300°C at 10°C/min and held for 3 min. The MSD acquired data in scan mode from m/z 50 to m/z 500 using HP Chemstation® software (Santa Clara, CA, USA). Each sample extract was characterized by full scan GC-MS for retention time and electron impact mass spectra comparison of testosterone to certified reference material according to WADA criteria.[19] Peak purity of testosterone was confirmed by analysis of spectra across the peak to identify extraneous ion contributions from possible co-eluting compounds. Gas chromatography–combustion-isotope ratio mass spectrometry An Agilent 6890 GC (Santa Clara, CA, USA) equipped with a Gerstel CIS 4 PTV inlet controlled by Gerstel Maestro software (Mülheim, Germany) and Thermo A200S autosampler coupled to a Thermo GC-C III interface and Thermo Delta Plus IRMS (Bremen, Germany) was used for compound specific carbon isotope analysis of testosterone. The GC conditions were equivalent to those described for full scan GC-MS except injection was carried out in solvent vent mode (70°C (held 1.0 min) to 120°C at 1°C/s (held 0.3 min) then to 300°C at 12°C/s (held for 5 min) and a range of injection volumes were possible if necessary
Table 1. Summary of the legitimate pharmaceutical testosterone preparations analyzed Product Androderm Testogel Axiron Androgel Andriol Testocaps Andriol Testocaps Andriol Testocaps -
Ester Dose - 24.3 mg - 50 mg - 30 mg - 50 mg Undecanoate 40 mg Undecanoate 40 mg Undecanoate 40 mg Propionate 25 mg Propionate 25 mg Undecanoate 250 mg
wileyonlinelibrary.com/journal/dta
Form Transdermal Patch Transdermal Gel Transdermal Spray Transdermal Gel Oral oil Capsule Oral oil Capsule Oral oil Capsule I.M Ampoule I.M Ampoule I.M Ampoule
Manufacturer Watson Labs (Australia) Bayer Schering (Germany) Eli Lilly (Australia) Laboratories Besins-Iscovsco (Belgium) Organon (Netherlands) Organon (Netherlands) Organon (China) - (China) - (China) Xianju Pharma (China)
Copyright © 2014 John Wiley & Sons, Ltd.
13
Exp. date
δ CVPDB T ± 0.5 (‰)
09-2012 05-2013 01-2014 05-2010 08-2010 01-2013 01-2013 03-2014 06-2013 02-2015
-29.0 -29.2 -28.5 -29.0 -27.6 -29.2 -30.6 -29.5 -28.7 -28.1
Drug Test. Analysis (2014)
Stable carbon isotope ratio profiling of illicit testosterone preparations (0.5 μl–5 μl). Data was acquired using ISODAT® NT 2.0 software (ThermoScientific, Bremen, Germany). The oxidation reactor for combustion was operated at 940°C. High purity oxygen gas was flushed through the furnace for 30 min prior to a sequence and for 5 min every 15 injections during a sequence. The reduction reactor temperature was held at 650°C. The δ13C values, reported as per mille units (‰) relative to the Vienna Pee Dee Belemnite (VPDB) scale,[20] were measured relative to δ13C = 28.1 ‰ for the CO2 reference gas pulsed into the source at 200 s (Figure 1). The reference gas was calibrated using an approach adapted from Munton and colleagues[21] against the C3 alkane mix, 5α-androstane and CU/USADA 33–1 reference standards which are traceable to the carbon isotope ratio embodied in NBS-19, by which the VPBD scale is defined. A standard of testosterone (25 μg/ml) was analyzed at least every 15 injections for comparison to its assigned value ( 30.0 ± 0.4 ‰).[16] Figure 1 shows a GC-C-IRMS chromatogram with the retention time of testosterone (1165 s) from which identification was made with comparison to the full scan GC-MS analysis.
Drug Testing and Analysis
Quality control Validation data and measurement uncertainty calculations for the above methodology have been published previously in a peer reviewed publication.[16] To facilitate appropriate comparisons between that previous study and the work presented here, the same testosterone reference material and quality control sample were utilized. The mean obtained δ13C of the testosterone reference material ( 30.1 ± 0.5 ‰, n = 52) was not statistically different to the assigned value ( 30.0 ± 0.4 ‰) from the previous study using a one sample t-test (p = 0.8671). Likewise, the mean obtained δ13C of the Sustanon quality control preparation ( 29.8 ± 0.6 ‰, n = 29) was not statistically different to the assigned value ( 29.9 ± 0.3 ‰) using a one sample t-test (p = 0.2595). Statistical analysis Mean, median, standard deviations, t-tests and single-factor ANOVA were determined using Microsoft® Office for Mac Excel 2008 and GraphPad Prism 5 for Mac (GraphPad Software, Inc.).
Figure 1. GC-C-IRMS chromatogram of the testosterone reference standard (a) and typical extracted testosterone preparation (b). Note the presence of androstenedione eluting immediately after the testosterone peak. Small amounts are often found in testosterone preparations and reference materials.
Drug Test. Analysis (2014)
Copyright © 2014 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/dta
Drug Testing and Analysis
L. Brooker et al.
Results and discussion δ13C analysis of testosterone preparations by GC-C-IRMS Legitimate pharmaceutical preparations Ten recently obtained testosterone medical products were profiled from Australia (4), Belgium (2), and China (4). These included four transdermal testosterone formulations as well as oral and injectable testosterone ester preparations. The δ13C values ranged from 27.6 ‰ to 30.6 ‰ and were in agreement with earlier reports of pharmaceutical testosterone (Table 1). Depending on the dosage and the time of urine sample collection after administration, it would be expected that in the majority of doping cases where such products were used that GC-C-IRMS analysis of the resulting androgen metabolites would provide sufficient evidence of abuse from the perspective of current doping control protocols.
13
Figure 2. Frequency distribution of δ C values recorded for illicit testosterone preparations (n = 283).
