Science of Sport's Bio Passport Articles

[BroDeal]

Part 1
http://www.sportsscientists.com/2011/03/biological-passport-legal-scientific.html

Part 2
http://www.sportsscientists.com/2011/03/biological-passport-effective-fight-or.html

 

[Cobblestones]

Part 1
http://www.sportsscientists.com/2011/03/biological-passport-legal-scientific.html

Part 2
http://www.sportsscientists.com/2011/03/biological-passport-effective-fight-or.html

Great. I especially liked the bar graph of the second link. The passport has some effect. It's the same as after introducing the 49% rule. It did have an effect. It's not the last word, but it is a start.

It is clear that if you still manipulate blood, it has to be in smaller increments. Not pints. Maybe third or quarter pints. Eventually, the effect is too small, the risk too big, the cost too much and you're left with 'amateur hour' such as Ricco (which I hope serves as an example why this is not a good idea). I still think that the absolute top riders can afford a scientific top-notch, undetectable program. So the playing field is still far from level. Maybe even less so than before. But blood doping should be a bit less rampant than it was before.

 

[sometriguy]

Awesome, a lot of work goes into this!

It is greatly appreciated

Thanks Sportsscientist

 

[Merckx index]

This is an excellent pair of articles, I've always admired and respected their work, but I do have a few complaints. First, there is no mention of any parameters other than reticulocyte percentage, making the passport seem like nothing more than a sophisticated off-score. What exactly are these hormones, growth factors, etc,, that are measured, and how are they incorporated into the final score? I have never been able to find any information on this. Anyone here know?

On the other hand, the re-infusion of blood (the blue arrows) causes a drop in reticulocytes. Why? Because the cells that are being re-infused are "older" (they've been stored in a refrigerator!) and so the new blood, post infusion, has more red blood cells, but fewer of them are immature.

But also because the introduction of more cells into the bloodstream suppresses reticulocyte formation. Very important point not mentioned.

The Off-score is of interest because it would be able to pick up both withdrawal of blood (characterized by a rise in reticulocytes and a fall in Hb), as well as the re-infusion of blood (reticulocytes fall and Hb concentration rises).

Except that re-infusion can be hidden by using EPO to counteract the fall in reticulocytes, and dilution to mask the Hb rise. They do mention the problem of masking in the second article. They do acknowledge that the changes in parameters seen first, since the EPO test, and second, since the passport, may be partly because riders have become more sophisticated in their doping. I tend to agree with the speed zone analogy, that just having the test is a deterrent, that it doesn’t prevent doping but reduces how much occurs (e.g., how much blood is transfused). I have thought this for a long time, and I agree with them that the evidence supports it.

However, there is still a serious question of how many riders can be caught by this technique. The 8/11 they cited was an experiment which was pretty much set up to maximize the possibility of catching dopers. The article does not point out that cyclists normally may follow a withdrawal with an immediate transfusion. As we have discussed here before, this is probably what riders do throughout the season, constantly replacing blood withdrawn so that they can store at refrigerated temperatures indefinitely.

We can see how effective this is on the graph with two strikes. Note that on the far left side of the graph no strike was recorded despite two withdrawals and one transfusion. There was very little change in reticulocyte %. And I’m not even clear from this graph how long after the withdrawal the infusion was made. If it was made immediately, as I suspect this is what riders are doing, there would be essentially no change in reticulocytes.

And again, transfusion itself--which is the name of the game for PE-- can be hidden quite easily. In a response to a comment, Ross notes:

In theory, someone could for example manipulate the Hb level independent of the ret% (or vice-versa), causing the off-score to remain relatively constant (remember that the equation takes (10 x Hb - sqrt(ret%)). So if both go up or down, the OFF score would potentially remain relatively unchanged.

Yes, very easily. Hb level can be lowered by dilution, which does not change reticulocyte %, while ret % can be raised with EPO, which will not affect Hb. Withdrawals are much more difficult to hide, but if they are always accompanied by a transfusion, there is no problem.

