www.cyclingnews.com - the world centre of cycling

Special news feature, September 23, 2005

Serious concerns over urinary EPO test

Three recent cases in the endurance sport of triathlon have raised some major concerns about the reliability of the urinary EPO test, which has been used in the sport of cycling since 2001. Although researchers around the world are working on improving the test, several holes in its past methodology have thrown doubt on every positive case involving the drug. This places the World Anti-Doping Agency and national sports federations under potentially crippling legal pressure, should wronged athletes sue for damages. But WADA believes it is safe from such challenges, as Cyclingnews' Chief Online Editor Dr Jeff Jones reports.

The key points, based on Peltre and Thormann's recommendations published in March, 2003

False EPO positives can arise from proteinuria, if an athlete produces certain types of proteins after intense exercise that the test picks up.
The antibody used is not specific enough, meaning that it will identify non-EPO proteins as EPO.
Different labs can give different results for the same sample, sometimes calling a negative a positive, and vice versa.
There can be a large variability in results, even when the same lab does the testing
WADA changed the test criteria from quantitative to qualitative wihtout any scientific validation.
A new "two-dimensional" EPO test is undergoing validation studies at the moment to iron out the flaws of the first one.
WADA doesn't believe that it can be sued by athletes who wrongly tested positive. It maintains that the test has given valid results in the past.

Science is characterised by debate. If a theory or method is found to be inaccurate or false, it has to be modified or thrown away. This means that things we have believed to be true in the past, such as the sun and planets revolving around the earth, have had to be revised when better information comes to light. Anti-doping science is not exempt from this, with all the talk about retrospective testing of samples, as well as questioning existing positives.

The Lance Armstrong EPO-in-a-1999-B-sample affair, as well as the acquittals of triathletes Rutger Beke, Virginia Berasategui and Ibán Rodríguez for EPO abuse, have put the spotlight on the Recombinant EPO test, which up until now has been believed to have been quite robust, even if comparatively ineffectual given its maximum three day testing window. Dr Iñigo Mujika is one scientist who has taken a keen interest in the test, and has had plenty of experience. He is a Sports Physiologist from the Basque Country, who has spent time working as a Senior Physiologist at the Australian Institute of Sport in the lead up to the Athens Olympics, and now works as a physiologist and trainer for Euskaltel-Euskadi, as well as a coach of one of the triathletes mentioned, Virginia Berasategui. He has written a detailed report on the limitations of the test, which he forwarded to Cyclingnews.

A summary of Dr Mujika's report (Click here for the full report)

The urine test currently used to detect rEPO was developed by the French Anti-Doping Laboratory in Châtenay-Malabry, and combines isoelectrical focussing with a double immunal blotting. This makes it possible to distinguish between artificial rEPO and human EPO, which behave differently in an electric field. The test was introduced at the Sydney Olympics in 2000, after a study was published in Nature outlining the method. Since then, the lab has published two more papers in relation to the EPO test.

In early 2002, WADA determined that the EPO test should undergo an independent review, and appointed electrophoresis experts Dr. Gabriel Peltre (ESPCI, LECA, Paris, France) and Prof. Dr. Wolfgang Thormann (Department of Clinical Pharmacology of the University of Bern, Switzerland) to examine it in detail. The pair submitted their report to WADA a year later, in March 2003, and proposed a number of refinements and improvements to the test. There were seven main points, ranging from the assessment of the urine samples prior to analysis to a new approach for the interpretation of the scanned EPO profiles, as well as the use of more appropriate testing materials.

Regarding the assessment of urine samples, Drs. Peltre and Thormann recommended that an EPO profile should be obtained for each athlete - similar to the blood profiling that is currently being employed by the UCI - corresponding to different physical activities including effort and rest. Any abnormal deviation from this basal profile should be investigated and explained.