Illicit testosterone preparations Two hundred and eighty-three Australian and international illicit testosterone preparations were collected for this study. The Australian sample set was seized by the Australian Customs and Border Protection Service in Australia and was analyzed by the Australian Forensic Drug Laboratory between 2010 and 2012. This set of 131 samples therefore follows on directly from those samples previously reported for the period 2006 to 2009.[16] It is unknown exactly when the samples from the international sample set were collected, and so no trend over time of the results can be inferred. The label for each data set refers to where the appropriate law enforcement authority seized the sample – not where the testosterone or testosterone ester was manufactured. It may be that the testosterone was manufactured in Belgium or Germany for example, but it is also just as likely that the testosterone was produced in China, India, Thailand or Mexico and trafficked to the ‘country-of-seizure’. The δ13C values measured for each sample group therefore reflect the distribution of testosterone available on the black market in those countries. The characteristics of the sample population with regard to the form of preparation and ester type were very similar to the previous study.[16] Testosterone enanthate esters were by far the most prevalent (n = 121), with the ratio of cypionate/ enanthate/propionate/mixed ester types approximately 1:8:3:5. Likewise, the ratio of low purity oils (injectables) to high purity solids (bulk materials) was 3:1 (210:71). Only one gel sample was analyzed despite the emergence of legitimate transdermal preparations for medical use in the last five years. One aqueous suspension of testosterone was analyzed which was highly contaminated with androstenedione. HPLC purification of the prepared extract enabled separation of the testosterone from the androstenedione and allowed determination of the free testosterone without interference.[22] The δ13C values for the illicit testosterone sample population (n = 283) displayed a normal distribution (D’Agostino & Pearson omnibus normality test) with mean and median of 28.6 ‰, minimum of 32.9 ‰ and maximum of 23.4 ‰. The frequency distribution of the population is displayed in Figure 2. Only 13 out of 283 testosterone samples (4.6 %) were found to display δ13C values within reference intervals reported for endogenous urinary androgen metabolites ( 17.3 ‰ to 25.8 ‰). This is a lower proportion than that reported by Cawley et al. in 2010[16] and substantially lower than described by Forsdahl et al. in
wileyonlinelibrary.com/journal/dta
2011.[17] There was no apparent trend in the origin or form of the 13C enriched substrates – samples originated from Australia, Germany, and the United States and were a mixture of oils and powders and various types of esters. It has been postulated that the 13C content of samples such as these might reflect new source materials being used to supply the demand of counterfeit testosterone products.[23] A further 47 samples (16.6 %) recorded δ13C values in the range 25.9 ‰ to 27.0 ‰, which is a larger proportion than previously reported. As suggested previously, if these products were administered the determination of their synthetic origin could be difficult in some athletes with endogenous urinary steroids that are somewhat 13C-depleted (e.g. athletes from northern Europe/Scandinavia).[24] This would especially be the case considering the potential dilution effect provided by endogenous steroid precursors synthesized de novo contributing to the pool of urinary steroid metabolites. All country-of-origin sample sets were considered normally distributed except for Australia (p = 0.0097, D’Agostino & Pearson omnibus normality test). The mean δ13C for Australia, Belgium, Germany, Switzerland, and the USA were 28.6 ‰, -28.5 ‰, 28.4 ‰, -28.3 ‰ and 29.3 ‰, respectively and were not considered significantly different by one-way ANOVA (p = 0.1084). From their distributions, which are presented in boxplot form in Figure 3 it can be seen that the US population is skewed to more depleted δ13C than the other three populations. Comparison of the forms present (low purity oils vs high purity bulk chemicals) in the illicit testosterone population was also undertaken. Both populations were normally distributed (D’Agostino & Pearson omnibus normality test) and their mean δ13C (oil 28.4 ‰ and powder 28.9 ‰) were considered significantly different from each other (t-test, p = 0.0336). For reference, the single gel and aqueous suspension samples analyzed recorded δ13C of 29.4 ‰ and 28.7 ‰, respectively. To complete a succinct comparison of the δ13C values derived from different ester types, all samples that contained multiple ester types were pooled to form a group named ‘mix’. All free testosterone single esters not found to be either enanthate, propionate, or cypionate were grouped under ‘other’ as their numbers were too few to allow valid statistical analysis separately. In contrast to the findings of Cawley et al.,[16] analysis of the ester types did not reveal a significant difference between
Copyright © 2014 John Wiley & Sons, Ltd.
Drug Test. Analysis (2014)
Stable carbon isotope ratio profiling of illicit testosterone preparations
Drug Testing and Analysis 13
Table 3. Summary statistics from the stable isotope ratio (δ C) profiling of domestic and international illicit testosterone preparations Parameter n Mean S.D Maximum IQR 1 (25% interval) Median IQR 3 (75% interval) Minimum
13
δ CVPDB T (‰) 283 28.6 1.7 23.4 27.2 28.6 29.8 32.9
Conclusion
13
Figure 3. Box-plot analysis of δ C values recorded from the country-ofseizure in the illicit testosterone population (n = 283).
the mean δ13C of the enanthate (n = 121, -28.5 ‰), propionate (n = 50, -28.5 ‰) and cypionate (n = 16, -29.4 ‰), or the “mix” (n = 82, -28.6 ‰) and ‘other’ (n = 14, -28.3 ‰) groups (one-way ANOVA, p = 0.3736) (Figure 4).
This study presents the most comprehensive and up-to-date carbon isotope ratio profiling of illicit testosterone products for doping control intelligence purposes, with seizures from three different continents represented (Australia, Europe, and North America). The population statistics of the 283 products are summarized in Table 3. For comparison, there was no statistical difference found between the mean of this population ( 28.6 ‰) and the previous work by Cawley et al. ( 28.4 ‰) (t-test, p = 0.1340).[16] Although a lower percentage of materials displayed δ13C values of testosterone greater than or equal to 25.8 ‰ (4.6 % compared to 9 %), a larger percentage was found between 25.9 ‰ to 27.0 ‰ (16.7 % compared to 9 %). This represents a sizeable proportion of the illicit testosterone population consisting of relatively enriched preparations (~ 20 % greater than 27.0 ‰). While it is acknowledged that administration of such preparations could prove troublesome for doping control methodologies utilizing GC-C-IRMS, it must also be emphasized that the success of the CIR test also depends on the basal value of the endogenous reference compounds utilized and the degree of endogenous dilution present. These factors are controlled by the diet of the athlete and the relative size of the endogenous urinary steroid metabolite pool compared to the dosage of exogenous testosterone administered and the time the sample was collected after administration. So while a minority of 13C enriched preparations may cause false-negative results in some particular populations of athletes, the CIR test will remain particularly effective for the majority of preparations in others – such as South Africa or Northern America. Likewise, the increased use of out-of-competition testing and targeted testing brought about by the use of the athlete steroid biological passport will see further improvements in the success rate of the CIR test in doping control. Acknowledgements
13
Figure 4. Box-plot analysis of δ C values recorded from the main three individual ester types in the illicit testosterone population.