Finally, there is the question of timing. I couldn't access the whole original article, so I don't know the details of their study. In particular, I would want to know how long after a withdrawal or transfusion were the hemo parameters measured? What exactly is the window during which abnormalities can be detected? The longer the window, obviously, the greater the chance of detection. But given that the examples SoS showed were just barely significant, one has to assume that any time much longer after blood manipulation than was actually used, and there would be no positive.

 

[Escarabajo]

Great read. I haven't had the time to read the second article. Too much work around my office these days.

I have a question and maybe Cobbles or Merckx Index can respond it. When they flag the athletes do they use the 99% confidence limits or the 99.9% confidence limits? For flagging I'd use the 99% confidence limits. Otherwise a lot of cyclists will squeeze through the cracks when using the first filter. (I know that for banning they use the 99.9% limits but that does not mean that for flagging you can not use the other limit. The more you flag the more you scare the athletes. A simple letter is not costly). I don’t think the authors were very clear in that respect.

Note that with the 99% confidence range they can catch 8/11 (~73%) but that number later dropped to 5/11 (45%) using the 99.9% limits. The author does not mention if the athletes used for the study were trying to dilute or cover up the results or just were being careless; because that could drop the catching rate a lot more.

The conclusion of this Pottgiesser paper, incidentally, is that the off-score had high sensitivity in detecting autologous blood transfusions - in 11 cyclists, it caught 8 during this simulated season at a probability level of 99%. At 99.9%, as you might expect from more stringent limits, it picked up 5 out of 11 doping athletes over the 'season'. The only false positive came from Hb in that one subject, not from the Off-score, which was recommended for future use in the biological passport model

The first false positive that they are talking about is kind of silly. I would not even call it false positive and would not include it in the data base. It is clearly seen that it is that way because it is the first sample taken and measured.

 

[Cobblestones]

I agree mostly with everything Merckx said.

For the hormones, growth factors etc. there is a 'steroid profile' in the works (or maybe already in place) which does longitudinal testing of natural steroids (not clen or anything which is unnatural).

When you look at the graphs, with each testing, the confidence limits get a little bit tighter because the dataset becomes bigger and the true distribution of parameters can be pinned down more closely. The same statistical technique is apparently applied to the steroid profile, although I have never seen a dataset published. However, it is completely separate from the blood values because there's no direct physiological connection between blood parameters and steroid profile.

As to Escarabajo's question, I think they use the 99.9% limit. Now, statistically speaking, finding two independent data points above the 99% limit (but below the 99.9% limit), when seen together, are statistically more significant than one datapoint above the 99.9% limit. But I don't know whether such a view is adopted by the UCI/WADA.

And I agree in particular with one point by Merckx. I would like a clear statement how long after a transfusion and/or withdrawal it is detectable.

Let me also make one own comment. This is the first time I have seen mentioned (beside the Pechstein case, which was a mess), that high number of reticulocytes might be something they're going after. Up to now, there was only the off-score limit of 133 which basically only applies for a transfusion (similar in concept as the 49% crit rule). Apparently they are now also targeting the withdrawal periods which are followed by low hemoglobin concentrations and high retic counts.

 

[Escarabajo]

...

As to Escarabajo's question, I think they use the 99.9% limit. Now, statistically speaking, finding two independent data points above the 99% limit (but below the 99.9% limit), when seen together, are statistically more significant than one datapoint above the 99.9% limit. But I don't know whether such a view is adopted by the UCI/WADA.

...

This is what I am talking about, and I agree with you that 2-3 points breaking the 99% is more important than 1 point breaking the 99.9% rule. Analysis is more powerful when analyzed qualitatively than quantitatively, as the author says, but if they don't flag when 2 or 3 points break the 99% confidence levels we are going to let a lot of dopers through.

I might drop a note in the website.


Thanks.

 

[Merckx index]

The author does not mention if the athletes used for the study were trying to dilute or cover up the results or just were being careless; because that could drop the catching rate a lot more.

