The Athlete Biological Passport: a ‘magic bullet’ for EPO detection? Part 1 of 2

Published 06 February 2013 By: Emma Mason

The Athlete Biological Passport: a ‘magic bullet’ for EPO detection? Part 1 of 2

Twenty-twelve was the year sport dominated the news. Fans were spoilt for choice in a record-breaking year including the Tour De France, Euro 2012, The Ryder Cup and the London Olympics. However, 2012 also saw the fall from grace of a sporting hero, a man who transcended sport and whose achievements were revered by sporting fans and the general public alike: American cyclist Lance Armstrong.

Rumours of drug doping plagued Armstrong from the first of his (now expunged) seven Tour de France victories but he was defiant, repeatedly and aggressively denying the allegations. However, on the 10th October 2012 the United States Anti-Doping Agency (USADA) published their reasoned decision’, a catalogue of witness testimony and documentary evidence claiming Armstrong had been involved in a highly sophisticated and prolific drug taking programme. The sporting world waited just over three months to hear Armstrong confess in the highly publicised Oprah Winfrey interview that he had used performance enhancing drugs during each of his seven victories.

Amongst the drugs Armstrong admitted using was Erythropoietin or EPO. In an era of highly sophisticated scientific techniques a reasonable person may ask how it is possible that detection of EPO continues to be so problematic. This article will address the inherent difficulties associated with testing for EPO, the reasons the current testing method has attracted criticism and what must be learned to ensure the newly proposed Athlete Biological Passport (ABP) is a more successful, effective and trusted procedure.


EPO Background

EPO is a hormone naturally generated in adult kidneys and is the body’s principal regulator of the biological process responsible for the amount of circulatory red blood cells. Thus, an increase of EPO present in the body will result in an overall rise of red blood cells, the amount of oxygen available to the muscle and lead to a distinct boost in aerobic performance. EPO was first marketed in 1988 as a therapeutic treatment for patients suffering from chronic renal problems. The anti-doping community, aware of the performance enhancing potential, declared it a prohibited substance in 1989. However, no harmonised, effective analytical detection method was confirmed until 2000. During the interim period abuse of the drug was reported to be rife1 .


EPO Testing Background

Developing an effective test for EPO was a great challenge. The drug itself is specifically designed to have as few differences between the natural and synthesised form as possible. Testing is further hampered by the relatively short half-life of EPO in bodily fluids making detection difficult without prior knowledge of exactly when the drug was administered. Inherently low concentration of the drug in bodily fluids or wastes meant that detection via traditional mass spectroscopic routes, used to great effect for detection of steroids and stimulants, was not feasible.  

Initially, several indirect methods were piloted for detecting EPO abuse. For example, the ‘No Start” rule introduced by the Union Cycliste Internationale (‘UCI’) in 1997 tried to monitor blood parameter levels to deter EPO abuse. The test was conducted on race days and any cyclist showing a haematocrit level (the proportion of red blood cells to total blood volume) over 50% was instantly disqualified. However, there were a number of problems with this method. For example, natural variation in basal haematocrit levels meant that athletes with an inherently low level could administer EPO without ever being caught. Moreover, haematocrit levels are extremely sensitive to blood volume levels and athletes could easily disguise their doping activities through simple dilution of their blood.


Current EPO Testing Method: The EPO Test

A breakthrough in detection occurred when it was discovered that the naturally occurring and synthetic forms of the hormone have a slight difference in overall charge. The forms can thus be separated through an analytical technique known as isoelectric focussing. The test is able to directly detect the presence of EPO in urine as the difference in molecular charge separates the natural and synthetic forms giving rise to a profile of identifiable, characteristic bands on the laboratory membrane. An athlete’s band profile can then be compared to that of synthetic EPO standards and if the comparison complies with World Anti Doping Agency (‘WADA’) identification criteria then a positive test result is reported. The development of the analytical procedure was announced in June 2000 and first trialled at the Sydney Olympic Games.