Drug Test. Analysis (2014)
Stable isotope ratio analysis of illicit testosterone preparations was supported by funding from the Partnership for Clean Competition (PCC) Research Collaborative. The authors are most grateful to Jingzhu Wang and Youxuan Xu (National Anti-Doping Laboratory China Anti-Doping Agency, Beijing, China), Peter Van Eenoo (DoCoLab, Gent, Belgium), Detlef Thieme (Institute of Doping Analysis and Sports Biochemistry, Kreischa, Germany), Francios Marclay and Martial Saugy (Laboratoire Suisse d’Analyse du Dopage, Lausanne, Switzerland), Tim Laussmann (Centre for Education and Science of the Federal Finance Administration, Customs Laboratory Cologne, Germany)
Copyright © 2014 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/dta
Drug Testing and Analysis
L. Brooker et al.
and Jeffery Comparin (Special Testing and Research Laboratory, United States Drug Enforcement Administration, Dallas, Virginia, United States of America) for the provision of testosterone ester samples that enabled the completion of the international study component. Many thanks also to Guro Forsdahl and Gunter Gmeiner (Seibersdorf Labor GmbH Doping Control Laboratory, Seibersdorf, Austria) for sharing their enriched testosterone sample population with ASDTL for comparative analysis (data not shown). Testosterone preparations seized at the Australian border were submitted for analysis at the Australian Forensic Drug Laboratory (National Measurement Institute, North Ryde, Australia) by the Australian Customs and Border Protection Service. We gratefully acknowledge the contribution to this work of our colleagues at the Australian Forensic Drug Laboratory (AFDL), particularly Ryan Anderson and Chris Donnelly for the initial HPLC determination of the ester type and purity and for collating samples for subsequent GC-C-IRMS analysis in our laboratory. General thanks also to Aaron Heagney, Helen Salouros, Hilton Swan and Michael Collins from AFDL for encouraging an interest in forensic IRMS at NMI. The authors acknowledge the gifts of steroid reference materials with certified δ13C values by Thomas Brenna at Cornell University and the United States Anti-Doping Agency.
[9] [10]
[11] [12]
[13] [14] [15] [16] [17]
References [1] A.T. Cawley, U. Flenker. The application of carbon isotope ratio mass spectrometry to doping control. J. Mass Spectrom. 2008, 43, 854. [2] T. Piper, C. Emery, M. Saugy. Recent developments in the use of isotope ratio mass spectrometry in sports drug testing. Anal. Bioanal. Chem. 2011, 401, 433. 13 [3] U. Flenker, U. Güntner, W. Schänzer. δ C-Values of endogenous urinary steroids. Steroids 2008, 73, 408. [4] T. Piper, U. Mareck, H. Geyer, U. Flenker, M. Thevis, P. Platen, 13 12 W. Schänzer. Determination of C/ C ratios of endogenous urinary steroids: Method validation, reference population and application to doping control purposes, Rapid Commun. Mass Spectrom. 2008, 22, 2161. [5] A. Cawley, G. Trout, R. Kazlauskas, C. Howe, A. George. Carbon 13 isotope ratio (δ C) values of urinary steroids for doping control in sport. Steroids 2009, 74, 379. [6] E. Strahm, C. Emery, M. Saugy, J. Dvorak, C. Saudan. Detection of testosterone administration based on the carbon isotope ratio profiling of endogenous steroids: International reference populations of professional soccer players. Brit. J. Sports Med. 2009, DOI: 10.1136/ bjsm.2009.058669 [7] C. Buisson, C. Mongongu, C. Frelat, M. Jean-Baptiste, J. de Ceaurriz. Isotope ratio mass spectrometry analysis of the oxidation products of the main and minor metabolites of hydrocortisone and cortisone for antidoping controls. Steroids 2009, 74, 393. [8] L. Brooker, M.K. Parr, A. Cawley, U. Flenker, C. Howe, R. Kazlauskas, W. Schänzer, A. George. Development of criteria for the detection
wileyonlinelibrary.com/journal/dta
[18] [19]
[20]
[21]
[22]
[23] [24]
of adrenosterone administration by gas chromatography–mass spectrometry and gas chromatography-combustion-isotope ratio mass spectrometry for doping control. Drug Test. Anal. 2009, 1, 587. T.G. Sobolevskii, I.S. Prasalov, G.M. Rodchenkov. Carbon isotope mass spectrometry in doping control. J. Anal. Chem. 2010, 65, 825. L. Brooker, A. Cawley, R. Kazlauskas, C. Goebel, A. George. Carbon isotope ratio analysis of endogenous glucocorticoid urinary metabolites after cortisone acetate and adrenosterone administration for doping control. Drug Test. Anal. 2012, 4, 951. G. Green, R. Aguilera, B. Ahrens, B. Starcevic, F. Kurtzman, J. Su, D. Catlin. The influence of diet on iostope ratio mass spectrometry values. Clin. J. Sports Med. 2009, 19, 287. P. Van Renterghem, M. Polet, L. Brooker, W. Van Gansbeke, P. Van Eenoo. Development of a GC/C/IRMS method - Confirmation of a novel steroid profiling approach in doping control. Steroids 2012, 77, 1050. C.H.L. Shackleton, A. Phillips, T. Chang, Y. Li. Confirming testosterone administration by isotope ratio mass spectrometric anaysis of urinary androstanediols. Steroids 1997, 62, 379. M. Ueki, M. Okano. Analysis of exogenous dehydroepiandrosterone excretion in urine by gas chromatography/combustion/isotope ratio mass spectrometry. Rapid Commun. Mass Spectrom. 1999, 13, 2237. 13 12 X. de la Torre, J.C. Gonzàlez, S. Pichini, J.A. Pascual, J. Segura. C/ C Isotope ratio MS analysis of testosterone, in chemicals and pharmaceutical preparations. J. Pharmaceut. Biomed. 2001, 24, 645. A. Cawley, M. Collins, R. Kazlauskas, D.J. Handelsman, R. Heywood, M. Longworth, A. Arenas-Queralt. Stable isotope ratio profiling of testosterone preparations. Drug Test. Anal. 2010, 2, 557–567. G. Forsdahl, C. Osteicher, M. Koller, G. Gmeiner. Carbon isotope ratio determination and investigation of seized testosterone preparations. Drug Test. Anal. 2011, 3, 814. Y. Zhang, H.J. Tobias, T. Brenna. Steroid isotopic standards for gas chromatography-combustion isotope ratio mass spectrometry (GCC-IRMS). Steroids 2009, 74, 369. World Anti-Doping Agency. Identification crieria for qualitative assays incorporating column chromatography and mass spectrometry - TD2010IDCR. Available at: http://www.wada-ama. org/Documents/World_Anti-Doping_Program/WADP-IS-Laboratories/ Technical_Documents/WADA_TD2010IDCRv1.0_Identification Criteria for Qualitative Assays_May 08 2010_EN.doc.pdf [26 May 2012]. H. Craig. Isotopic standards for carbon and oxygen and correction factors for mass spectrometric analysis of carbon dioxide. Geochem. Cosmochim. Acta 1957, 12, 133. E. Munton, F.