Monday, May 21, 2007

Hearing - Mon Meier-Augenstein III

Third time's a charm? Let's see if we can get anywhere now.



YOUNG: objection to style of presentation, for example, page 10, and page 15, which are just examples of what's going on. Counsel is perfect free to ask opinions or show chromatograms. It is improper direct with text answer to a question, then ask the question, and then have him read off the slide. That's not direct testimony. That's whas 90% of this is about. If they want to ask him questions, fine, but they can't have answers on screen before he answers.

[more]


SUH: presentation is by the Doctor; they are his words, not ours. All of these have previously been provided by the Doktor. The use of presentations has been done before, by Dr. Catlin. Throughout USADAs case they have done a summary, asking "are they reliable", getting answers "yes", and moving on.

It will be deeply helpful to the panel to understand how chromatograms are analyzed. Consistent with the expedited nature of arbitration, that USADA has always cited, we are trying to do this in that context. If this is cutoff, the panel will not have vital information about our argument.

YOUNG: there is comment that all this is provided to USADA. Doctor never filed a report. If you look at slide 58, this is conclusions to be drawn from mix cal acetate. He can be asked to look at chromatograms. We're trying to address the fact his testimony is prepackaged in a slide show before the questions are asked.

[ silence, panel confers ]

[ it's official: some of the press want to go home ]

CAMPBELL: time concern. Are you at the point you may not be able to put on your case?

SUH: it depends how many interruptions we have. We could get to closing if we were allowed to proceed.

BARNETT: we're willing to give you 3-4 of our hours if that will help. We want the evidence in.

[ he says unconvincingly ]

SUH: that would help.

MCLAREN: you can show the graphs, but don't show the conclusions until he gives them.



l

q: have you seen some of these issues in the chromatograms (CG's) in this case?
a: no, they've been run on a different machine, so there isn't the detail that even shows what is going on. In this slide, you can see the subtraction and the peak start stops. Had to make some working assumptions that the practice was consistent and done throughout, and that it's done the same way. But sometimes the CGs have auto, sometimes zero...

q: quality of CG?
a: not very good.

q: EX 110 LNDD 1791, USADA 343. LNDD1798 Describe what's in the circles.
a: they all show subtraction.

q: which is this?
a: some F2 from other positive sample.

q: so this is a good one?
a: yes, good separation.

q: which is this?
a: this is Landis' f2.

q: floyd's show the poor CG you said earlier. Which is this?
a: another where the CG is even worse than Floyd's, shows how BG sub works with that kind of peak.

q: does BG sub matter?
a: if you start to cut away peak area, you lose peak information.

q: next slide...
a: this is blowup of previous, where BG sub changes the number.

q: next slide, who prepared this.
a: I did, from the data printouts by LNDD.

page 17 -- delta C balues of BG during analysis of A sample as reported by LNDD. Goes from -64 to -52 from IS, Etrio Andro, 5B, 5A Pdiol, 11k

a: sloping background problems.

q: this is the a sample. [ as listed above ]
a: numbers are BG C13 numbers.

q: shows A F2 line, A f2 line, Blu F2, blu f3, explain A f3.
a: the dotted line is a trend line showing a perpetually changing background.

q: meaning?
a: effects the results. Subtraction of something constantly changing from adjacent peaks.

q: don't know whether to subtract this or that value?
a: correct; the sw algorithm picks. but the 44/45 trace shows it looking flat, when it isn't.

q: that's the one Brenna used to say the baseline was OK.
a: yes, I believe he's wrong.

q: so we have changing background...
a: for the respondant's samples.

q: but flat for the blanks?
a: yes.

q: did you make a similar for the B?
a: yes.

q: same problems?
a: yes. virtually the same regressions on the slope. all made on assumptions that peak identity are correct, which I can't vouch for, because it was next to impossible to confirm the identities.

q: so you're saying you don't even know if what they are calling the metabolites is correct?
a: corrrect.

q: Quote from USADA brief, para 42. "GC-MS gold standard"
a: yes, in principle.

q: 42, relative retention time.
a: yes.

q: para 52, the LNDD criteria are the WADA criteria.... retention time must fall within the range...
a: yes.

q: in order to properly identify if you have 5a, 5b.
a: yes.

refers to block diagram

q: you have to compare the retention times from this to that phase, and make sure they are within a certain amount.
a: yes.

q: look at the TD2003IDCR, identifying qualitative assays. "shall not differe by more than 1% or +/- .2 minutes, whichever is smaller.
a: correct.

q: EX 24, USADA 130. explain.
a: shows retention times for compounds compared to internal standard.

