Projets de recherche

Code: T20M01MT

Higenamine, also known as norcolaurine, is a non-selective β2-agonist naturally occuring in different plants such as Nandida domestica, Tinospora crispa, and Annona quamosa [1-5]. Due to its bronchodilatative and stimulating effects, the misuse of higenamine in sports is prohibited at all times [6]. According to the relevant WADA technical document, a reporting level of 10 ng/mL (50% of the MInimum Required Performance Level (MRPL)) applies for the detection fo β2-agonists in doping control urine samples [7]. From 2016 till 2019, 201 samples werer reported as an adverse analytical finding (AAF) for higenamine [8-11].

Although higenamine has never been approved as a drug by the US Food and Drug Administration (FDA) [11], it plays an important role in traditional Chinese herbal medicine [1, 4]. Moreover, it was found tto be an (un)labeled ingredient of different weight loss and sports supplements [2-4, 12], which have caused several cases of unintentional doping during the last years [13-18].

As also some tropical fruit plants of the Annonaceae family were founhd to contain higenamine [2, 5], the aim of this study was to investigate whether the ingestion of such fruits can lead to AAFs in sports. The results of this reserach project can be of great value for anti-doping routine work, as they will help to ensure fair result management and decision-making processes in case of higenamine findings in sports drug testing programs.

Therefore, elimination studies were conducted in order to characterize the time-dependent urinary excretion of higenamine. Previous unpublished reports for L. C. Cameron's laboratory showed that the consumption of A. muricata (n=3) or A. squamosa (n=4) produced detectable higenamine in 100% of the subjects' urine (n=7). Otherwise, higenamine was not detectable in the C. papaya (control) group (n=3). In total, two administration studies with single dose of a fruit puree from Anonna muricata (330 g) and Anonna cherimola (330 g) were conducted, and the collected urine samples wre analyzed by means of liquid chromatography - tandem mass spectrometry (LC-MS/MS). HIgenamine detection times in general and, specifically, urinary concentrations of higenamine were desirable in support of an improved interpretation of AAFs, especially when scenarios of proven supplement contamination are debated and supplement administration protocols exist.

Main findings

The β2-agonist higenamine is prohibited in sports at all times and available, amongst others, as ingreditnets in various dietary supplements.
Further, a variety of plants including tropical fruits of the Annonaceae family have been reported to contain higenamine, and the aim of this study is to investigate whether the ingestion of such fruits can lead to AAFs in sports. For that purpose, single-dose elimination studies were conducted with A. muricata and a second higenamine-containing fruit (A. cherimola). Post-administration urine samples were analyzed concerning their higenamine content and assessed with regards to the currently enforced reporting level of 10 ng/mL.
All study volunteers produced urine samples containing higenamine after ingestion of the fruit preparations; however, under the chosen conditions, all observed urinary concentrations ranged exclusively below 5 ng/mL and, thus, the established MRL.

The herein obtained data can be considered in result management and decision-making processes in case of higenamine findings in sports drug testing programs. The outcome supports the position that single-dose administrations of these fruit species are rather unlikely to lead to AAFs in sports. Yet, substantial variability of the natural higenamine content in fruits exists, potentially influenced by seasonal/spatiotemporal factors, and also fruit processing and storage might affect the overall dietary higenamine availability. Hence, whilst less likely, it cannot be excluded that under specific circumstances the current MRL is exceeded by nutrition-derived higenamine uptake.

Code: DBS20AS1AT

Due to the aimed implementation of dried blood spot (DBS) sampling in doping controls for the upcoming Olympic events (Summer Olympic / Paralympic Games 2020 in Tokyo and Winter Olympic / Paralympic Games 2022 in Bejing), several targeted research projects were initiated by WADA’s DBS steering committee early 2020. To support realization of the ambitious timeline, the following projects are being applied for as targeted research. The first subproject deals with the stability of DBS samples. Of particular interest is how the stability of the substances in the dried state appears on the paper of the card. Factors such as temperature, light, and humidity should also be considered. Several model compounds from various classes of prohibited substances will be included in this study.