I'm quite sure they didn't, which is why, in agreement with you, I mentioned that I thought this method probably would not catch many actual riders blood doping (a point I've made here before in connection with Bert's case). Though I could not access the whole article, my impression from the abstract and the SoS discussion was that these were just ordinary individuals subjected to withdrawals/transfusions at known times, with no attempts at masking. AFAIK, the subjects weren't even athletes who would have known about masking, and would not have had access, for example, to EPO, or wanted to use it. And I can't emphasize strongly enough that these changes can be reduced to virtually zero by making withdrawals and transfusions at the same time, as their own data show.

The study was blind, in that the investigator taking and analyzing the blood samples did not know when the subjects had withdrawals and transfusions. This investigator then apparently made the call of whether blood manipulation had occurred at a certain point.

When you look at the graphs, with each testing, the confidence limits get a little bit tighter because the dataset becomes bigger and the true distribution of parameters can be pinned down more closely.

It will be interesting to see just how tight or narrow the limits can get when data are obtained from riders over a very long period of time, that is, several years. One would certainly expect some improvement in identifying an anomaly, as a rider's individual pattern becomes better established. Then again, if a rider is blood doping over this period, and several times does not get flagged, that could actually help him relax the limits, as the change is wrongly attributed to normal fluctuations!

And I agree in particular with one point by Merckx. I would like a clear statement how long after a transfusion and/or withdrawal it is detectable.

Don't know if they addressed this specifically in the paper, but their data would surely illuminate this question. I believe the study was set up so that the relationship between blood manipulations and analysis was more or less random, so one would expect some data points at several different times after a withdrawal or transfusion.

Apparently they are now also targeting the withdrawal periods which are followed by low hemoglobin concentrations and high retic counts.

This is something they should have been doing all along. As I have noted before, withdrawals are much more difficult to mask than transfusions, and while I believe riders in the know mostly withdraw and transfuse together, there is likely to be some point, maybe in the offseason, when the process begins with a withdrawal with no corresponding transfusion. But then again, as LMG pointed out, it would easily be possible to start with a very small withdrawal, one that would not trigger a significant change in parameters, and work your way up.

If the doping doctors aren't already aware of data like these, they certainly are now! I'm sure they have the answer to the question of how long after a blood manipulation the parameters remain abnormal.

 

[Tyler'sTwin]

Does it matter if you can't pick up when a rider switches the old blood in his fridge for a fresh bag (of course you can't as long as the amount is roughly the same)? A blood bag wont magically appear in your fridge. In order to go X ml up in a race, you have to go X ml down at some point.

Good point about working yourself up from a smaller amount. For example, you could maybe start with 150 ml and then increase the amount withdrawn by that much everytime you withdraw/reinfuse. It would be a more drawn out process though. Not sure how it would affect training.

How effective is saline at lowering Hb and how long does it last? There has to be a reason why HES was preferred before it was tested for.

 

[athletepassport]

For the hormones, growth factors etc. there is a 'steroid profile' in the works (or maybe already in place) which does longitudinal testing of natural steroids (not clen or anything which is unnatural).

The urinary steroid profile forms the steroid passport, growth factors the endocrine module of the passport. See papers by RI Holt in Pubmed.

The same statistical technique is apparently applied to the steroid profile, although I have never seen a dataset published.

search images in google using steroid passport

As to Escarabajo's question, I think they use the 99.9% limit. Now, statistically speaking, finding two independent data points above the 99% limit (but below the 99.9% limit), when seen together, are statistically more significant than one datapoint above the 99.9% limit. But I don't know whether such a view is adopted by the UCI/WADA.

It is. See the colorbars in the figures shown in the SoS article. An efficient way to detect microdosing.

 

[Merckx index]

How effective is saline at lowering Hb and how long does it last? There has to be a reason why HES was preferred before it was tested for.

The main advantage of HES is that you don’t have to go to the bother of an infusion (which of course means you need some privacy for a while, as well as storing a relatively large bag of fluid). The effects of HES may also last longer, since as long as it’s present in your blood, the extra fluid will be maintained. When you take a saline infusion, your Hb and HT will immediately drop, but over time the extra fluid will be absorbed by your body, and those parameters will go back up.

The urinary steroid profile forms the steroid passport, growth factors the endocrine module of the passport. See papers by RI Holt in Pubmed.