False positives influenced by proteinuria

"If people have a pronounced post-exercise proteinuria, one lab interprets the result as a positive test ...but another laboratory considers the test as negative. "

- Prof. Bogaerts on one of the EPO test's shorctomings

There appears to be little knowledge about the EPO isoforms that are measured by the test. The two experts recommended that research on their function, catabolism, half-life, exact pI and age, gender and genetic influences, should be undertaken. For example, there are now at least three cases of triathletes naturally producing EPO-like proteins after after intense exercise - something that has never been taken into account until now, and the test has already been used to convict dozens of athletes since 2000. Therefore, the test appears to have an unknown false positive rate, and any athlete who feels they have wrongly tested positive for EPO in the last four years should be able to prove their innocence and sue - someone - for damages.

The expertise of the labs also comes into play here, as some have identified positives where none exist. A high protein concentration in the urine sample can be one of the flags for a false positive, and in such cases, some labs (particularly those of Châtenay-Malabry in the outskirts of Paris, Oslo and Barcelona) try to eliminate the non-EPO proteins before analysis.

Prof. Bogaerts was appointed by the Disciplinary Council of the Flemish Community in Belgium to examine the case of Rutger Beke's false positive. In an interview with Sporza TV on August 2, 2005, Prof. Bogaerts declared, "If people have a pronounced post-exercise proteinuria, one lab interprets the result as a positive test, such as those of Ghent and Cologne, but another laboratory considers the test as negative. I am certain that they have acted honestly, but the EPO test in itself is not reliable."

A WADA financed study was carried out in 2002 by the Australian Sports Drug Testing Laboratory (R. Kazlauskas, G. J. Trout, C. Howe, K. Nelson, J. Trotman), entitled "The effects of factors such as exercise and disease on the distribution of urinary erythropoietin isoforms." It determined that the concentration of EPO in urine is highly variable, and for some individuals can vary by more than a factor of four from one collection to the next. Also, that extreme long duration exercise can produce a small but significant increase in the percent basic isoforms found in the urine. But in the case of the samples coming from diseased subjects (anaemia resulting from severe kidney disease), the test was unsuitable because of the high protein content in the samples.

Better antibodies

"We have identified several urinary proteins with which the monoclonal EPO antibody used in the current test has non-specific binding."

- Australian research group Proteome Systems has studied the major flaw in the test

Besides exercise induced proteinuria, the second, and probably the biggest worry with the EPO test, is the use of an antibody that is non-specific for rEPO identification, i.e. it can identify other proteins in the sample. This is termed cross reactivity, and was noted in Peltre and Thormann's report. "Antibodies to EPO, monoclonals or polyclonals, should be tested for their fine specificity and their potential capacity of distinguishing between endogenous and exogenous EPO," wrote Peltre and Thormann. The main antibody used in the test, called AE7A5 and produced by the laboratory R & D Systems, has not been designed to be used in anti-doping testing.

A number of research groups and individual scientists have pointed to this as the main flaw in the EPO test. Prof. Bogaerts said in his Sporza interview, "Laboratories in Leuven have shown that one of the reagents which is used in the test is not specific enough. This means that the antibody picks up other proteins." Thus, if a specific antibody had been used, it would have been practically impossible for Rutger Beke to have shown a positive result for EPO, even after intense exercise.

Similarly, a very recent study by Australian researchers (Khan, Grinyer, Truong, Breen, Packer. New Urinary EPO drug testing method using two-dimensional gel electrophoresis. Clin. Chim. Acta 358 (1-2): 119-130, Aug. 2005) states that, "We have identified several urinary proteins with which the monoclonal EPO antibody used in the current test has non-specific binding. The iso-electric points of these cross-reactive proteins overlap with HuEPO and rHuEPO." In other words, false positives could result from the antibody erroneously identifying natural EPO and other unrelated proteins in a urine sample. A more specific antibody does not seem to have been developed.

Qualitative or quantitative?

Up until now, the urinary EPO test has been called "qualitative", but is in fact quantitative, as there has to be a minimum percentage (80% ) of basic isoforms for a sample to be classed as positive. The Châtenay-Malabry laboratory in Paris is even more stringent, requiring 85% of the bands to be basic for a positive. It should be noted that 80% is quite high when considering normal urine: someone could take EPO and still pass the test if their basic bands percentage was 79%. Thus, the test can give rise to false negatives as well.