-H. Liu, E.J. Murby, D.B. Hibbert. Certification of steroid carbon isotope ratios in a freeze-dried human urine reference material. Drug Test. Anal. 2012, 4, 928. M.K. Kioussi, Y.S. Angelis, A.T. Cawley, M. Koupparis, R. Kazlauskas, J.T. Brenna, C. Georgakopoulos. External calibration in gas chromatographycombustion-isotope ratio mass spectrometry measurements of endogenous androgenic anabolic steroids in sports doping control. J. Chromatogr. A 2011, 1218, 5675. M.R. Graham, P. Ryan, J.S. Baker, B. Davies, N.-E. Thomas, S.-M. Cooper, P. Evans, S. Easmon, C.J. Walker, D. Cowan, A.T. Kicman. Counterfeiting in performance- and image-enhancing drugs. Drug Test. Anal. 2009, 1, 135. 13 12 T. Piper, U. Flenker, U. Mareck, W. Schänzer. C/ C Ratios of endogenous urinary steroids investigated for doping control purposes. Drug Test. Anal. 2009, 1, 65.
Copyright © 2014 John Wiley & Sons, Ltd.
Drug Test. Analysis (2014)
Research article Received: 29 June 2013
Revised: 8 August 2013
Accepted: 9 August 2013
Published online in Wiley Online Library
(www.drugtestinganalysis.com) DOI 10.1002/dta.1533
Stable carbon isotope ratio profiling of illicit testosterone preparations – domestic and international seizures† Lance Brooker,* Adam Cawley, Jason Drury, Claire Edey, Nicole Hasick and Catrin Goebel Gas chromatography-combustion-isotope ratio mass spectrometry (GC-C-IRMS) is now established as a robust and mature analytical technique for the doping control of endogenous anabolic androgenic steroids in human sport. It relies on the assumption that the carbon isotope ratios of naturally produced steroids are significantly different to synthetically manufactured testosterone or testosterone prohormones used in commercial medical or dietary supplement products. Recent publications in this journal have highlighted the existence of black market testosterone preparations with carbon isotope ratios within the range reported for endogenous steroids (i.e. δ13C ≥ 25.8 ‰). In this study, we set out to profile domestic and international law enforcement seizures of illicit testosterone products to monitor the prevalence of ‘enriched’ substrates – which if administered to human subjects would be considered problematic for the use of current GC-C-IRMS methodologies for the doping control of testosterone in sport. The distribution of δ13C values for this illicit testosterone sample population (n = 283) ranged from 23.4 ‰ to 32.9 ‰ with mean and median of 28.6 ‰ – comparable to previous work. However, only 13 out of 283 testosterone samples (4.6 %) were found to display δ13C values ≥ 25.8 ‰, confirming that in the vast majority of cases of illicit testosterone administration, current GC-CIRMS doping control procedures would be capable of confirming misuse. Copyright © 2014 John Wiley & Sons, Ltd. Keywords: forensic science; doping control; anabolic androgenic steroids; testosterone; isotope ratio mass spectrometry
Introduction Gas chromatography-combustion-isotope ratio mass spectrometry (GC-C-IRMS) is now established as a robust and mature analytical technique for the doping control of endogenous anabolic androgenic steroids in human sport.[1,2] The carbon isotope ratio (CIR) test relies on the assumption that the δ13C of naturally produced steroids are significantly different to synthetically manufactured testosterone or testosterone prohormones used in commercial medical or dietary supplement products. In recent years, a multitude of GC-C-IRMS studies have been conducted by anti-doping laboratories on urine samples collected from drug-free human subjects.[3–12] Despite differences in sample preparation approaches, these reference population studies have consistently demonstrated that the normal population of δ13C of urinary steroids is found to be greater than or equal to 25.8 ‰. Likewise, studies have been undertaken to investigate the range of δ13C expected for synthetic testosterone originating from both legitimate medical and illicit black market products. Early studies demonstrated the principle that exogenous testosterone was depleted in 13C, finding that the tested materials had δ13C less than 25.9 ‰.[13–15] A comprehensive study undertaken in 2010 at the National Measurement Institute by Cawley et al. profiled legitimate pharmaceutical and veterinary formulations and illicit testosterone products seized at the border by the Australian Customs and Border Protection Service.[16] While the δ13C of legitimate preparations ranged as expected from 26.6 ‰ to 31.8 ‰, 24 out of 266 seized illicit testosterone products were found to have δ13C greater than or equal to 25.8 ‰. It was noted that many of these products had common
Drug Test. Analysis (2014)
features, for instance deliberate mislabelling of product details, seemingly to avoid border controls. Recently, Forsdahl et al. have published research pertaining to a similar case study.[17] Of 30 black market testosterone products seized in Austria and analyzed by GC-C-IRMS, more than half were within the range reported for endogenous steroids (the full data set ranged from 23.6 ‰ to 29.4 ‰). In the research presented below, further profiling of illicit testosterone preparations seized from Australia and around the world is undertaken to estimate the prevalence of ‘enriched’ substrates such as described by Cawley et al.[16] and Forsdahl et al.[17] which are considered problematic for the use of GC-C-IRMS methodologies for the doping control of testosterone in sport.