Thanks a lot for this link. When I checked out some of Holt’s papers, I realized I had read about the hormone approach before, the use of IGF-I and some other hormones as markers for growth hormone use. IMDLO (in my data-limited opinion!), the hormone passport may prove to be more effective than the blood passport. As we have discussed in this thread and others, there are established ways to beat the blood passport. I don’t believe the hormone changes following GH use can be masked or eliminated so easily.

The steroid passport, as I understand it, is basically an extension of the T/E test that used to be used exclusively, and is now the initial screening step, in the synthetic T-test. Look for products of steroids in the urine. The T/E test can be foiled by taking E, but assuming this steroid passport looks at a number of markers, it may be difficult to mask use. Again, based on my limited knowledge of test details, I think this may be more effective than the blood passport.

 

[Race Radio]

I agree mostly with everything Merckx said.

For the hormones, growth factors etc. there is a 'steroid profile' in the works (or maybe already in place) which does longitudinal testing of natural steroids (not clen or anything which is unnatural).

These profiles start next season, however they have been collecting and testing samples this year in order to build a profile. Some internal testing programs also do this already.

 

[lean mean &green]

passport a crude measure but the best we've got for now

from memory (sorry, no time for those links) i don't think the steroid profile will be ready for primetime until around 2012. UCI page is vague, says they're doing it without assistance from the steering group for now (ie w/o 3rd party help in the form of WADA and other experts)

i did a quick skim of WADA's 2010 ABP Operating Guidelines. they outline specifics of hematalogical parameters but saw no mention of the "endocrine module" whatsoever. sad really, because the manipulation of anabolic drive should be monitored with the same, or preferably more veracity, as the hematological profile is at present.

a couple of other issues worth mentioning:

- i think we'll see total hemaglobin mass added to the profile and probably much sooner than i originally thought. with that, it's likely we'll see a new and improved OFF-score for lack of a better term, based upon mass and retics which could really disrupt this transfusion circus.

link to everything you ever wanted to know about tHb-mass, enjoy!

another helpful link on the variability, or lack thereof, of tHb mass. even though some published passport data doesn't fall outside of the 99.9% cutoff it still looks suspicious when you realize how little tHb mass oscillates - that doesn't translate directly to Hb and retics but it's to be considered. the feedback for O2 homeostasis is extremely sensitive and precise. natural erythropoesis to correct imbalance is always ongoing and very gradual.

EDIT: ninja'd by RR on the steroid profile timeline, FWIW i think he's correct.

 

[Merckx index]

- i think we'll see total hemaglobin mass added to the profile and probably much sooner than i originally thought. with that, it's likely we'll see a new and improved OFF-score for lack of a better term, based upon mass and retics which could really disrupt this transfusion circus.

Yes, that would help a lot. That would take dilution off the table, and make transfusion almost as difficult to hide as withdrawal. Though again, as long as riders are doing both together, there still is virtually no way to catch them. Except when they transfuse before a major race, and there is a window of a day or more before they follow with a withdrawal. But even then, a hard race will have the effect of lowering somewhat the elevated parameters, and of course if it's a stage race, each stage will have that effect.

 

[lean mean &green]

Yes, that would help a lot. That would take dilution off the table, and make transfusion almost as difficult to hide as withdrawal. Though again, as long as riders are doing both together, there still is virtually no way to catch them. Except when they transfuse before a major race, and there is a window of a day or more before they follow with a withdrawal. But even then, a hard race will have the effect of lowering somewhat the elevated parameters, and of course if it's a stage race, each stage will have that effect.

athletes would have to be on a never ending mistake free transfusion cycle even throughout the off season. tHb-mass only changes by +/- 3% all year. your values in January should be similar to those in July.

tHb-mass also remains pretty steady thru-out consecutive racing days.

tHb-mass is a game changer.

 

[Tyler'sTwin]

Yes, that would help a lot. That would take dilution off the table, and make transfusion almost as difficult to hide as withdrawal. Though again, as long as riders are doing both together, there still is virtually no way to catch them. Except when they transfuse before a major race, and there is a window of a day or more before they follow with a withdrawal. But even then, a hard race will have the effect of lowering somewhat the elevated parameters, and of course if it's a stage race, each stage will have that effect.