The EPO Test is still in use today and was part of the anti-doping testing battery employed at the London Olympics 2. The test has provided the anti-doping community with a significantly improved weapon to detect use of the drug. Nonetheless, the test did not have a smooth introduction and several issues outlined below have led to its scientific and legal credibility being called into question.


Problems Arising

The problems with the current test fall into three broad categories: 

  1. Actions of the anti-doping community
  2. Inherent difficulties of detecting the drug
  3. Actions of the athletes

 It should be noted that only the first problem is directly related to the current EPO test. Points two and three are characteristic difficulties facing any method and will therefore continue to be a challenge for detection through the new ABP.


1. Actions of the anti-doping community

In 2001, the first positive EPO test was declared by the Lausanne WADA laboratory and concerned the urine sample given by Danish cyclist Bo Hamburger. The cyclist successfully appealed the decision to the sporting arbitral tribunal the Court of Arbitration for Sport (CAS) due to laboratory failure to apply uniform identification criteria to the cyclist’s A and B samples. At that time, identification criteria required 80% of the bands to be in the basic region 3 . While the A sample was found to be clearly above this level, the B sample had 78.6% of bands in the basic region. Nonetheless, the Lausanne laboratory had chosen to declare a positive sample as there was no scientific justification for an 80% threshold and that a level marginally below 80% was reliable enough to allow one to assume a positive result” 4. CAS rejected this argument on the basis that this constituted a failure to adhere to criteria set by their own laboratory. In addition, the court highlighted that threshold levels were not set at 80% for all WADA laboratories. For example, the laboratory in Paris set the limit of percentage basic isoforms to be at 85%. The incongruities between both the threshold limits set by WADA laboratories and the incorrect, inconsistent application of those criteria to athlete samples led the case to be dismissed. The decision caused the EPO test, WADA and the anti-doping practitioners to lose scientific and legal credibility.

An independent review of the test in 2003 5 helped to highlight areas for improvement and has engaged the scientific community in gradual enhancement of the test’s sensitivity and specificity. Nonetheless, errors in procedure and analysis of the EPO test have continued to plague the anti-doping community and damage the test’s reputation. For example, sample degradation caused band profiles of an innocent athlete to match that of one who had doped 6. Furthermore, atypical profiles generated through strenuous exercise 7 and inconsistent analysis of the same samples by different laboratories 8 could create false positives. In light of the above it is not difficult to see why confidence in the test has been shaken.


2. Inherent Difficulties

Detection Window

As mentioned above, one of the fundamental problems with testing for EPO is the short half-life of the drug. This reduces the time window within which the drug or its metabolites can be detected in bodily waste. It is generally accepted that first generation EPO has a half-life of 3 days and that after this time the chance of detection is significantly reduced 9. Moreover, research has shown that micro dosing could reduce the window further to only 12-18 hours post injection 10 . Longer detection windows of up to seven days 11 have been seen in the second and third generation EPO drugs, NESP and CERA respectively. However, even the increased chance of detection during the longer window has not proved a successful deterrent 12.

It must also be remembered that the modern athlete is well versed in anti-doping testing regimes. The short detection window allowed the development of doping programmes such as Armstrong’s that ensure maximum physical benefit out of competition while minimal chance of the drug metabolites remaining in the system when competition commences.


Drug Development

While the development of EPO has plagued the anti-doping community it is important to remember the wider picture: the drug was introduced as and continues to be used as an important therapeutic treatment. The reliability of the EPO test is also affected by development of medicinal drugs with the same physiological effect but contrary isoelectric banding pattern. The EPO test utilises a comparison to specific, known banding patterns and thus a different drug with a different banding pattern could lead to difficulty in the drug’s detection. For example, biosimilars are drugs produced following the expiration of the original patent but owing to dissimilar techniques used by pharmaceutical companies they are not chemically identical to the original drug and thus require the test criteria to be adapted 13 . Furthermore, regulation of such drugs is less stringent in emerging markets 14 and consequently athletes, who are not restricted to one supplier, have access to a growing range of EPO biosimilars. Further medical advancements such as second and third generation EPO have forced adaptation of the identification criteria to ensure detection. Over the years the EPO test criteria have been adapted to detect the varying EPO drugs but against an influx of drugs this has often been referred to as a game of ‘cat and mouse’: until the anti-doping community is aware that a certain drug is being abused it is unable to adapt the scientific technique.