Experimental Reference materials Certified reference materials of testosterone, testosterone cypionate, testosterone decanoate, testosterone enanthate, and testosterone propionate were obtained from Chemical Reference Materials, National Measurement Institute (North Ryde, Australia).
* Correspondence to: Lance Brooker, Australian Sports Drug Testing Laboratory, National Measurement Institute, 105 Delhi Road, North Ryde, NSW 2113, Australia. E-mail: [email protected] †
Presented as part of the special issue; Advances in sports drug testing, 2013. Australian Sports Drug Testing Laboratory (ASDTL), National Measurement Institute, North Ryde, Australia
Copyright © 2014 John Wiley & Sons, Ltd.
Drug Testing and Analysis
L. Brooker et al.
Other steroid standards were obtained from Steraloids (Newport, RI, USA) and Sigma Chemical Co. (St Louis, MO, USA). 5αAndrostan-3β-ol-acetate, androsterone acetate, cholestane and 11-oxoetiocholanolone acetate (CU/USADA 33–1) were gifted from the Division of Nutritional Sciences, Cornell University (Ithaca, NY, USA).[18] C3 alkane mix and 5α-androstane were purchased from the Department of Geological Sciences, Indiana University (Bloomington, IN, USA). Chemicals and reagents All general laboratory chemicals and reagents were of analytical grade. All organic solvents were of HPLC grade from Merck (Darmstadt, Germany). Water was obtained using a Millipore filtration system (Bedford, MA, USA). Trimethylchlorosilane (TMCS) was purchased from Grace (Deerfield, PA, USA). Carbon dioxide (Food Fresh CO2), ultra high purity helium (> 99.999 %) and high purity oxygen (> 99.5 %) were purchased from BOC gases (Sydney, NSW, Australia). Testosterone preparations Sustanon 250 from Organon (Oss, Netherlands) [Batch #612603, Expiry 09–2008] containing a mix of testosterone isocaproate, propionate, decanoate, and phenylpropionate esters was analyzed with every batch as a quality control sample. Ten commercial pharmaceutical testosterone preparations sourced from Australia, Belgium and China were analyzed (Table 1). 283 illicit testosterone preparations that were seized by relevant authorities in the countries specified were analyzed (Table 2). They comprised of both low purity oils (injectables) and high purity powders (bulk chemicals). Sample preparation Initial purity estimation and ester identification was carried out by HPLC-UV analysis while subsequent hydrolysis of testosterone esters for GC-C-IRMS analysis was undertaken as previously described.[16] In brief, 0.5 mg of testosterone ester was subjected to acidic hydrolysis with trimethylchlorosilane in methanol. The reaction was stopped by addition of carbonate buffer and the liberated testosterone was extracted with hexane. Evaporation of the organic extract afforded a residue that was reconstituted in cyclohexane for GC-MS and GC-C-IRMS analysis. A blank steroid free sample and a Sustanon 250 were included as quality controls in every batch of 20 samples prepared.
Table 2. Summary of the country-of-seizure and form of illicit testosterone preparations analyzed Country
Total
Oil
Powder
Other
Australia Belgium Germany Switzerland USA Total
131 19 71 33 29 283
99 7 49 33 22 210
31 12 21
1 1
7 71
2
Gas chromatography–mass spectrometry An Agilent 6890 GC coupled to an Agilent 5973 MSD (Santa Clara, CA, USA) was used. The carrier gas was helium with a flow rate of 1.5 ml/min for an initial inlet pressure of 14.1 psi (constant flow). The injection volume was 1 μl (9:1 split) at 280°C. The GC column (0.25 mm I.D x 30 m) was a J&W DB17MS cross-linked 50% phenyl-methyl siloxane (0.25 μm film thickness). The column temperature was programmed from 70°C (1.5 min) to 250°C at 35°C/min, to 275°C at 2°C/min, then to 300°C at 10°C/min and held for 3 min. The MSD acquired data in scan mode from m/z 50 to m/z 500 using HP Chemstation® software (Santa Clara, CA, USA). Each sample extract was characterized by full scan GC-MS for retention time and electron impact mass spectra comparison of testosterone to certified reference material according to WADA criteria.[19] Peak purity of testosterone was confirmed by analysis of spectra across the peak to identify extraneous ion contributions from possible co-eluting compounds. Gas chromatography–combustion-isotope ratio mass spectrometry An Agilent 6890 GC (Santa Clara, CA, USA) equipped with a Gerstel CIS 4 PTV inlet controlled by Gerstel Maestro software (Mülheim, Germany) and Thermo A200S autosampler coupled to a Thermo GC-C III interface and Thermo Delta Plus IRMS (Bremen, Germany) was used for compound specific carbon isotope analysis of testosterone. The GC conditions were equivalent to those described for full scan GC-MS except injection was carried out in solvent vent mode (70°C (held 1.0 min) to 120°C at 1°C/s (held 0.3 min) then to 300°C at 12°C/s (held for 5 min) and a range of injection volumes were possible if necessary
Table 1. Summary of the legitimate pharmaceutical testosterone preparations analyzed Product Androderm Testogel Axiron Androgel Andriol Testocaps Andriol Testocaps Andriol Testocaps -
Ester Dose - 24.3 mg - 50 mg - 30 mg - 50 mg Undecanoate 40 mg Undecanoate 40 mg Undecanoate 40 mg Propionate 25 mg Propionate 25 mg Undecanoate 250 mg
wileyonlinelibrary.com/journal/dta
Form Transdermal Patch Transdermal Gel Transdermal Spray Transdermal Gel Oral oil Capsule Oral oil Capsule Oral oil Capsule I.M Ampoule I.M Ampoule I.M Ampoule
Manufacturer Watson Labs (Australia) Bayer Schering (Germany) Eli Lilly (Australia) Laboratories Besins-Iscovsco (Belgium) Organon (Netherlands) Organon (Netherlands) Organon (China) - (China) - (China) Xianju Pharma (China)
Copyright © 2014 John Wiley & Sons, Ltd.