No, it's much better than that, because changes in tHb mass cannot be explained by variations in plasma.

Well, so what? You can catch them when they are going X ml of blood down and you can catch them when they are going X ml of blood up, which is unavoidable when blood doping (and the latter is the whole point of it). When riders do both together it's simply a way of prolonging storage. It's the method of long term storage that you have to settle for if you can't freeze blood. It's hardly a big problem that this zero-sum game isn't detectable.

 

[Escarabajo]

...

a couple of other issues worth mentioning:

- i think we'll see total hemaglobin mass added to the profile and probably much sooner than i originally thought. with that, it's likely we'll see a new and improved OFF-score for lack of a better term, based upon mass and retics which could really disrupt this transfusion circus.

link to everything you ever wanted to know about tHb-mass, enjoy!

another helpful link on the variability, or lack thereof, of tHb mass. ....

Thanks. Good stuff, although I am a little confused since some of that information I could not understand.

This test was discussed in this forum before here:

http://forum.cyclingnews.com/showthread.php?t=3479

In my ignorant opinion it looks like they are going for the small transfusions per a 3 week period. maybe >6 times per GT. Logistically it looks like very difficult to do but the top fish can always pay these expensive programs. I say this because at the end of the presentation it says that the HB mass test is less effective wit less amount of blood infusion.

Maybe if we combine the whole array of tests like the blood passport + Plastisizer test + HB Mass test we get a more certain probability of catching who is doping during a GT.

 

[Merckx index]

athletes would have to be on a never ending mistake free transfusion cycle even throughout the off season. tHb-mass only changes by +/- 3% all year. your values in January should be similar to those in July.

tHb-mass also remains pretty steady thru-out consecutive racing days.

tHb-mass is a game changer.

Maybe. But LMG, you were the one who pointed out to me that riders could start in the off season with very small amounts of blood withdrawn/transfused, and work up. 3% of total blood volume is 150-200 ml., one could certainly begin with that or less. One could also get a significant PE effect with volumes on that order.

If it is a game-changer, I think it will catch riders not during the off-season, but when they transfuse before a major race or GT stage, without an immediately preceding withdrawal. I think that's when they're most vulnerable.

But also keep in mind that riders are strongly opposed to this CO test, they claim, correctly or not, that it may be detrimental to their performance. This test has been around for a long time, I think they need a better one.

 

[lean mean &green]

Maybe. But LMG, you were the one who pointed out to me that riders could start in the off season with very small amounts of blood withdrawn/transfused, and work up. 3% of total blood volume is 150-200 ml., one could certainly begin with that or less. One could also get a significant PE effect with volumes on that order.

If it is a game-changer, I think it will catch riders not during the off-season, but when they transfuse before a major race or GT stage, without an immediately preceding withdrawal. I think that's when they're most vulnerable.

But also keep in mind that riders are strongly opposed to this CO test, they claim, correctly or not, that it may be detrimental to their performance. This test has been around for a long time, I think they need a better one.

i resisted responding b/c it's hard to do so without getting into the nitty gritty of how i think autologous transfusions occur and that would turn this into the longest post ever. you are correct, i implied that if you were exercising extreme caution (not necessarily the case) you could begin an autologous transfusion cycle with small amounts but the goal would still be to pair withdraw/reinfusion/dilution/microdosing in such a way as to avoid detection and gradually build up banked blood volumes of a full unit or more. IOW, always withdraw slightly more than you reinfuse and correct a little quicker with microdosing - if you do it right there aren't any big withdraws to cover up and there's hardly even a need to dodge DCO's.

tHb-mass makes reinfusing that banked blood in larger amounts during a peak/priority event nearly impossible - not the case at present.