In January 2005, WADA recommended that the 80% basic bands criterion should no longer be used, and that a more qualitative system should be used:

1. In the basic area there must be at least 3 acceptable, consecutive bands assigned as 1, 2, 3 or 4 in the corresponding reference preparation.

2. The 2 most intense bands either measured by densitometry or assessed visually in the basic area must be consecutive and the most intense band must be 1, 2 or 3.

3. The two most intense bands in the basic area must be more intense than any other band in the endogenous area either measured by densitometry or assessed visually.

None of these criteria were subject to scientific review, but were unilaterally adopted by WADA, it seems. On September 5, the President of the Spanish National Anti-Doping Commission (NAC) sent a communication to the Disciplinary Committee of the Spanish Triathlon Federation, in which they were advised that the World Anti-Doping Agency (WADA) phoned the accredited laboratory in Madrid on August 31 to communicate new instructions to modify the evaluation criteria for detection of urinary EPO. The new criteria have not been published by WADA and are therefore not known.

Huge variability in results

"None of the recommendations of the independent experts has been put into practice to date."

- Why has WADA been sitting on this for so long?

Of the 10 urine samples collected and analysed for rEPO after Ironman Lanzarote on Saturday, May 21, 2005, two were declared "positive" by the accredited laboratory of Madrid - those of Virginia Berasategui and Ibán Rodríguez. In addition to that, one was "non detectable", three were "not qualifiable", and four more "negative". Taking the not qualifiable and undetectable samples into account, between 20 and 60 percent of the samples could have been positive for EPO, a potentially huge incidence.

However, the samples took nearly three days to arrive in Madrid, and no information has been provided about the transport and storage conditions. The pre-test protein concentration of the samples is unknown, and could have exceeded the reliability limits of the test, given the points made above. Finally, according to the results of the analyses carried out on the A sample (first analysis and confirmation analysis) and B sample of the two athletes who tested "positive", the following percentages of EPO in the basic area were recorded:

Athlete 1: 63.15%; 68.21%; 54.72%
Athlete 2: 50.40%; 57.77%; 67.13%

Neither of these samples is close to the "80% basic bands" criteria required until January 2005, and still recognised by the Lausanne Court of Arbitration for Sport. The variability of the results between samples A and B is 33.2% in the case of Athlete 2 and 13.3% in the case of Athlete 1, showing that the test used up until now is unreliable.

Why has it taken so long?

Despite the clear limitations of the urinary EPO test, which have been known since 2003, Dr Mujika writes that "none of the recommendations of the independent experts has been put into practice to date, and the only change that has been applied to the test since March 2003 (date of publication of the Peltre and Thormann report) until today is the addition to the analytical procedure of the so-called 'stability test', subsequent to the case of athlete Bernard Lagat, in which the African runner was initially accused of doping with rEPO after the analysis of the A sample, then acquitted when the B sample showed negative results. Interestingly, this modification has never been validated in the usual way of every scientific method, that is by publication in a renown scientific journal submitted to peer review. It is thus obvious that WADA is a closed organization that does not follow the international rules of scientific publication procedures."

Furthermore, the Peltre and Thormann report was removed from WADA's website in late 2004, soon after Belgian scientists announced their intention to look into Rutger Beke's false positive.

A new test

"I wouldn't say that the problems with the old test cast doubt on all positives under it."

- Proteome Systems' Nicolle Packer on the ramifications of a faulty EPO test

WADA or no, science progresses, and things are looking rosier for the future of urinary EPO testing. Labs around the world are working on improving the test, or even developing a new one. The Châtenay-Malabry lab has already been doing this, and its study of the 1999 Tour samples was part of its research, even if the results were somewhat surprising. But it was using the existing test and refining it with what appeared to be more sensitive analysis.

A biotech company in Sydney, Australia, called Proteome Systems, has developed an improved test, taking into account the limitations of the previous one. Project leader Nicolle Packer applied for a grant from the Australian government to develop a new test, as she could see that their expertise could be applied to improving the urinary test originally developed by Australian National Measurement Institute. As previously referenced, the group published its results under the title, "New Urinary EPO drug testing method using two-dimensional gel electrophoresis." Clin. Chim. Acta 358 (1-2): 119-130, Aug. 2005.