13
Exp. date
δ CVPDB T ± 0.5 (‰)
09-2012 05-2013 01-2014 05-2010 08-2010 01-2013 01-2013 03-2014 06-2013 02-2015
-29.0 -29.2 -28.5 -29.0 -27.6 -29.2 -30.6 -29.5 -28.7 -28.1
Drug Test. Analysis (2014)
Stable carbon isotope ratio profiling of illicit testosterone preparations (0.5 μl–5 μl). Data was acquired using ISODAT® NT 2.0 software (ThermoScientific, Bremen, Germany). The oxidation reactor for combustion was operated at 940°C. High purity oxygen gas was flushed through the furnace for 30 min prior to a sequence and for 5 min every 15 injections during a sequence. The reduction reactor temperature was held at 650°C. The δ13C values, reported as per mille units (‰) relative to the Vienna Pee Dee Belemnite (VPDB) scale,[20] were measured relative to δ13C = 28.1 ‰ for the CO2 reference gas pulsed into the source at 200 s (Figure 1). The reference gas was calibrated using an approach adapted from Munton and colleagues[21] against the C3 alkane mix, 5α-androstane and CU/USADA 33–1 reference standards which are traceable to the carbon isotope ratio embodied in NBS-19, by which the VPBD scale is defined. A standard of testosterone (25 μg/ml) was analyzed at least every 15 injections for comparison to its assigned value ( 30.0 ± 0.4 ‰).[16] Figure 1 shows a GC-C-IRMS chromatogram with the retention time of testosterone (1165 s) from which identification was made with comparison to the full scan GC-MS analysis.
Drug Testing and Analysis
Quality control Validation data and measurement uncertainty calculations for the above methodology have been published previously in a peer reviewed publication.[16] To facilitate appropriate comparisons between that previous study and the work presented here, the same testosterone reference material and quality control sample were utilized. The mean obtained δ13C of the testosterone reference material ( 30.1 ± 0.5 ‰, n = 52) was not statistically different to the assigned value ( 30.0 ± 0.4 ‰) from the previous study using a one sample t-test (p = 0.8671). Likewise, the mean obtained δ13C of the Sustanon quality control preparation ( 29.8 ± 0.6 ‰, n = 29) was not statistically different to the assigned value ( 29.9 ± 0.3 ‰) using a one sample t-test (p = 0.2595). Statistical analysis Mean, median, standard deviations, t-tests and single-factor ANOVA were determined using Microsoft® Office for Mac Excel 2008 and GraphPad Prism 5 for Mac (GraphPad Software, Inc.).
Figure 1. GC-C-IRMS chromatogram of the testosterone reference standard (a) and typical extracted testosterone preparation (b). Note the presence of androstenedione eluting immediately after the testosterone peak. Small amounts are often found in testosterone preparations and reference materials.
Drug Test. Analysis (2014)
Copyright © 2014 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/dta
Drug Testing and Analysis
L. Brooker et al.
Results and discussion δ13C analysis of testosterone preparations by GC-C-IRMS Legitimate pharmaceutical preparations Ten recently obtained testosterone medical products were profiled from Australia (4), Belgium (2), and China (4). These included four transdermal testosterone formulations as well as oral and injectable testosterone ester preparations. The δ13C values ranged from 27.6 ‰ to 30.6 ‰ and were in agreement with earlier reports of pharmaceutical testosterone (Table 1). Depending on the dosage and the time of urine sample collection after administration, it would be expected that in the majority of doping cases where such products were used that GC-C-IRMS analysis of the resulting androgen metabolites would provide sufficient evidence of abuse from the perspective of current doping control protocols.
13
Figure 2. Frequency distribution of δ C values recorded for illicit testosterone preparations (n = 283).
Illicit testosterone preparations Two hundred and eighty-three Australian and international illicit testosterone preparations were collected for this study. The Australian sample set was seized by the Australian Customs and Border Protection Service in Australia and was analyzed by the Australian Forensic Drug Laboratory between 2010 and 2012. This set of 131 samples therefore follows on directly from those samples previously reported for the period 2006 to 2009.[16] It is unknown exactly when the samples from the international sample set were collected, and so no trend over time of the results can be inferred. The label for each data set refers to where the appropriate law enforcement authority seized the sample – not where the testosterone or testosterone ester was manufactured. It may be that the testosterone was manufactured in Belgium or Germany for example, but it is also just as likely that the testosterone was produced in China, India, Thailand or Mexico and trafficked to the ‘country-of-seizure’. The δ13C values measured for each sample group therefore reflect the distribution of testosterone available on the black market in those countries. The characteristics of the sample population with regard to the form of preparation and ester type were very similar to the previous study.[16] Testosterone enanthate esters were by far the most prevalent (n = 121), with the ratio of cypionate/ enanthate/propionate/mixed ester types approximately 1:8:3:5. Likewise, the ratio of low purity oils (injectables) to high purity solids (bulk materials) was 3:1 (210:71). Only one gel sample was analyzed despite the emergence of legitimate transdermal preparations for medical use in the last five years. One aqueous suspension of testosterone was analyzed which was highly contaminated with androstenedione. HPLC purification of the prepared extract enabled separation of the testosterone from the androstenedione and allowed determination of the free testosterone without interference.[22] The δ13C values for the illicit testosterone sample population (n = 283) displayed a normal distribution (D’Agostino & Pearson omnibus normality test) with mean and median of 28.6 ‰, minimum of 32.9 ‰ and maximum of 23.4 ‰. The frequency distribution of the population is displayed in Figure 2. Only 13 out of 283 testosterone samples (4.6 %) were found to display δ13C values within reference intervals reported for endogenous urinary androgen metabolites ( 17.3 ‰ to 25.8 ‰). This is a lower proportion than that reported by Cawley et al. in 2010[16] and substantially lower than described by Forsdahl et al. in
wileyonlinelibrary.com/journal/dta
2011.[17] There was no apparent trend in the origin or form of the 13C enriched substrates – samples originated from Australia, Germany, and the United States and were a mixture of oils and powders and various types of esters. It has been postulated that the 13C content of samples such as these might reflect new source materials being used to supply the demand of counterfeit testosterone products.[23] A further 47 samples (16.6 %) recorded δ13C values in the range 25.9 ‰ to 27.0 ‰, which is a larger proportion than previously reported. As suggested previously, if these products were administered the determination of their synthetic origin could be difficult in some athletes with endogenous urinary steroids that are somewhat 13C-depleted (e.g. athletes from northern Europe/Scandinavia).[24] This would especially be the case considering the potential dilution effect provided by endogenous steroid precursors synthesized de novo contributing to the pool of urinary steroid metabolites. All country-of-origin sample sets were considered normally distributed except for Australia (p = 0.0097, D’Agostino & Pearson omnibus normality test). The mean δ13C for Australia, Belgium, Germany, Switzerland, and the USA were 28.6 ‰, -28.5 ‰, 28.4 ‰, -28.3 ‰ and 29.3 ‰, respectively and were not considered significantly different by one-way ANOVA (p = 0.1084). From their distributions, which are presented in boxplot form in Figure 3 it can be seen that the US population is skewed to more depleted δ13C than the other three populations. Comparison of the forms present (low purity oils vs high purity bulk chemicals) in the illicit testosterone population was also undertaken. Both populations were normally distributed (D’Agostino & Pearson omnibus normality test) and their mean δ13C (oil 28.4 ‰ and powder 28.9 ‰) were considered significantly different from each other (t-test, p = 0.0336). For reference, the single gel and aqueous suspension samples analyzed recorded δ13C of 29.4 ‰ and 28.7 ‰, respectively. To complete a succinct comparison of the δ13C values derived from different ester types, all samples that contained multiple ester types were pooled to form a group named ‘mix’. All free testosterone single esters not found to be either enanthate, propionate, or cypionate were grouped under ‘other’ as their numbers were too few to allow valid statistical analysis separately. In contrast to the findings of Cawley et al.,[16] analysis of the ester types did not reveal a significant difference between
Copyright © 2014 John Wiley & Sons, Ltd.