Main Findings

The stability of prohibited substances on dried blood spots (DBS) is generally assumed as superior to the storage under liquid conditions. Nevertheless, stability is not warranted per se for all target analytes and all conditions. Under consideration of the reanalysis of samples that were fortified with model compounds (originally stored for 365 days at room temperature, 4 °C or -20°C with/without desiccant) from various classes (anabolics, peptide hormones, β2-agonists, metabolic modulators, diuretics, stimulants, narcotics, glucocorticoids and β –blockers) and subsequently stored for another two years at -20°C showed excellent stability. All model compounds included were still detectable. Additionally, the potential impact of an intercontinental flight was found to be negligible for the selected model drugs. Another newly performed stability study over five months (at room temperature, 4 °C or -20°C with/without desiccant) with new model compounds (incl. HIF-stabilizers etc.) showed that storing at room temperature without desiccant and exposure to light will cause considerable losses for some of the highly volatile compounds. Results for storing at 4°C and -20°C showed equivalent results to the formerly performed study. Thus, storage at 4° C (or -20°C) with desiccant in the dark is recommended for long-term storage of DBS doping control samples.

Code: DBS20CT34JM

The minimally invasive dried blood spot (DBS) technique has the potential to improve the time-and-cost efficiency compared to traditional matrices in doping control. Understanding the athletes’ preferred sampling site will support the drafting of WADA Collection Guidelines/ISTI in the implementation process. Furthermore, the potential impact of the sampling site e.g. finger vs. arm on the concentration of target analytes needs to be established. Doping Control Officers (DCOs) will collect capillary blood from the finger (finger-prick) and from the upper arm (specific collection device) from 108 athletes (males and females) of various sport disciplines. The DCOs will record the time needed to collect the samples, register the number of unsuccessful attempts and evaluate the usability of the collection devices and whether they prefer the collection of DBS from the fingertip or the upper arm. The athletes will fill out a questionnaire regarding the perception and painfulness of the two DBS collections and whether the collections processes have had any impact on their sport activities afterwards. The lab staff will to fill out a questionnaire once they have received the samples to understand the suitability of the samples for analysis. The DBS samples will be analysed for the concentration of endogenous testosterone.

Main findings

Currently several DBS collection devices exists allowing collection of capillary blood from different anatomical sites. Nevertheless, the suitability for collecting DBS samples in an anti-doping context depends on the sample collection experience by the athletes, doping control officers (DCO) and the handling of the sample by the laboratory staff. Furthermore, agreement between quantitative measurements is important if more than one collection method (device and/or sampling site) is approved. In this project, a total of 108 matched DBS samples from the fingertip (HemaSpot HF; lancet device) and the upper arm (Tasso-M20; microneedle device) were collected from 49 female and 59 male national level athletes from various sports (handball, weightlifting, football, running and wheelchair rugby). Following sampling, the collection process was evaluated by athletes and DCOs. Furthermore, upon reception of the samples, the laboratory staff compared the quality and usability of the samples from the two devices. The testosterone concentration was measured in all samples and the correlation between concentrations determined.

Overall, the DBS sampling was associated with minimal sensation of pain and a high general experience by the athletes, but the perceived pain was rated lower (-0.4 ± 1.6, p < 0.05) and the general experience rated higher (+0.6 ± 2.3, p < 0.001) during upper-arm DBS collection than during DBS collection from the fingertip. Likewise, the DCOs rated the general experience with the upper-arm DBS collection higher (+1.6 ± 1.1, p < 0.01) than the fingerprick DBS collection, partly due to problems occurring more frequently during the DBS collection from the fingertip (28% of collections) than from the upper arm (6% of collections). Both procedures were equally fast, lasting only around two minutes on average, which is a great advantage compared to urine. When choosing, the great majority of DCOs and athletes, independent of gender and discipline, preferred the automated DBS collection from the upper arm over the manual collection from the fingertip, and both DBS collections over conventional sample collection methods (urine and venous blood collection).

