Saturday, August 05, 2006

Sad End for Landis?

So the B sample is positive too. Doesn't look like Landis has much of a chance to clear his name. I don't know much about doping and cycling, but somehow all of it seems quite shady, including the people in the anti-doping agencies. Something is rather strange with the way Landis allegedly doped as well. First, what does synthetic testosterone do anyway? The joke I heard is that it only helps if there's a hot chick at the top of the mountain. :) Delphine, any thoughts on this, since you are our resident expert? What would actually happen to someone if their testosterone level is 3 times higher than the natural limit in the human body? Would there be any outwardly observable signs? If there should be, then maybe people should be taking another look at the tapes from stage 17.

The other thing is, I thought that one of the arguments used by people who claim Armstrong is a dopee despite never failing a drug test is that there are dozens of banned substances that cannot be detected by any test. If that's the case, why would Landis choose testosterone, and the synthetic stuff at that, when it's almost certain that he'd get caught? It's not like he doesn't know he'll be tested right after the stage when he won. Landis is either incredibly stupid (unlikely), or desperate (possible), or there's something else going on. And let's not forget that he's had a reputation for being a very clean rider. The case against him just seems too neat.

This brings me to the shadiness of the anti-doping guys. In the recent case where Armstrong's sample from 1997 (or there abouts) was tested positive for EPO, the court actually found the anti-doping agency and the lab that performed the test grossly negligent in their handling of the samples. It was unclear who had access to it and for how long. That very same lab was used to test Landis's A sample. My question is, who's in charge of handling the samples and can they be trusted? There seems almost an irrepressible giddiness in the way the positive test was leaked to the press. In a way, the more doping there is and the greater the magnitude of the bust, the more important and powerful the anti-doping guys become. In my book, that's a conflict of interest and it taints every result that they produce.

What do you guys think?

22 Comments:

Blogger finou said...

ok so I make no pretense of being an expert on these maters at all but here's my take on it.

Elevated levels of testoterone in one day would most likely just increase endurance a whole lot. Over a long period of time, it's an anabolic steroid so it would help build muscle mass (but I don't think that's what helped Landis since it was only a 1 day spike) Also, just to be nitpicky, there is no "limit" to the levels of testosterone, there's just normal human "levels" of testosterone.

I think the case against Landis pretty strong. He may not have known that he got fed something (that's happened before) though. On why he could have done this and think he could get away with it: the synthetic testosterone test they did is fairly new and a lot more accurate than old methods and also I would also assume that people aren't very good at estimating dosages that can show up on newer (less used) drug test and maybe don't think that a certain level will show up. The Landis case is very different from the Armstrong case. For Landis, it was an official drug test in an international standard testing facility (little room for error with little time lag between the collecting and testing of samples).

For Armstong, this was in a research lab and someone "let slip" the results on some badly labeled tube. The slip to the press was gross incompetence/negligence/unethical behavior. But about the quality of the result being bad, I work in a research lab with frozen samples and I can see how mix ups with old tubes can happen. This is also why I don't believe those results when only 1 of Armstrong's samples come back positive (That's why you have multiple samples to make sure everything is confirmed and you don't mix samples). The research facility and the official testing setups are very different in my opinion. Part of the point of that research lab looking at old samples, btw, was to test whether people use substances they know at the time can't be traced. Regardless of the whole Armstrong fiasco, they found that the answer is big definitive yes. Lots of the old samples from many riders had EPO and that's kind of sad.

I know some atheletes are knowingly using drugs but I want to know how much coaches and managers are also to blame. There's a pretty famous case of an East German women's swimming champion, for example, who sued the country of Germany (and won) when she (actually he, after sex change) figured out that all her (his) health problems were due to the drugs she had been unknowningly slipped when she was young...

8/05/2006 02:22:00 PM  
Blogger Vincent said...

So now one expert (finou) has weighted in on this site. If only we had a Mennonite cyclist expert as well....

8/05/2006 05:48:00 PM  
Blogger Eric said...

But if we're willing to entertain the idea that one sample could have been contaminated, then it seems to me that both could have been contaminated... Especially, if it was deliberate. Given the stories of crime labs being irresponsible and even corrupt, I think this is a reasonable question.

8/05/2006 09:15:00 PM  
Blogger Qian said...

Thanks for the explanation, Delphine. I think it's plausible that a coach or trainer slipped Landis the testosterone. In a sport as apparently riddled with doping as cycling, it's remarkable how clean everyone generally is during the actual tour. Everyone is probably just more careful. However, I can imagine after Landis bonked in stage 16 one of the coaches deciding that he's got nothing to lose anyway and giving him the steroid just so he can make up some time and still earn a respectable finish. But it worked too well and Landis actually won the stage, thereby guaranteeing the doping test. I find it hard to imagine that if Landis did the doping himself that he would take that kind of risk.