Drug Test. Analysis (2014)
Stable carbon isotope ratio profiling of illicit testosterone preparations
Drug Testing and Analysis 13
Table 3. Summary statistics from the stable isotope ratio (δ C) profiling of domestic and international illicit testosterone preparations Parameter n Mean S.D Maximum IQR 1 (25% interval) Median IQR 3 (75% interval) Minimum
13
δ CVPDB T (‰) 283 28.6 1.7 23.4 27.2 28.6 29.8 32.9
Conclusion
13
Figure 3. Box-plot analysis of δ C values recorded from the country-ofseizure in the illicit testosterone population (n = 283).
the mean δ13C of the enanthate (n = 121, -28.5 ‰), propionate (n = 50, -28.5 ‰) and cypionate (n = 16, -29.4 ‰), or the “mix” (n = 82, -28.6 ‰) and ‘other’ (n = 14, -28.3 ‰) groups (one-way ANOVA, p = 0.3736) (Figure 4).
This study presents the most comprehensive and up-to-date carbon isotope ratio profiling of illicit testosterone products for doping control intelligence purposes, with seizures from three different continents represented (Australia, Europe, and North America). The population statistics of the 283 products are summarized in Table 3. For comparison, there was no statistical difference found between the mean of this population ( 28.6 ‰) and the previous work by Cawley et al. ( 28.4 ‰) (t-test, p = 0.1340).[16] Although a lower percentage of materials displayed δ13C values of testosterone greater than or equal to 25.8 ‰ (4.6 % compared to 9 %), a larger percentage was found between 25.9 ‰ to 27.0 ‰ (16.7 % compared to 9 %). This represents a sizeable proportion of the illicit testosterone population consisting of relatively enriched preparations (~ 20 % greater than 27.0 ‰). While it is acknowledged that administration of such preparations could prove troublesome for doping control methodologies utilizing GC-C-IRMS, it must also be emphasized that the success of the CIR test also depends on the basal value of the endogenous reference compounds utilized and the degree of endogenous dilution present. These factors are controlled by the diet of the athlete and the relative size of the endogenous urinary steroid metabolite pool compared to the dosage of exogenous testosterone administered and the time the sample was collected after administration. So while a minority of 13C enriched preparations may cause false-negative results in some particular populations of athletes, the CIR test will remain particularly effective for the majority of preparations in others – such as South Africa or Northern America. Likewise, the increased use of out-of-competition testing and targeted testing brought about by the use of the athlete steroid biological passport will see further improvements in the success rate of the CIR test in doping control. Acknowledgements
13
Figure 4. Box-plot analysis of δ C values recorded from the main three individual ester types in the illicit testosterone population.
Drug Test. Analysis (2014)
Stable isotope ratio analysis of illicit testosterone preparations was supported by funding from the Partnership for Clean Competition (PCC) Research Collaborative. The authors are most grateful to Jingzhu Wang and Youxuan Xu (National Anti-Doping Laboratory China Anti-Doping Agency, Beijing, China), Peter Van Eenoo (DoCoLab, Gent, Belgium), Detlef Thieme (Institute of Doping Analysis and Sports Biochemistry, Kreischa, Germany), Francios Marclay and Martial Saugy (Laboratoire Suisse d’Analyse du Dopage, Lausanne, Switzerland), Tim Laussmann (Centre for Education and Science of the Federal Finance Administration, Customs Laboratory Cologne, Germany)
Copyright © 2014 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/dta
Drug Testing and Analysis
L. Brooker et al.
and Jeffery Comparin (Special Testing and Research Laboratory, United States Drug Enforcement Administration, Dallas, Virginia, United States of America) for the provision of testosterone ester samples that enabled the completion of the international study component. Many thanks also to Guro Forsdahl and Gunter Gmeiner (Seibersdorf Labor GmbH Doping Control Laboratory, Seibersdorf, Austria) for sharing their enriched testosterone sample population with ASDTL for comparative analysis (data not shown). Testosterone preparations seized at the Australian border were submitted for analysis at the Australian Forensic Drug Laboratory (National Measurement Institute, North Ryde, Australia) by the Australian Customs and Border Protection Service. We gratefully acknowledge the contribution to this work of our colleagues at the Australian Forensic Drug Laboratory (AFDL), particularly Ryan Anderson and Chris Donnelly for the initial HPLC determination of the ester type and purity and for collating samples for subsequent GC-C-IRMS analysis in our laboratory. General thanks also to Aaron Heagney, Helen Salouros, Hilton Swan and Michael Collins from AFDL for encouraging an interest in forensic IRMS at NMI. The authors acknowledge the gifts of steroid reference materials with certified δ13C values by Thomas Brenna at Cornell University and the United States Anti-Doping Agency.