Both devices provided easy access to the DBS sample and the overall analysis time was not affected by the choice of DBS material, however, the Tasso-M20 device was the preferred device amongst the four analysts preparing the samples due to less sample handling prior to sample preparation. Endogenous testosterone was quantified in DBS samples from both devices with good repeatability (RSD < 5%) and reproducibility (RSD < 10%). The quantitative analyses showed good correlation between samples collected from the fingertip and the upper arm from athletes (r = 0.921, p<0.0001), as well as non-significant difference of the median testosterone concentrations (1.70 ng/mL for Tasso and 1.67 ng/mL for HemaSpot, p = 0.503). A small, measured bias between Tasso and HemaSpot (-7.45%) was observed, likely due to unsatisfactory volume control in the HemaSpot compared to the Tasso device. Collectively, the results suggest that there is no physiological difference in the basal testosterone concentrations in capillary blood samples collected from the fingertip and the upper arm.

In conclusion, both an automated upper-arm DBS collection device [Tasso-M20] and a manual fingerprick DBS system [HemaSpot HF] could be used for DBS collection in an anti-doping setting. However, for subsequent quantitative analyses, the volumetric control of the Tasso-M20 spots seemed more robust than for HemaSpot HF. Both DCOs and laboratory personnel seem to prefer the upper-arm DBS collection with the Tasso-M20 device. If more than one collection method (device and/or sampling site) is approved, it appears crucial that the analytical assays are validated and calibrated on the respective DBS-devices/materials.

Code: DBS20AS7JM

The minimally invasive dried blood spot (DBS) technique has the potential to improve the time-and-cost efficiency compared to traditional matrices in doping control. The potential impact of the sampling site e.g. finger vs. arm on the concentration of target analytes needs to be established, especially when analyzing for threshold substances prohibited in-competition only. Eight healthy male volunteers will receive a single oral administration of ~20 mg (‘low dosage’) and 60 mg (‘high dosage’) of ephedrine in a randomized crossover design with one week between the interventions. Parallel DBS samples from the fingertip and upper arm will be collected at 0 (pre-administration control sample), 1, 2, 4, 6 and 8 hours post-administration. From the DBS samples the ephedrine concentration will be determined. Additionally, venous blood samples will be collected through a peripheral venous catheter on the same time points to compare the DBS concentrations of ephedrine with those in plasma.

Main Findings

Dried blood spot (DBS) testing allows for fast, easy, and minimally invasive collection of microvolumes of blood. In an anti-doping context, DBS testing has particularly relevance for substances prohibited in-competition only, as it can determine the presence of pharmacologically relevant blood concentrations during the in-competition period. A wide range of collection methods and devices exist for DBS collection allowing collection of capillary blood from different anatomical sites, but the possibility to use different devices in an anti-doping setting would rely on agreement in substance concentrations between sampling sites and between devices. Furthermore, it is of interest to evaluate the agreement between concentrations of target analytes in DBS and conventional venous plasma samples. Herein, we collected matched upper-arm DBS, finger prick DBS and venous plasma samples from 8 healthy, male subjects in an 8-hour period following oral administrations of 20 mg (‘low dose’) and 60 mg (‘high dose’) of ephedrine. We observed no consistent trend in the dependence of ephedrine concentration on blood sampling site or sampling device, and the correlations between ephedrine concentrations in finger prick DBS and upper-arm DBS were very high (Pearson’s r > 0.80) after both low and high dose administration. These results indicate that DBS originating from finger prick and automated upper-arm collection, along with conventional venous blood samples, can be used for quantification of ephedrine in doping control.

Code: DBS20AS7AT

This project will assess the introduction of cut-off limits for substances that are only prohibited in competition (stimulants, corticoids, cannabinoids, and narcotics). Data about the pharmacologically relevant levels of each substance are required to interpret the measured results correctly.