On a related tangent, why has it been so hard to control doping in cycling? I remember a big doping scandal in swimming a few years back when just about the whole Chinese women's national team was banned because the coach was caught with steroids in his suitcase. Since then, the sport has been pretty clean (and the Chinese women have been far less competitive). In cycling, is it because everybody is doing it or is it because there's too much money involved?

Oh, on a different related tangent, there's been talk of testing for steroid use in golf, which is just the most asinine thing in the world. What's next, doping tests for billiards and horse shoes?

8/05/2006 09:20:00 PM  
Blogger Eric said...

From the story link...

'"It goes without saying that for us Floyd Landis is no longer the winner of the 2006 Tour de France," race director Christian Prudhomme told The Associated Press.... "We can't imagine a different outcome," he said.'

It's statements like that that make me suspicous of the authorities. It seems to me that the race director should be very familiar with the official policies and proceedure for resolving such claims. For the race director to publically state that he has reached a conclusion before that takes place, suggests to me a significant bias at best.

Is the testing done in France? Or is it sent elsewhere? Sorry, finou, but I can't help but wonder if there could be an element of French anti-Americanism.

8/05/2006 09:21:00 PM  
Blogger Ben said...

The whole case just seems odd. If I recall correctly, someone from Landis' side (a bit biased) was claiming that steroids take months to make an effect, and the only short-term benefit might be a slight increase in recovery time. But why use a method that's so sure to be detected?

I was reading a comment on Tyler Hamilton's blog (currently serving a suspension for doping) that was rather interesting. Most of these methods are indectable since they use the body's own hormones. Ulrich and Basso who were cut in operation Puerto before the tour, presumably wouldn't have been caught using any of these drug tests.

You can't help but feel sorry for Landis, as he stands to loose millions from this. Whether he's innocent, doped unknowingly, or made a rash decision, we'll probably never know.

8/07/2006 06:53:00 AM  
Blogger Qian said...

That's another problem I have with this whole anti-doping thing. There's really no way to prove one's innocence since the accusation alone can taint an athlete for life. In Landis's case, it's going to not only destroy his career (and the millions Ben mentioned) but he'll probably go down in sports history as one of the most infamous cheaters of all time. That's a very severe punishment when we have no real way of determining his ultimate guilt or innocence. What matters is not just whether he was doped on that day or not, but who was responsible. At least in criminal or civil law, every party, from the cops to the prosecutors, to the jury and the judge are closely scrutinized through a well defined process. But in doping we already have an apparent history of questionable ethical behavior by anti-doping officials who are in effect judge, jury, and executioner, yet who's watching the watchers? Where are the checks and balances in this system?

8/07/2006 08:54:00 AM  
Blogger finou said...

About Eric's comment on anti-American sentiments being to blame: I don't think so. Anti-American sentiment is not that strong at all. (I would say it's rather like anti-French feelings here.) I believe the testing facility is located in France. However, it's under strict international doping agency control. The people working there are all from lots of different countries (including the US). If there is corruption at that facility then it stands to reason that it's the same at any anti-doping test site anywhere including here.

However, I do think the leak to the media on Armstrong's case might have to do with strong anti-Armstrong (not american) sentiments (they really don't like him because he has supposedly acted too cocky sometimes by saying that there was no one else in the field that could pose a challenge for him).

The point of the 2 samples is that they are tested indenpendently by different people and use slightly different tests. The reasoning is that if there is some experimental error with the first sample or that for some reason it was a false positive, the B sample should come back negative since different people test both samples and they do more tests with the second. It's not perfect but it has been shown to catch some false positives before.

Finally, I really don't like Landis's side's arguments. Those kind of make me feel that they are guilty... I don't buy the whole "it takes months for steriods to show up". That's completely not true. How long it takes for the drug to show up has to do with the dosage and method of taking it and you should be able to take some and have it show up within 30min... The boost of endurance Landis would get from a short term testosterone boost would have given him an edge and that's all there is to it. Everyone is very well trained and evenly matched so every little bit helps no?

8/07/2006 11:41:00 AM  
Blogger Ben said...

Yeah, I don't like Landis' arguments either. The idea that this level was caused by alchool, or produced naturally just seems wrong. If I recall correctly, the ratio (of testosterone to epitestosterone) found in sample A was something like 11:1 which sounds way too high for things like that. On the other hand, if he believes he's innocent surely he'd explore all the options...

