Detection of recombinant EPO and innocent athletes

I belong to a group of four Norwegian scientists (Jon Nissen-Meyer, Tore Skotland, Bjarne Østerud and myself) who have been engaged, for about ten years, in evaluating the scientific methods employed to detect EPO and EPO-like substances in athletes. The methods used have changed and the scientific criteria for deciding a positive analysis have been variable, arbitrary and subject to some criticism. What worries us is that the scientific aspects of the evaluations have been neglected, at least in part, and the decisions appear to have been reached based on the assumptions that the anti-doping experts know what they are doing and therefore cannot, and should not, be challenged.

The technical method to detect EPO has always involved polyacrylamide gel electrophoresis. Before 2014, Isoelectric focusing (IEF) was used. This is a method where the EPO-like substances are separated based on their isoelectric point. The supplementary method is PAGE, with the detergent SDS or Sarkosyl, which separates proteins based on molecular weight. After 2014, IEF was considered less accurate and sensitive than PAGE, a change noted in the WADA technical document TD2014EPO, so from then on PAGE was the preferred method. The opposite characterizations were given before the switch, i.e. then IEF was the better method of the two. At present, the IEF is not much used, since PAGE is considered more sensitive and specific.

Two of us in the group (Skotland, Østerud) were involved in the CAS hearing against the Norwegian walker Erik Tysse. He was sanctioned based on IEF analyses, in spite of his SAR-PAGE analysis being negative. The analyses were of poor quality and the results were not reproducible and contradicted one another. The two of us presented evidence that, in our opinion, gave strong arguments against the presence of a doping agent. In hindsight, it is fair to say that Skotland and Østerud’s expertise was not appreciated and their expertise was overshadowed by the apparently authoritative statements from WADA’s experts.

Later, the four of us have been engaged in the EPO-allegations against the Irish sprinter Stephen Colvert, the German middle-distance runner Benedikt Karus, the Czech triathlete Vojtech Sommer, the Portuguese cyclist Andre Cardoso and, this year, the Norwegian boxer Hadi Srour. In the cases of Colvert, Sommer and Srour we have, in particular, spent some time in analyzing the data from the different antidoping laboratories, using accepted, well-known quantitative methods. In our opinion, the experiments have been carried out relatively well, but we have severe problems in accepting the interpretations. It is obvious that the WADA-accredited labs have problems in interpreting and following the criteria for determining whether the PAGE analyses are positive or negative. The main problem is that the technical document (TD2014EPO) does not give clear indications of what exactly these criteria are. In fact, the decision is left to the antidoping experts, who can call a positive if they believe that they see a smear of mixed bands in the PAGE gel extending above the band for normal, human EPO. This is the region where recombinant EPO can be expected to appear. Our argument, as expressed in our expert statement in the hearing for Hadi Srour, is that any decision requires a quantitative, objective measure of the level of gel staining in the region, i.e. how much EPO-like material is present in the band representing normal, physiological EPO and how much in the region above (rEPO)? To leave this decision to an arbitrary visual inspection of the gels is unscientific. Results from the Norwegian antidoping laboratory and from our own analyses clearly demonstrate that there is no more material in the rEPO-region in the athlete’s urine than in a control sample without rEPO, a negative control (one example shown in the Figure below).