Researchers led by Dr. Carsten Lundby at the Copehagen Muscle Research Center have just published a paper on testing for EPO via urine sample. Lundby and company administered recombinant EPO to eight male subjects to track the performance-enhancing effects of the drug. While the results of this study are plenty revealing in their own right, in the course of performing this study the researchers decided to send the samples to two laboratories to see if the subjects would test positive using the current test protocol approved by the World Anti-Doping Association (WADA). The results of this study were just published by the Journal of Applied Physiology, a peer-reviewed journal.

The seven-week study began with a two week “boosting” period in which subjects received high doses of EPO, followed by a two week “maintenance” period of reduced EPO administration and ending with a three week “post” period. During this time some 32 urine samples were sent to each lab. In the paper the labs aren’t identified; they are just labeled “A” and “B.” Lab A found all samples from the boosting period positive. However, it only found two of 24 maintenance-phase samples positive.

Much worse was the record of the lab labeled “B” in the study. It didn’t identify a single sample as positive, and only determined only seven of the eight boosting period samples as suspicious; no other samples raised proverbial red flags for the lab. Lab B has since been identified as the Laboratory for Doping Analysis (LDA) in Cologne, Germany. The LDA is a WADA-accredited lab. It’s director, Wilhelm Schanzer said the study’s finding that the lab could not accurately identify recombinant EPO is “outright false.” He went on to say, “It’s not true that you could take EPO and not be detected.”

Schanzer went on to claim that because the samples were for a research project his lab didn’t perform all the tests necessary to verify a positive finding. He said under normal testing procedures his lab would have detected EPO accurately. He makes a great claim, but there is a problem with it: all of the samples were submitted blind. LDA weren’t told anything about the samples that might tip off the lab’s staff.

Perhaps more chillingly: One sample taken during the post period when theoretically no EPO should have been detectable was deemed positive by one of the labs.

BKW spoke to Paul Strauss of the Agency for Cycling Ethics to get some perspective on the issue. He began by saying, “WADA needs to look at this very seriously.”

The first observation he made was to note that the markers used to distinguish EPO depend on its production. He says the test WADA uses is optimized to find Amgen-produced EPO, while EPO produced in Mexico or China, and recombinant EPO can all escape detection if lab technicians only look for Amgen EPO.

Strauss also said that the criteria for a positive result can and do vary from one lab to another.

Ultimately, it may be that in the short term the best way to deliver clean riders to the start of races will come through programs such as Strauss’ Agency for Cycling Ethics, Paul Scott’s Scott Analytics or Danish physician Rasmus Damsgaard’s testing program.

“Longitudinal analysis which uses statistics to compare individuals and populations of athletes is very effective in raising a suspicion of doping in a particular athlete,” Strauss said.

However, even longitudinal testing has its own drawbacks. Strauss continued, “Its weakness is that it is not specific as to the exact drug being used. This leads to the problem of pursuing a sanctionable event for doping on a non-analytical adverse finding.”

It may be a foregone conclusion that catching athletes using performance enhancing drugs will remain an imperfect science. The question that remains: What we are willing to accept as the margin of error—the innocent or the guilty? What is the greater injustice: To allow some cheaters to escape detection and gain wins that shouldn’t rightfully be theirs, or to wrongly convict the occasional athlete who didn’t break the rules?

If we look to legal systems for parallel, this is where the United States and some European countries differ significantly. The American view of justice is that no innocent person should be convicted (in theory, if not in actual practice), while many countries, such as France with its Napoleonic Code, would rather scoop up a few innocents along with all the guilty. This characterization paints with a broad brush, but it seems a helpful way to frame what ought to be a conversation for how drug testing should be considered.

Even if Landis had succeeded in his defense, the result would hardly have been as damning as this study which was funded in part by the Danish anti-doping agency. The message is simple: Use recombinant EPO and finish your boost phase before the Tour starts; we won’t catch you.