I did find this article on testosterone and endurance which makes for an interesting read. For an event like cycling that is almost pure endurance (well, except for the sprinters), testosterone by itself doesn't seem to do much good.

Another reason for the two samples is that chain of custody is very important. Keeping the two samples seperate makes it less likely that one individual could contaminate both samples.

8/07/2006 01:27:00 PM  
Blogger Justin said...

According to various articles, including the one Ben referenced, alcohol has been shown to have a significant impact on T/E ratios and baseline T/E ratios seem to vary widely across the population. In the New York Times version of the article Ben linked to, it was also mentioned that T/E ratios increase as the result of winning a sporting event, presumably a once-in-a-lifetime ride following a crushing defeat the previous day would have an effect.

My fear is that, in order to keep pace with cheaters that are willing to experiment with substances and methods that have not been fully scientifically tested and validated, anti-doping agencies are relying on protocols whose rate of false positives is unknown. While a 4:1 T/E ratio (or the previous 6:1 ratio) will work to catch the cheats, without a controlled study that examines all the factors that might change the result, I have a hard time accepting the punishment Landis will receive. "Beyond a reasonable doubt" seems like the appropriate standard before we villify an athlete (and rob him of 2 years of his career).

8/07/2006 02:17:00 PM  
Blogger Eric said...

I'm glad to hear that they intentionally have lab workers from a variety of nationalities.

Finou, I didn't mean to imply that there was lots of anti-Americanism. When in France/Europe most people have been quite nice to me (maybe even moreso than when I'm a tourist in America), but it only takes one bad apple to corrupt the whole process.

Two samples seems strange to me. I would think it would make sense to have at least three, otherwise how do you know which test/lab to beleive?

Finou, do you know if they are able to work with very small samples (like when they do DNA tests and start with PCR), so they could split the "B" sample several ways and send it to several labs for different tests? Or do they just need the whole sample to detect trace ammounts of stuff?

Also, I thought that they tested riders at several times throughout the tour. Why not go back to a few previous tests from other stages (or other tours?) and see what Landis's ratios were. If they were always high, then either he's always been doping or he has a naturally high rate.

I agree the defense from Landis's camp seems a bit out there/desparate. But you could argue that that's what we should expect from people who haven't been doing and were unprepared for the accusations. If you were taking something, then you'd probably be well versed on alternative explanations. Or maybe they're pretending to be clueless.

8/07/2006 10:28:00 PM  
Blogger Vincent said...

I got the impression that they do have baseline values for Landis from other days, but perhaps I'm misremembering. In any case, what strikes me as odd about all this is that neither Floyd's protestations nor the case against him (Why would he take something for one day that would have negligible effect on his performance for that day but that would cause him to fail a drug test?) really pass the smell test. Something funny is going on, but I don't know what. Maybe he's guilty, or maybe he's just a really poor speaker.

8/07/2006 11:41:00 PM  
Blogger Justin said...

Apparently, performance enhancers are now a problem in chess tournaments.

Of course, they're talking about technological, rather than biological, performance enhancers.

8/08/2006 04:48:00 PM  
Blogger finou said...

The reason I'm pretty convinced that he took or somebody gave him something is that for the B sample they test the isotopes of Carbon in the testosterone. Since synthetic testosterone is made from soy (I think) it actually has a different ratio than what's naturally produced by the body. So basically, if the B sample test is also positive, they know that the ratio of testosterone to epitestosterone is off in both and also know it's synthetic testosterone and so it can't be a product of normal human day to day variation.

About the two samples: I assume that they do multiple repeats and a battery of test on both samples. It's just that the samples are kept completely seperate and tested by independent groups. I suppose they maybe could split the samples to send to 3 indenpendent groups but that's going to be expensive and I am not sure if the athletes can pee that much ;)

8/08/2006 04:50:00 PM  
Blogger Eric said...

I agree that an isotope ratio that matches that of the synthetic, but not the human testosterone would be "beyond a reasonable doubt" that it wasn't just due to him having high testosterone that day. But still, there's a problem since it's not clear if he took it knowingly, took it unknowingly (possibly by a supporter or or possibly by a detractor), or if the sample was contaminated after being out of his body.

WRT chess... Clearly they should hold the entire match inside a Faraday cage. :)

8/08/2006 11:55:00 PM  
Blogger Justin said...

How long-lived is cholesterol in the body? Particularly in the body of a Tour racer, which is presumably metabolizing like crazy?