Main Findings

In competition sport drug testing represents an important aspect in doping controls. Here the analysis of blood samples owns a considerable benefit compared to urine samples due to the determination of actually relevant blood concentrations during the competition. DBS sampling represents a simple, reliable, cost-efficient and robust approach to ideally support the classical urine analysis, especially for in competition testing. In the present study potential analytical cut-off limits for stimulants, corticoids, cannabinoids, and narcotics are proposed. The values are based on literature review for pharmacokinetic data as well as already existing levels valid for driving under the influence of drugs.

Code: DBS20AS2AT

This project will investigate the possibility to measure the hematocrit value of the dried blood spot (DBS) sample by near-infrared spectroscopy (NIR). This can have an influence on the result, especially when the data are evaluated quantitatively. For this purpose, the results of the NIR measurements will be compared with reference data (Sysmex etc.) obtained simultaneously.

Main Findings

In contrast to established blood sampling strategies (yielding serum or EDTA plasma), DBS consist of whole blood. Therefore, the knowledge about the hematocrit (as percentage of red blood cells) of the DBS sample represents an important parameter especially for quantitative results interpretation. Here, the hematocrit measurement with near-infrared spectroscopy (NIR) from cellulose-based DBS paper represents a valid and reliable approach. After complete drying of the cards, the non-destructive NIR-analysis enables robust hematocrit (Hct) measurements over weeks and presumably months. The correction for the actual hematocrit of the finger prick DBS samples facilitates the accurate correlation to the resulting and comparable plasma levels of the respective drugs. With regard to in-competition DBS sampling, this possibility will enhance the result interpretation significantly. In the present project, different whole blood sampling strategies (venous, finger prick, TAP, Tasso, capillary) and the subsequent hematocrit measurement (Sysmex, NIR, centrifugation) were compared. All measurements for venous EDTA-blood and capillary finger blood (finger prick) were < 10% of relative deviation compared to the ‘true’ Sysmex value. This is also true regardless of the method of measurement. However, collection of capillary blood on heparin (here using the TAP device) and subsequent analysis by NIR led to greater Hct values and a mean relative deviation ˃10%. Transfer to other laboratory is not hindered, because NIR analysis is based on calibration-based technology, which can be adapted and transferred to any other instrument using the same technology. It was additionally shown that the repeated exposure to NIR does not have a measurable impact on the subsequent chemical analysis of the prohibited drugs.

Code: DBS20AS4DE

Recent work in the anti-doping field has clearly shown that the future of blood collections is dependent on direct capillary blood collection, moving away from traditional venous draws. In that light, this project will focus only on the detectability of endogenous (blood EPO, or bEPO) and recombinant EPO in samples collected directly with capillary collection devices and will avoid venous blood spotted onto a matrix. Blood samples, both from control volunteers and patients who have been administered rEPO, will be collected using Tasso and OneDraw devices, and finger pricks.

Extraction of blood (and EPO) from the spot will be optimized using in-house protocols. Once extracted, two methods of EPO purification will be attempted. First, a conventional magnetic bead immunopurification method utilizing anti-EPO antibodies to capture the EPO in the dried blood sample will be tested (adapted from Desharnais, 2017). Next, the commercially available MAIIA EPO purification gel kit will also be utilized for efficacy. Once purified, samples will be analyzed via SAR-PAGE and Western Blotting, using the biotinylated monoclonal anti-EPO antibody described in previous Cologne Workshops (Reichel 2018, Dehnes 2020) to provide a higher quality protein signal. Using the methods described above, the sensitivity and specificity of bEPO and rEPO in DBS will be characterized.

Main findings

This project was executed as a follow-up verification from work completed in the Barcelona laboratory in 2018 in which various ESAs were detected in dried blood spots using electrophoretic methods. The purpose of this project was to verify these results in a second laboratory and to answer the following questions:

1. What is the optimal method to extract and purify EPO from a dried capillary blood spot?

2. How does the optimal method perform in samples collected directly from capillary blood, specifically in samples from a drug administration?