[9] [10]
[11] [12]
[13] [14] [15] [16] [17]
References [1] A.T. Cawley, U. Flenker. The application of carbon isotope ratio mass spectrometry to doping control. J. Mass Spectrom. 2008, 43, 854. [2] T. Piper, C. Emery, M. Saugy. Recent developments in the use of isotope ratio mass spectrometry in sports drug testing. Anal. Bioanal. Chem. 2011, 401, 433. 13 [3] U. Flenker, U. Güntner, W. Schänzer. δ C-Values of endogenous urinary steroids. Steroids 2008, 73, 408. [4] T. Piper, U. Mareck, H. Geyer, U. Flenker, M. Thevis, P. Platen, 13 12 W. Schänzer. Determination of C/ C ratios of endogenous urinary steroids: Method validation, reference population and application to doping control purposes, Rapid Commun. Mass Spectrom. 2008, 22, 2161. [5] A. Cawley, G. Trout, R. Kazlauskas, C. Howe, A. George. Carbon 13 isotope ratio (δ C) values of urinary steroids for doping control in sport. Steroids 2009, 74, 379. [6] E. Strahm, C. Emery, M. Saugy, J. Dvorak, C. Saudan. Detection of testosterone administration based on the carbon isotope ratio profiling of endogenous steroids: International reference populations of professional soccer players. Brit. J. Sports Med. 2009, DOI: 10.1136/ bjsm.2009.058669 [7] C. Buisson, C. Mongongu, C. Frelat, M. Jean-Baptiste, J. de Ceaurriz. Isotope ratio mass spectrometry analysis of the oxidation products of the main and minor metabolites of hydrocortisone and cortisone for antidoping controls. Steroids 2009, 74, 393. [8] L. Brooker, M.K. Parr, A. Cawley, U. Flenker, C. Howe, R. Kazlauskas, W. Schänzer, A. George. Development of criteria for the detection
wileyonlinelibrary.com/journal/dta
[18] [19]
[20]
[21]
[22]
[23] [24]
of adrenosterone administration by gas chromatography–mass spectrometry and gas chromatography-combustion-isotope ratio mass spectrometry for doping control. Drug Test. Anal. 2009, 1, 587. T.G. Sobolevskii, I.S. Prasalov, G.M. Rodchenkov. Carbon isotope mass spectrometry in doping control. J. Anal. Chem. 2010, 65, 825. L. Brooker, A. Cawley, R. Kazlauskas, C. Goebel, A. George. Carbon isotope ratio analysis of endogenous glucocorticoid urinary metabolites after cortisone acetate and adrenosterone administration for doping control. Drug Test. Anal. 2012, 4, 951. G. Green, R. Aguilera, B. Ahrens, B. Starcevic, F. Kurtzman, J. Su, D. Catlin. The influence of diet on iostope ratio mass spectrometry values. Clin. J. Sports Med. 2009, 19, 287. P. Van Renterghem, M. Polet, L. Brooker, W. Van Gansbeke, P. Van Eenoo. Development of a GC/C/IRMS method - Confirmation of a novel steroid profiling approach in doping control. Steroids 2012, 77, 1050. C.H.L. Shackleton, A. Phillips, T. Chang, Y. Li. Confirming testosterone administration by isotope ratio mass spectrometric anaysis of urinary androstanediols. Steroids 1997, 62, 379. M. Ueki, M. Okano. Analysis of exogenous dehydroepiandrosterone excretion in urine by gas chromatography/combustion/isotope ratio mass spectrometry. Rapid Commun. Mass Spectrom. 1999, 13, 2237. 13 12 X. de la Torre, J.C. Gonzàlez, S. Pichini, J.A. Pascual, J. Segura. C/ C Isotope ratio MS analysis of testosterone, in chemicals and pharmaceutical preparations. J. Pharmaceut. Biomed. 2001, 24, 645. A. Cawley, M. Collins, R. Kazlauskas, D.J. Handelsman, R. Heywood, M. Longworth, A. Arenas-Queralt. Stable isotope ratio profiling of testosterone preparations. Drug Test. Anal. 2010, 2, 557–567. G. Forsdahl, C. Osteicher, M. Koller, G. Gmeiner. Carbon isotope ratio determination and investigation of seized testosterone preparations. Drug Test. Anal. 2011, 3, 814. Y. Zhang, H.J. Tobias, T. Brenna. Steroid isotopic standards for gas chromatography-combustion isotope ratio mass spectrometry (GCC-IRMS). Steroids 2009, 74, 369. World Anti-Doping Agency. Identification crieria for qualitative assays incorporating column chromatography and mass spectrometry - TD2010IDCR. Available at: http://www.wada-ama. org/Documents/World_Anti-Doping_Program/WADP-IS-Laboratories/ Technical_Documents/WADA_TD2010IDCRv1.0_Identification Criteria for Qualitative Assays_May 08 2010_EN.doc.pdf [26 May 2012]. H. Craig. Isotopic standards for carbon and oxygen and correction factors for mass spectrometric analysis of carbon dioxide. Geochem. Cosmochim. Acta 1957, 12, 133. E. Munton, F.-H. Liu, E.J. Murby, D.B. Hibbert. Certification of steroid carbon isotope ratios in a freeze-dried human urine reference material. Drug Test. Anal. 2012, 4, 928. M.K. Kioussi, Y.S. Angelis, A.T. Cawley, M. Koupparis, R. Kazlauskas, J.T. Brenna, C. Georgakopoulos. External calibration in gas chromatographycombustion-isotope ratio mass spectrometry measurements of endogenous androgenic anabolic steroids in sports doping control. J. Chromatogr. A 2011, 1218, 5675. M.R. Graham, P. Ryan, J.S. Baker, B. Davies, N.-E. Thomas, S.-M. Cooper, P. Evans, S. Easmon, C.J. Walker, D. Cowan, A.T. Kicman. Counterfeiting in performance- and image-enhancing drugs. Drug Test. Anal. 2009, 1, 135. 13 12 T. Piper, U. Flenker, U. Mareck, W. Schänzer. C/ C Ratios of endogenous urinary steroids investigated for doping control purposes. Drug Test. Anal. 2009, 1, 65.
Copyright © 2014 John Wiley & Sons, Ltd.
Drug Test. Analysis (2014)
