Abstract
The use of performance-enhancing drugs has been a problem throughout the history of competitive sports. In Olympic sports doping was officially banned in 1967 by the IOC (International Olympics Committee) Medical Commission. The World Anti-Doping Agency (WADA) was established in 1999; it is an international independent agency with authority from the IOC to carry out doping controls, its aim being to create a doping-free culture in sport. It monitors the World Anti-Doping Code, a document which seeks to harmonise the antidoping policies in all sports and across all countries. 1 The rules of the code are applied by the Court of Arbitration for Sport (an international arbitration body set up to settle disputes related to sport) and other internal bodies of national antidoping organisations. Most international federations of Olympic sports, national Olympic committees and other sports organisations had implemented the code even prior to its introduction in 2004.
The monitoring of compliance to this code requires much work: scientific, legal and administrative. However, research into the effectiveness of current monitoring arrangements shows that there is much work yet to be done. Alaranta et al 2 questioned 446 athletes anonymously and found that 30% said they personally know at least one other athlete who takes performance-enhancing drugs banned by WADA. A major obstacle for WADA to overcome is that many drugs, especially anabolic-androgenic steroids (AAS), are used out of competition in the training phase before entering competitive events. Out of competition testing is expensive to conduct as many athletes travel extensively between competitions and to training destinations. Testing in-competition for substances administered out-of competition, however, will probably not detect these substances as they may have short detection windows when measuring them in the usual samples (predominantly urine samples). The problem is the difficulty of reliably testing whether banned substances have been used in the training period before an event; such detection is currently more or less “time dependent”, depending on the substances being tested for.
Considering this issue, keratinaceous samples (e.g. hair) seem to offer an advantage, as detection windows in such samples have been shown to be far longer for most drugs than urine samples. Drugs are taken up into hair at the hair root and into nail matrices at the proximal nail fold and at the nail bed. 3 Their location in the sample is useful to estimate the dates of administration. The average rate of growth of both hair and nails and the factors affecting this rate have been well documented, therefore if a hair or nail sample is divided into narrow sections the distance of the substance from the hair root or the proximal nail fold divided by the average rate of growth gives an estimate of the dates of administration. This is more complex in nails since a proportion of the drug is also taken up at the nail bed; however, peak concentrations within the nail sections could still be feasibly used for estimation of the time lapse since administration. Varying exposure to drugs resulting in varying concentrations of drugs extracted from the nail plate over periods of months has been previously documented. 4 In hair, the mechanism of uptake and drug binding in the hair shaft is well researched and analysis of hair samples for illicit drug use is admissible in a court of law. For nails the mechanism of uptake and drug binding in the matrix is less well characterised, and requires further research if nail clippings were to be analysed for illicit drugs. Hair is the keratinaceous material of choice for many drug detection laboratories, as its length generally gives a long detection window, and research into analytical methodologies and the interpretation of results is well documented as well as the legal admissibility. However, in an area such as doping in sport, hair testing would be easily evaded by shaving the head, and an added complication is that a side-effect of AAS, drugs commonly used out of competition, is hair loss. Nails are also non-intrusive and easily extractable keratinaceous samples. A literature search revealed that testosterone, an endogenous steroid, has previously been extracted and quantified in human nail clippings. 5
A study was designed with the aim to develop a method that would be efficient and sensitive enough to be able to extract and detect AAS that are abused in the practice of doping in sport from nail clipping samples. For an initial study, three AAS with widely ranging chemical properties were chosen to be measured: namely testosterone, testosterone propionate and stanozolol (from Sigma-Aldrich, Poole, UK). Ethics approval was given from the Queen Mary University of London Research Ethics Committee. Samples were collected from male volunteers, who had been taking AAS over the last six months, as well as comparative samples from volunteers without any history of AAS use. An extraction method and a high-performance liquid chromatographic (HPLC) method were developed, analysed for performance and efficiency, and subsequently LC-ESI-ion-trap-MS (HPLC coupled to an electrospray ionisation mass spectrometry detector) was used to analyse some samples.
The extraction method yielded reasonable recoveries for all three analytes; however, analysis of the recoveries was complicated by lack of precision of the assay due to the concentration range investigated being close to the lower limits of quantitation for the analytes with ultra violet (UV) detection. UV detection was found to be unsuitable for the application of measuring AAS in nails. The LC method was optimised, and limit of detection, lower limit of quantification, repeatability, reproducibility, selectivity and robustness of the method were determined. The LC method was then transferred to a liquid chromatographic-mass spectrometry (LC-MS) system, and the volunteer samples were analysed.
Results from the LC-MS analysis showed that the sensitivity of this method unfortunately did not meet acceptable standards for quantifying the concentrations of the three AAS in the samples. Testosterone was detected in all the samples analysed, both from volunteers with a positive history of AAS use and from participants with no history of AAS use. This is to be expected as testosterone is an endogenous steroid; however, the method lacked the sensitivity to demonstrate any real difference between those with a history of testosterone or testosterone pro-hormone use and the controls, or between males and females. Testosterone propionate was not detected in any of the samples, including a negative result from one volunteer who had been using testosterone propionate for six months prior to sample collection. This could be due to extraction losses at the low concentration expected; testosterone esters are more easily broken up under the conditions used in the method than the other two analytes. For the third analyte, the protonated ion of stanozolol was shown to be present at a level distinguishable from the baseline in one sample from a volunteer, who admitted to taking stanozolol regularly over the prior six months (the only volunteer with a positive history of stanozolol use). This peak was not present in any of the control samples or any samples with a negative history of stanozolol use.
The positive sample analysed was from a participant with a history of high, continuous doses. It is unlikely that this method would detect an acute administration of AAS, the concentration of stanozolol in this sample is around 700 pg/mL of extract and for a single dose of this substance a concentration of approximately 100 times lower than this could be expected, which would be below the limit of detection using this method, although more sensitive methods may be developed. No articles concerning hair analysis for AAS could be found claiming to detect single dose administration. This limitation of the method, however, is not a problem for applicability to antidoping testing as AAS and most other banned performance-enhancing drugs do not have especially high efficacies, so their use by competitors is usually continuous dose or by cycling during training periods rather than by single dose administration, although obviously athletes use them as sparingly as possible when attempting to evade detection.
The methods investigated in this preliminary study did not have acceptable sensitivity levels required for the determination of the concentrations expected in the sample type (although they did meet the minimum WADA standard detection capabilities of 2 ng/mL). However, the ESI-ion-trap MS detection did show qualitatively the presence of stanozolol and testosterone although at unquantifiable concentrations. These results provide evidence that AAS, both endogenous and exogenous, do get taken up into the nail plate after administration. This evidence demonstrates that nail clipping analysis warrants further investigation. The method developed in this study can be refined, in terms both of extraction and detection. More sensitive mass spectrometers are becoming available and with improved sample clean-up steps it is expected that AAS could be quantified in such samples.
With such improvements, nail samples could be a useful additional sample type to combat doping in sport. With a longer detection window covering out of competition periods and nail samples being difficult to adulterate they offer considerable advantages in policing antidoping policies.