Extraction and purification using StemCell ELISA and anti-EPO magnetic bead technology were studied in this project. Although MAIIA technology was also proposed as a purification method, this technology is currently under study in other laboratories and therefore was omitted from this project. Initial and follow-up studies both showed superior extraction and purification using the StemCell ELISA method, and thus this method was deemed more appropriate for ESA analysis in DBS.

Next, a clinical study was designed in which nine male subjects each received a single dose of EPOGEN (epoetin alfa) at a dose of 40 IU/kg s.c. Finger prick (spotted onto DMPK-C cards) and capillary blood from Tasso M20 devices was collected from each patient before their dose and then intermittently up to 72 hours after their dose. All samples were extracted using the StemCell ELISA method and were analyzed by SAR-PAGE and Western Blotting according to laboratory standard operating procedures. On average, rEPO was detectable in both finger prick and Tasso M20 samples up to 48 hours.

In most cases, the rEPO detected in the samples appeared on the gels as a double band, with the recombinant portion migrating on the gel similarly to the Dynepo standard and the endogenous portion migrating further. While this has been seen in urine and blood samples following controlled administrations, it differs from the conventional ‘mixed band’ or ‘smeared band’ commonly seen with rEPO-positive samples.

Finally, while rEPO was detectable in capillary blood samples, due to factors including low concentrations of EPO in the samples and low sample volume, gel quality was an issue throughout the study. In many instances, the EPO content on the gel was low enough such that the contrast between EPO and background provided interpretation issues. It is expected that using larger sample volumes should eliminate some of these concerns.

Overall, results from this study validate the data observed in the Barcelona in 2018, showing that doses of rEPO can be detected in DBS. Prior to implementing into routine doping

control, method improvement is recommended regarding increasing sample volumes (pending MAIIA purification results). Additionally, it is recommended that a technical document be drafted by the EPO Working Group outlining sample preparation proced

Code: DBS20AS5AT

This project deals with the stability of steroid esters on DBS cards, which represent a frequently misused class of prohibited compounds. As is well known, these esters are hydrolyzed by endogenous esterases in the blood (even after sampling in serum or plasma) and this can falsify the results accordingly. Whether and to what extent this cleavage also occurs when using DBS will be investigated here.

Main findings

Anabolic steroids represent one of the most frequently misused class of compounds in sports due to their performance enhancing properties. These steroids are often applied as esters with variable length of the non-polar ester side chain that have a considerable impact of the pharmacology of the drug. These steroid esters are known for their degradation due to endogenous esterases, which are present in blood also after the sampling process. The addition of esterase inhibiting agents (such as sodium fluoride) enables a slight inhibition of the degradation in the sampling device. The present study was conducted to confirm the potential of DBS sampling enabling the storage of DBS samples with subsequent analysis of intact steroid esters. Interestingly, the storage of whole blood samples as DBS (without any esterase inhibitors) will provide a useful approach to ensure better stability during long-term storage compared to classic liquid whole blood samples. Also the application of EDTA whole blood samples (e.g. derived from the athlete biological passport sampling) to DBS cards enables the storage for subsequent analysis of steroid esters. Here, the time period from sampling until spotting should be minimized (e.g. < 2 hours). This study shows also that storage at 4°C or -20°C with a desiccant in the dark provides the best results for all steroid esters. Here even after five months of storage the esters were still detectable on the DBS cards.