My understanding is that the isotope ratio test compares the person's carbon isotope ratio in testosterone to the person's ratio in cholesterol. That would seem to imply that there is nothing inherently natural about any particular isotope ratio and that it is conceivable that a person that was buring 10,000 calories a day and burning energy reserves would potentially have different carbon sources for different substances. My understanding is that the isotope analysis for urine is further clouded by the fact that they're looking at breakdown products over the time that the urine was produced which is less accurate than looking for the actual stuff in blood.

While I'd tend to believe a carbon ratio analysis for, say, my urine, since I'm consuming and burning basically the same amount and not doing anything metabolistically interesting, I am less confident that the analsis is foolproof for elite athletes whose bodies are doing some crazy things. I'm confident that the anti-doping agencies cannot do a scientific analysis of the rate of false positives and false negatives for a given test among elite athletes. Relying on extrapolating Landis's hormone levels after Stage 17 of the Tour de France from my hormone levels sitting on the couch (or even running on a treadmill) since we're both males of roughly the same age, I would tend to take the results with a grain of salt.

8/09/2006 10:40:00 AM  
Blogger finou said...

Cholesterol and steroid hormones are actually very related. In the body, steroid hormones are made from cholesterol which means the two should have pretty much the exact same ratio of carbon. I'm pretty sure that for cholesterol-like molecules (including steroids), the ratio in urine is pretty much going to mirror what's in the blood very closely (just based on how kidneys work... it's not like those molecules get actively pumped out/in). Now, I agreee that levels of single types of molecules would be very different in athletes compared to me sitting on my butt but that's why they normalize to other related molecules (i.e. the levels in the athlete's urine might be much higher than mine but ratios should be pretty similar). Also, they don't look at the baseline of everybody as the control. They are comparing the test to the results of the previous days for that person. This is all besides the fact that people (and animals in general) usually are all have the same isotope ratio anyway, btw.

The only reason the synthetic is different is that it's derived from soy (or yams) which gets a lot of this carbon from the air (or maybe it was the soil or something like that) which has a very different isotope ratio because it gets more of the lighter one (or heavier one... I don't remember the exact details). However, it's apparently really obvious using any run of the mill mass spec and the test has very low false positive rates from what I've read. There's just no physical way that the carbon ratio in the steroid would be so different than the cholesterol if you look at the limit they set for what they call a positive result.

8/09/2006 01:24:00 PM  
Blogger Justin said...

Indulge a non-biologist for a moment...

Wouldn't the ratio of carbon isotopes in a human be directly related to the food that they consumed (you are what you eat and all that)? If so, we'd expect a meat eater and a vegetarian to have different carbon ratios in general.

Assuming that is true, if we have an athlete that is buring 10,000 calories a day, that would suggest to my non-biologist mind that there is a relatively high amount of "isotope turnover" going on. It would seem plausible to me that, in that sort of situation, you may have cholesterol created x hours ago when your body was burning some fat from a steak dinner and testosterone created y hours ago when your body was burning a vegetable-based energy bar floating around at the same time, which would presumably screw up the isotope ratio.

Also, there is an American Statistical Association paper that talks about problems with the old 6:1 T/E ratio (let alone the new 4:1 ratio) and suggests that much higher ratios (in excess of 15/1) are more appropriate.

8/09/2006 02:13:00 PM  
Blogger Justin said...

The right link to the ASA paper

8/09/2006 05:09:00 PM  
Blogger Eric said...

I don't know about cholesterol or steriods. But I do remember reading that isotopes ratio in bones and teeth are related to diet. In particular, herbivores and carinvores are distinguishable from prehistoric bone samples. I think they even claim to measure when a species started to become omniverous or switch to a predominantly XXX diet based on isotope ratios (I want to say O and C, but I could be wrong) that are intermediate. But I'm not sure how relevant this is, given that isotope ratios in urine samples could change on much faster time scales than bone samples.

8/10/2006 01:58:00 AM  
Blogger finou said...

right so the ratio in a vegitatrian and a non-vegi person should be slightly different but the ratio on cholesterol and steriods in the body should be pretty much the same. I could be wrong but I'm pretty sure that you can't turnover those fast enough so that the ratios on the two molecules would be vastly different even if you are burning a bunch of calories because you are an athlete. The fact remains that you are making steroids from cholesterol so it would stand to reason actually that in an athelete where maybe they are producing more testosterone the ratios between the cholesterol and the testosterone should be even closer than in an average person where there may be a lag between when the cholesterol from the food is gotten and when it's used to produce hormones.

Also, I believe soy and yams have pretty different ratio from most other plants so even a vegitarian would not have the ratio that is indicative of synthetic hormones.

8/10/2006 10:11:00 AM  
Blogger finou said...

Here's a NY Times article on testosterone.

8/11/2006 11:36:00 AM  

Post a Comment

<< Home