Packer explained to Cyclingnews that the current test is a "one-dimensional separation" for rEPO vs hEPO. The two proteins have identical amino acid chains, and differ only in the sugars that the cells producing them attach round the protein. The charge on the protein is different because the sugars are different, therefore if you separate on charge, you get different 'ladders' for artificial and natural forms of EPO.

"Several steps in the current test are difficult or inconclusive," said Packer. "You can imagine that with so many proteins in urine, the level of EPO you're trying to detect is very small. With our [initial] grant we did a proof of concept showing that the two-dimensional separation was more reliable. We also found that the antibody used in the current test will sometimes pick up other proteins."

Packer added that she thought it was only cases such as Rutger Beke's that could be problematic. "I wouldn't say that the problems with the old test cast doubt on all positives under it," she said.

Proteome Systems has made several improvements in sample preparation, detection and separation. The new test is currently undergoing validation trials at the National Measurement Institute in Australia and the German Anti-Doping Laboratory in Cologne. Packer believes that the new test will be as a validation test, and the old one-dimensional test will be used as a screening test. "If you can do two tests, it's better than one," she concluded.

WADA's response

"On what basis would you come after WADA?"

- WADA president Dick Pound believes his organisation is safe from athletes' legal action

In a telephone conference on September 15, WADA's Dick Pound dismissed suggestions that the EPO test is unreliable, or presented any kind of legal problem. "Taken from 10,000 feet, we're not concerned about the viability of the test at all," said Pound. "The test is quite reliable. What is interesting, and is something that we have to continue to follow, is whether or not there is some exercise induced variations that might occur, that might lead a lab without too much experience in these kinds of tests to conclude that there's a positive when maybe there is not.

"So what we've sent out to all of our laboratories is, if you get profiles that are somewhat different from the standard positive profiles, check it with one of the two or three labs that have the particular expertise in this. Go to the specialists before determining whether or not you're faced with a positive sample.

As far as athletes suing WADA for damages over false positives, Pound doesn't believe it will work. "I think if you've signed onto this and you're governed by the code, your recourse is to the Court of Arbitration for Sport," he said. "You don't go running off to the state courts."

Referring to Beke's case, where he intends to sue WADA for €125,000," Pound said, "This one so far, as far as I know, has been confined to the courts in a particular part of Belgium. We've had some contact from the lawyer of the athlete, whose name is Beke, using the shotgun approach. Maybe as a lawyer he went to school in the States, suing everyone in sight. We said 'Well, do whatever you have to do, but we don't have any responsibility'. And furthermore, we said 'On what basis would you come after WADA?' They refused to give us any documents, they refused to permit the government agencies to release any information to us, then at the same time they have this preposterous claim against us. It's out there, but it's not something over which we're losing any sleep at all."

Can an athlete sue any organisations for a loss in income? "I don't know the answer to that. I suspect if you're claiming, you would have to show there's been fault on the part of somebody. Scientific disagreement may or may not be fault. But certainly it's not the fault of an organisation like WADA, which approves a test that has significant scientific support form around the world. There may be some factual things, I suppose you might have the possibility that a lab does a really shoddy job as opposed to a matter of opinion, but that's getting pretty hypothetical."

Cyclingnews coverage of the L'Equipe allegations

June 27, 2006 - Carmichael defends Armstrong, Armstrong answers L'Equipe & LeMond
June 26, 2006 - LeMond: "Armstrong threatened my life"
June 19, 2006 - Armstrong calls for Pound's exit
June 18, 2006 - Lance Armstrong's open letter against Dick Pound
June 4, 2006 - UCI hits back at WADA
June 3, 2006 - WADA slams the Vrijman report
June 2, 2006 - L'Equipe stands by its story, UCI supports Vrijman's findings
June 1, 2006 - UCI, WADA and Armstrong react to Vrijman's report
May 31, 2006 - UCI lawyer asks for Armstrong's name to be cleared
May 14, 2006 - Two more weeks for Armstrong investigation

Click here for full coverage of the L'Equipe allegations.

More Cyclingnews features

Recently on Cyclingnews.com