Code: R19M02RF

Nearly 30 000 blood samples are collected yearly for the athlete biological passport (ABP). Therefore, accurate and precise measurement of blood variable with low bias are paramount to ensure the indirect detection and targeting potential of the ABP. The blood matrix as a suspension of living cells in plasma ensures oxygen transport to the working muscles thanks to the red blood cells. While circulation allows to maintain a constant composition of the blood, variations in the fluid balance may inevitably alter variables for which concentration values are reported (e.g., hemoglobin concentration ([Hb]) or hemotocrit (Hct)) while absolute measures (e.g., reticulocytes percentage (Ret%)) remain stable. In this context, numerous confounding factors affect blood variables and robust procedures are needed to limit pre-analytical variations of blood vaiables when analyzed for the ABP. Currently, the guidelines specify that 2h waiting is necessary after any physical exercise and require the athletes to be seated for 10 min before sample collection. The aim of this study is to investigate the influence of body position prior to and during phlebotomy (i.e. seated vs. supine) on blood variables reported in the ABP by collecting successive samples over 90 min in either position. Additionally, this study assesses if a short position change (e.g., walking a short distance from a waiting room to the sample collection site) influences the reading and may thus be acceptabe in the context of an antidoping blood sample collection sequence.

Main findings

The Athlete's Biological Passport (ABP) is a tool for the indirect detection of blood doping. Current guidelines from the World Anti-Doping Agency (WADA) require a delay of 2 hours after any physical exercise and to be seated for 10 minutes prior to any blood sampling to obtain a valid measurement.

Accurate and precise measurement of blood variables with low bias are paramount to ensure the indirect detection and targeting potential of the ABP. The blood matrix as a suspension of living cells in plasma transports oxygen to the working muscles. Variations in the fluid balance and thus plasma volume may inevitable alter variables for which concentration values are reported (e.g., haemoglobin concentration ([Hb]) or haematocrit (Hct)) while absolute measures (e.g., reticulocytes ercentage (Ret%)) remain stable.

Since body position prior to and during phlebotomy may influence the outcome, this study compared blood biomarker variations with changes in body position during blood sample collection. Ten successive venous blood samples from 38 subjects of 3 groups (elite cyclists, apnea divers and controls) in three situations (seated, after a 50 m walk, and supine) were collected and analyzed via flow cytometry. While reticulocytes percentage was unchanged in all conditions, haemoglobin concentration and hematocrit were stable after at least 10 min in a seated position. Due to shifts in plasma volume, the measure were significantly higher after changing posture for a short walk, but readjusted to previous levels after only 5 min. Supine position caused generally lower values after 10-30 min.

In conclusion, our study indicates that standing up shortly during an antidoping blood collection process (and walking up to 50 m to change seats for example) significantly alters the [Hb} and Hct values in athletes and healthy subjects. Values however stabilized after 5 min upon returning in a seated position. Blood sampled in a supine position may result in lower [HB] and Hct values that can affect an ABP profile. Blood samples for anti-doping purposes in the context of the ABP should therefore not be collected in a supine position. If a subject has to stand up shortly after having waited for 10 min (e.g., to change seat from a waiting room to the phlebotomy location), acceptable samples could be obtained after 5 min in a seated position. These findings can complement the current WADA guidelines for blood sampling in the context of the ABP.

Code: 19A09XD

The relationship among the dehydroepiandrosterone (DHEA) and its metabolites 7-hydroxylated and 7-keto have been widely described leading to think that their presence in urine supposes a DHEA abuse. On the other hand, we have confirmed the 3-deoxylation forming structures androst-3,5-diene-7-keto under acidic conditions; in that sense, the synthesis of arimistane is possible from 7-keto-DHEA under the conditions of the common procedures applied to detect steroids in antidoping laboratories. 7-keto-DHEA is well known by having no influence on androgens or estrogens metabolism; and that the configuration androst-3,5-diene-7-keto is essential to inhibition of aromatase occurs.

Preliminary results (n=1) showed that after a single oral dose of 7-keto-DHEA no alterations of the endogenous steroid profile occurred, the presence of arimistane and its main metabolite (7β-hydroxilated metabolite), and finally we obtained negative results for pseudo-endogenous steroids (i.e. testosterone, its main metabolites and DHEA) after the analysis by GC/C/IRMS. Several reduced, oxidized metabolites and poly-hydroxylated metabolites were found using GC/QTOF, most of them were not present in negative urine. These poly-hydroxylated metabolites and other isomers founded have not been described for DHEA so, probably could be specific of the 7-keto-DHEA metabolism.