3. Actions of the Athletes

There is no doubt that the athletes themselves have devised original and effective ways to deceive the test. These may be practical avoidances such as adapting drug administration programmes to ensure metabolites are out of the system by competition, or they may be scientific. The use of micro-dosing leads to a detection window so short that it makes detection near impossible. Furthermore, athletes have been reported to use anticoagulants to conceal EPO abuse. One anticoagulant, Heparin, was found to destroy the banding profile making analysis of the sample impossible 15 . This required the anti-doping scientists to adapt the test and carry out an additional, selective purification stage prior to sample analysis.


An Alternative approach

Despite the efforts of the anti-doping community, there are still problems that have tainted the EPO test’s reputation, perhaps permanently. The test’s development came at a time when the anti-doping community, overwhelmed by the prevalence of EPO abuse, was desperate for a solution. More precisely, it was looking for a classical direct detection method for a very advanced, modern form of doping. As a result, it appears the test was introduced with inadequate assessment and inevitably modifications and improvements have had to be made publicly in response to failures. As it stands, the current version of the EPO test is the most sophisticated, sensitive and specific available. However, following years of controversy and the revelations in USADA’s ‘reasoned decision’ of continual, undetected EPO abuse from the self proclaimed ‘most tested athlete ever’16, it is probably fair to say that confidence in the test’s fairness and effectiveness has never been lower.

As such, there is a strong school of thought that it is time to look at other methods of detection for EPO. There has been extensive research into the possibility of using various direct methods such as mass spectroscopy utilised to great effect in the detection of steroids and stimulants. Unfortunately, these are still reliant on detection of the expelled metabolite in urinary waste where the concentrations of the EPO protein are intrinsically low rendering the sensitivity inadequate 17. Over the past ten years, however, there has been a dramatic shift in the attitude of the anti-doping community toward indirect methods of doping detection. This has led to development of the Athlete Biological Passport (ABP) that purports to identify use of performance enhancing drugs through longitudinal testing of the athlete’s blood parameters. Traditionally, the anti-doping community has favoured a quantitative approach to testing that provides a single piece of scientific evidence on which to convict a doping athlete; in the case of EPO, the banding profile. Whilst this method of prosecution has long been deemed acceptable procedure by the anti-doping regulators there has been growing concern within the wider scientific community that this is unfair. The accused athlete is left unable to pursue any line of defence other than that the Adverse Analytical Finding (‘AFF’) was caused by procedural errors 18. The athlete is not able to question the scientific credibility of the test, which, when taking into consideration the potential for false positives, is unjust. Furthermore, it does nothing to improve the scientific or legal credibility of anti-doping science.  

There have been suggestions that anti-doping should look to align its methods with those employed by forensic science when convicting criminals based on DNA evidence. In forensic science, a conviction cannot be made based solely on a DNA match. Instead, the evidence must be accompanied by statistical estimates of random match probability (a measure of the likelihood that the DNA match is a random coincidence) and of the probability of false positives. Guilt or innocence of the accused is then decided on by judge or jury based on DNA evidence in conjunction with a probability ratio 19. DNA identification is achieved through empirical testing and a statistical analysis of population genetics.

In the part 2 of this article the author will the look into detail at the Athlete Biological Passport procedure.



Related Articles


Emma Mason

Emma Mason

Emma is a trainee solicitor in Squire Patton Boggs’ sports litigation department who has completed seats in corporate, international dispute resolution and a secondment to Chelsea Football Club. During her traineeship Emma has, from a sporting perspective, assisted with the sale of a Championship football club and the provision of advice to various International Federations and Premier League football clubs.