[ uses laser ]

q: "Dr. Brenna said they have retention times that match on previous GCMS."
a: applies to any situation.

q: have you looked at the relative retention times for Landis' samples?
a: yes.

q: do they comply with IDCR?
a: no, they do not.

q: lots of data...

[ shows from 5-8% differences in retention times ]

q: for A sample, explain what the last column is.
a: % differenct of relative retention times

[ this 5.7% to 6.7% ]

q: supposed to be within 1%.
a: yes. this is a huge difference. I have no confidence in the peak identification whatsoever.

q: isn't 6% close enough to 1%
a: it's 600% different than 1%

q: but you know?
a: no I don't.

q: forget about the rule, you know, it's kind of the right place.
a: It is not identified, proof positive. It misses the whole point of relative retention times.

q: lower quadrant, absolute retention time, A sample.
a: varies from 6.15 to 8.33 minutes.

q: so 27.53 and 19.06 on one value?
a: yes.

q: requirement?
a: 0.2 minutes.

q: this difference is?
a: 6 minutes.

q: percentage?
a: astronomical -- 4000%

q: isn't that close enough?
a: no.

q: next column, B sample. relative in %. What's the requirement?
a: 1%.
[ varies 6.1% to 7.2%]

q: B sample absolutes - requirement?
[ 6.31 minutes to 8.64 minutes ]
q: 0.2 minutes


q: which chromatogram?
a: F3 fraction USADA 349, B sample; virtually impossible to say which peak is the internal standard; this is the one with the 5a and 5b?

q: can't you tell by looking?
a: no, I can't. given a difference of 4 minutes, there's only a presumption.

q: so add 240 seconds, you end up in this bunch.
a: benefit of hindsight.

[ lost it ]

a: illustrates similar peaks at different times, to show identification problems.

q: so this looks sorta like this one.
a: yes.

q: so you can't just look at it.
a: appearances would be foolish.

Slide 28

q: why this table?
a: illustration of peak overlap effects on reported deltas.

[ shows -27.4 reported at -32.1 ]

[ shows a fatty acid chromatogram ]

q: what does this show?
a: bad chromatogram, good example of strongly overlapping peaks. This slopes up due to column bleed; there is something else.

q: we've heard if small peaks co-elute, it doesn't matter
a: if you know what you are dealing with, yes, but not if you don't.

[ Landis is focused ]

a: if it's hugely different than your target peak, you can't assume the contribution is minor.

[ slide labelled consequences for delta C13 values. from previous fatty acid slide ]

a: so this compares what you get from purely automatic analysis; we see three injections. Software get's a funny number?

q: isn't it close enough?
a: no!

q: opinion on the chromotography here, and the influence on the results.
a: no confidence. peaks have disappeared, co-eluted, overlapped, long tails downward sloping baselines.

q: A sample F1 USADA 158 what do you see:
a: good, stable baseline, no interference, good separation. daylight between peaks.

q: compare fo USADA 165 and 167.
a: the circled area shows them just about separated in the GCMS, the IRMS shows first peak bleeding into second.

q: affect?
a: will affect C13 numbers for both. The first will look more positive, the latter more negative.

q: USADA 171. circles
a: clearly visible shoulders between 5a and 5b, and 8 minor peaks following.

q: GC/MS chromatogram.
a: 4 peaks not 8 following; disappearing shoulders; where did they go?

q: it's so small it doesnt' matter?
a: can't draw any conclusions, only speculate.