3% increases are seen over a longer timeline - not suddenly right before an event or in the middle of a stage race. i would think that when cross referenced with retics, 150-200 mL of whole blood (approx 30g Hb-mass) would be enough to get you into trouble.

could a rider transfuse less than 150 mL and get away with it? possibly. would that result in noticeable PE? yep, some. would a talented athlete unwilling to take those measures now be able to compete? maybe, but we're getting a lot closer. it's reasonable to think that risk reward will shift in a measurable way and the graph below (from OP SoS articles) suggests that the even small innovations can act as an effective limiter. mass measurements force some to think about whether or not the benefits of such small transfusions are no longer worth the risk.

also, tHb-mass irregualarities are a stronger and more defensible position for authorities pursuing a case of ABP abnormalities than we have at present. based upon the independent observer report last summer i'd say that the UCI is stuck in no man's land at the moment. they're commited to the ABP but not nearly enough for it to make a profound difference. new and better parameters are needed and sampling is too infrequent and predictable. at the same time they appear to have diverted resources away from testing blood and urine for specific substances. some see it as convenient negligence but it might just be a difficult period of transition. i honestly don't know.

 

[lean mean &green]

Thanks. Good stuff, although I am a little confused since some of that information I could not understand.

This test was discussed in this forum before here:

http://forum.cyclingnews.com/showthread.php?t=3479

In my ignorant opinion it looks like they are going for the small transfusions per a 3 week period. maybe >6 times per GT. Logistically it looks like very difficult to do but the top fish can always pay these expensive programs. I say this because at the end of the presentation it says that the HB mass test is less effective wit less amount of blood infusion.

Maybe if we combine the whole array of tests like the blood passport + Plastisizer test + HB Mass test we get a more certain probability of catching who is doping during a GT.

your opinion doesn't sound ignorant to me at all. we're merely adding mass to the passport as a more reliable measure. it's a process. testing error will decrease and data collection over time will add to it's effectiveness.

that's a good thread begun by Krebs Cycle and there were some great questions, many of them are answered or have begun to be answered in my first link. i think i know who KC is IRL (the clinic buzzword this week) and he's very credible.

 

[Cobblestones]

that's a good thread begun by Krebs Cycle and there were some great questions, many of them are answered or have begun to be answered in my first link. i think i know who KC is IRL (the clinic buzzword this week) and he's very credible.

We had a few very knowledgeable posters at that time. Krebs cycle, Mitochondrion (??) and a few others (names escape me). I haven't seen them lately.

 

[python]

http://forum.cyclingnews.com/showthread.php?t=3479

biopassport has been the centre piece of (at least) a half a dozen of other threads.

each time some article appeared, there was a new thread. imo, there's NOTHING new, including the haemoglobin mass test suggestions. there were plenty of links to the methodologies, applications, interpretation interviews, limitations etc...

except, of course, the cas validation decision.

the science of sport fellas did an excellent job. but again i saw all of that before.

 

[athletepassport]

we're merely adding mass to the passport as a more reliable measure.

the issue is that riders cannot be forced to breath carbon monoxide. And to the best of my knowledge there is no other reliable method to measure the mass of hemoglobin.

 

[lean mean &green]

the issue is that riders cannot be forced to breath carbon monoxide. And to the best of my knowledge there is no other reliable method to measure the mass of hemoglobin.

in larger amounts carbon monoxide is poisonous and potentally deadly but it's around us in small amounts everyday, it's a small byproduct of our own metabolism and is produced by industry and automotive engines. the amts used here are innocuous and a more thorough explanation supported by data is found in my first link.

there are some small implications immediately following the analysis but there's no reason not to test shortly after the completion of an athletic event. it's no more invasive than asking for a blood or urine sample and the process takes less than 15 min.

 

[athletepassport]

in larger amounts carbon monoxide is poisonous and potentally deadly but it's around us in small amounts everyday, it's a small byproduct of our own metabolism and is produced by industry and automotive engines. the amts used here are innocuous and a more thorough explanation supported by data is found in my first link.

there are some small implications immediately following the analysis but there's no reason not to test shortly after the completion of an athletic event. it's no more invasive than asking for a blood or urine sample and the process takes less than 15 min.

Wrong. The dose does exceed the ceiling limit or STEL fixed by several organizations:
http://www.transducertech.com/pocketco/govco.html