It is then necessary to increase the volunteers looking for the individual variability in order (1) to confirm these results by GC/C/IRMS, (2) to confirm the true origin of the arimistane after the administration of 7-keto-DHEA searching for a suitable assay, (3) to look for the specific metabolites that discriminates between DHEA and 7-keto-DHEA administration, and finally (4) to correctly assign 7-keto-DHEA on the WADA Prohibited List considering the antiestrogenic properties according its chemical structure.

Main Findings

Preliminary data showed some cross-metabolic findings among DHEA, arimistane and 7-keto-DHEA (three compounds belong to two different sections of the WADA Prohibited List), so it is necessary to ascertain specific metabolic findings that can be assigned to each administration. Arismistane, at a single dose, does not provoke evident alterations in the urinary endogenous steroid profile. Arimistane itself is only detectable in the sulfate fraction more as an artifact or degradation product than to an actual phase II metabolite. The analysis of the sulfate fraction has some limitations. Taking into account the C3 de-oxygenation of the androst-3,5-diene-7-oxo structures under acidic conditions, its presence in the sulfated fraction may be due to the C3 de-oxygenation of endogenous 7-keto-DHEA. Twelve arimistane metabolites were described after GC and LC analysis. The main metabolite 7β-hydroxy-arimistane showed the longest-term excretion. Contrary to GC, traces of arimistane PC were observed by LC-MS analysis. Hydroxylated metabolites of arimistane in C2 was proposed although additional spectrometric techniques must be applied to finally elucidate the structure. Alternatively, the chemical synthesis of the compound is needed. The presence of arimistane in concentrations higher than its 7β-hydroxy metabolite could be an adequate marker because, after the administration of arimistane, itself is almost undetectable in urine. The detection of 7-keto-DHEA administration based on the endogenous steroid profile of ABP and the consequent analysis by GC-C-IRMS failed. IRMS data supported the fact that there is no back-formation of DHEA from 7-keto-DHEA. Ten metabolites proposed by GC-MS and LC-MS analyses were proposed, in addition to 7-keto-DHEA itself. They showed considerable responses in both the free+glucuronated and sulfate fractions. Among them, there were four metabolites excreted for longer times and with higher responses. Additional spectrometric techniques or the syntheses of the proposed structures are needed for a definitive confirmation of the configurations. Although considered a degradation metabolite, arimistane was observed in samples post- 7-keto-DHEA administration. After the analysis of samples from 5 volunteers, no arimistane metabolite (7β-hydroxy) was found. The metabolites are specific enough to avoid GC-C-IRMS confirmations. Artifacts may be produced during sample preparation and instrumental analysis.

After the GC analysis of trimethylsylil derivatives of 7-keto-DHEA 7α-OH-DHEA, arimistane was identified. Arimistane signal was proportional to the concentration of derivatization mixture used. Meanwhile, 7β-hydroxy-DHEA showed no degradation. Analysis by LC instruments (avoiding derivatization and high temperatures of the injector port) showed that 7-keto-DHEA preparations in protic solvents as MeOH favor the dehydration of the molecule, forming arimistane. This not occurs with aprotic solvents as DMSO. Approaches based on enzymatic hydrolysis using β-glucuronidase (E.coli) or glucuronidase/arylsulfatase (H. pomatia) did not favor the formation of arimistane. Nevertheless, the hydrolysis under strongly acidic conditions favors the complete degradation of 7-keto-DHEA into arimistane.

The correct detection of 7-keto-DHEA by using a suitable analytical procedure could avoid the risk of false positive findings for aromatase inhibitor arimistane and, at the same time could avoid the unnecessary application of IRMS since specific metabolites could be found.