q: good enough?
a: terribly sorry, you don't work on asusmptions.

q: they're all cheaters.
a: they all have rights to a fair hearing and access to the data. I don't know what these are, given the discrepency in the retention times.

q: slide 39, USADA 348, f3 sample B sample. GCMS
a: yes.

q: USADA 349, disappearing peaks, impact?
a: can't say. it's either part of one peak or the other.

q: 5b tails into 5a.
a: it will shift the delta values reported.

q: symmetry is important?
a: yes, if you don't know what the sample is.

q: opinion about whether LNDD controls work for 5a?
a: no, where is it in this chromatogram?

q: so you don't know if the IS is the IS?
a: right.

q: assuming they picked the right IS, did they find it with the right specs?
a: some fall outside their specs.

Q: opinion about whether data of IS should be considered for QC?
a: should be considered. If I read transcripts right, even ??? thinks they should be considered.

q: slide 51, USADA 161
a: A sample fraction shows IS at -31.64; accepted range is -29.96 to -30.96 out by .47/mil.

q: slide 52

[ Internal standard story - A graph showing the IS drops way out on the F1, way up on the Blu F2]

a: they are not run consecutively,

q: coincidence the time is 6 x the run time?
a: time has no documentation.

q: what is vert axis?
a: limits of acceptability, and some out of range.

q: close enough?
a: should re-run the sample. It's not close enough.

[ slide 54, internal standard story, B ]

q: what?
a: gap between first mix cal acetate and sample.

[ shows 2 out of range, bhe blu F3, and sample F1 ]

q: gap?
a: would let you run 5 analyses.

q: close enough?
a: no. out is out.

q: have you seen or read that mix cal acetate is a positive control, run before and after samples.
a: yes, and that it's run before and after, and there are at least two cases it is not the case.

q: why positive control.
a: the blanks are negative QC, known drug free, should never come up positive. Positive means a sample that is known positive.

[ MIX cal F1 sample -31.64, F2 blank -29.94, F1 sample -31.08, F3 blank -31.54 ]

q: Within range for mix-cal acetates.
a: yes.

q: why not relevant in the samples?
a: shows the machine is doing the right thing in the sample that has complex matrix.

q: so the mix-cal acet is clean.
a: shows it works under benign conditions.

q: but samples are dirty.
a: that's why you should run positive and negative QC that are matrix matched.

q: USADA 361. Who prepared this slide 60?
a: i did.

q: good chromo?
a: mix cal is fine. it would be boring if that wasn't ok. the other one has peak overlap.

a: lab has a standard that is better than they used here, but would give the lab better feedback. this is shooting fish in a barell. This is not challenging; running urine samples IS challenging.

q: USDAD 158 USADA 161, USADA 182 stacked, showing blank F1, Sample F1, Mic cal
a: sample has high sloping baseline.

q: what's the difference?
a: the standard is QC for instrument, depends on the quality of the CG.

q: In USADA's brief, they say mix cal's run immediately before and after, Para 79. Were they?
a: 5:40 minutes isn't immediatly after. 4:40 isn't immediately before, in my opinion.

q: Testosterong and GC IRMS. USADA ex 37.
a: yes, good overview.

q: slide 77.
a: simplified, during all this, the carbon skeleton doesn't change; only hydrogens. one peak always twice the size of the other in the 5b and 5a peaks. Useful for troubleshooting potential problems.

q: EX 1241, EX 40 "product shareing same precursor show have comparable skeletons"

[ from Shackleton ]

q: Are lndd's results consistent with this?
a: no; max in Shackleton if 1.74%;

Slide 79 USADA 1241

a:: LNDD's are 1.39% for athletes, 2.76% for T positive athletes.

[ Slides 80 ]

A: never exceed 2 per mil, in good studies.

q: slide 81, what does this show?
a: bunch of computation...

q: slide 82,
a: Differences between 5 5a differences are inconsistent with published studies.

q: had you seen in any study, a difference of this magnitude?
a: no.

q: what if you saw this result?
a: if I saw such results, I'd make sure it was right.

q: cologne testogel study. what conclusions?
a: surprising, results consistent with Shackleton and Aquilerra.

a: should look at difference, and not even close to 2 per mil. At any other point,

[ landis looks as lost as me. ]

q: did you read the irms could be explained by testogel, because of volunteer P3? What didn't USADA mention?
a: they looked at the two results, at an instant, but there is more information that is the differnce betwen the two, which is inline with the other papers, giving 3 studies, 3 methods, and they agree. Not one instance where the difference exceeds 2.2. So if you see a result that is 2x that, then I'd be very suspicious, and spend time to make sure it stands up to scruting.

SUH: Has a flight...

YOUNG: going to take some time

LUNCH, break till 12:45.

21 comments:

Anonymous said...

[ he says unconvincingly ]

It sounded pretty convincing to me, and it certainly didn't have to be done.

Anonymous said...
This comment has been removed by a blog administrator.
brian said...

this questioning now is very effective....makes me wonder why these crack lawyers didn't have the page #'s on the slide show to begin with. They almost shot themselves in the foot this morning.

Brian

Anonymous said...
This comment has been removed by a blog administrator.
Cheryl from Maryland said...

Hey science guys, how convincing are you finding Dr. M-A? He seems to me to be hitting home runs on loss of peak information due to overlapping peaks, shifting baselines, and retention times.

brian said...
This comment has been removed by a blog administrator.
Anonymous said...

wheres a copy of the slide show?

Chaz said...

It sounds like USADA is pretty confident that they are going to win and they don't want the public opinion to go the other way because the defence didn't have time to present their case. Who is doing the timekeeping, by the way, it doesn't seem that Landis has used that much more time then USADA.

Anonymous said...

They definitely have used more time than USADA. Cross examination has taken forever.

Chaz said...

More, yes, but 6 hours more??? Sounds like lawyers billable hours.

Anonymous said...

Where is Robby Ventura ? (Floyds coach). I put coaches and doctors in the same cat . I find it very suspisious that he hasnt been heard from in this preceding .He knows the whole story . Look for more heads to roll in the coming monthes in cycling .

Ron Cook said...

Anonymous said...
wheres a copy of the slide show?

Even though i can't access the video(too many users), if I log into the CVN page I can still see all of the slides.

Anonymous said...

The coach would not have anything to offer re the science. He could say that, to the best of his knowledge, he does not think Landis doped, but that is of little to no evidencery value.

brian said...

It seems like Alan Lim was doing more for Floyd than Robby. Robby was busy reporting for OLN. Alan was with Floyd all the time during the tour and did much of his power training and wind tunnel coachin in the pre-season. He is a scientist. I think he could add alot to this hearing.

Anonymous said...

Ventura appeared on Versus in March? on their Cyclysm show. While everyone else (Phil. Paul. Bob) was trying to discuss Floyd's current situation all RV could say was "What we should all be talking about is what a fantastic ride Floyd's stage 17 performance was, that's what we should be talking about". It made me wince - talk about a guy either i) still in denial; ii) with his head in the sand or iii) just totally clueless.

Duckstrap said...

To Cheryl from Maryland--I find M-A's testimony very convincing. These are the issues we have been hammering on for awhile on DPF. This is where Ayotte's "it's so because I say so" does not work for me. These tests are complex, and, though more or less standard, without appropriate care are there many ways to get very erronious results.

Anonymous said...

Clean vs dirty samples: Clean samples test the response of the detectors but offer no challenge to the software, i.e. analysis. Dirty samples should offer no additional challenge to the detectors but greatly challenges the software (signal processor). Different algorithms will produce different results on dirty samples and identical results on clean samples.
I am not an expert. Is there an expert that wants to comment?

Anonymous said...
This comment has been removed by a blog administrator.
Anonymous said...
This comment has been removed by a blog administrator.
Anonymous said...
This comment has been removed by a blog administrator.
Anonymous said...
This comment has been removed by a blog administrator.