DRUG TESTING ADVISORY BOARD
OPEN SESSION
March 6, 2001

Agenda Item: WELCOME

 

MR. STEPHENSON (HHS):   I want to congratulate every one of you who arrived here.  This was quite a challenge.  I wasn't sure if the snow was going to prevent people from attending the open session.

                All of the Board members and invited guests introduced themselves.

 

Agenda Item:  HHS UPDATE

MR. STEPHENSON:   As you know, we've gone through a process in testing review that started last September as a result of a particular case that had arisen from Delta Airlines in an FAA overview.  To date we have done probably a half a dozen interviews with the media on this, a couple with the Wall Street Journal and the New York Times, and each time the information that was requested was: What's the status?  What's the bottom line? What's the end?

                We are not there yet.  We have sent out a general framework letter.  It has been modified and delivered to each of the laboratories that had to do so, to the medical review officers that had received test results that needed to be adjusted.

                To date we have information that has been submitted from most of the medical review officers, but to the best of my knowledge still not yet all of them.  The idea is that we are working aggressively, both with DOT and with our contractor, Research Triangle Institute, to try to put together a consolidated status on that.  Our goal was to make sure that wherever there was an incorrect or an unsustainable test result for specimen validity, that those had been identified in all of our laboratories.  Every laboratory was inspected.  There was a self-reported audit, followed up by an on-site audit by an inspector(s).  Lists were assembled that gave the specimen ID number.  Remember we don't have names of individuals, so the link back to a person was through the medical review officer and ultimately could be through the employer.  Our goal is to make sure that everyone who has had one of these test results documented against them has been notified to the very best of our ability.  That's our overall goal.

                Additionally, we are changing areas that we found needed more attention.  In a separate section, we will talk with the members of our Drug Testing Advisory Board in that capacity, as our advisors about a Federal Register notice which will take what we had generated as program guidance and will convert it to modifications to our Mandatory Guidelines.  We're in an advanced stage of document preparation.  It's gone through a couple of reviews by our Office of General Counsel.  We have shared the text of it with the Department of Transportation.  Before it goes out for public comment, we will have to vet it through our own Department and through ONDCP, because in our sense this also affects what we do with our Federal executive branch agencies and the employees that are covered under our Federal drug-free workplace program.

                That's all I have to say on that subject at this time.

                I'd like to go on to one other thing which is current. You'll hear more about this later on as we go through the presentations.  But MDMA or ecstasy is an issue that is gaining more and more recognition as a drug problem.  It had not gotten our attention well, although it's been around for a long time, but it seems to be emerging from the general background noise and taking a place of concern by people who deal with young folks and raves, and then to a certain degree, as we'll hear later on, in military recruitment.  What we have had in the last week is a situation that was reported in the regional press of MDMA and GHB being linked between two Midway Airline pilots and the death of an aerobic instructor from Gold's Gym in the Raleigh-Durham area, North Carolina.  We have had a meeting with ONDCP and are looking at what we can do to facilitate the development of the appropriate screening and confirmation of technologies to use in our program and in DOT's, and at what we can do with our colleagues in DoD, and the Administrative Office of the U.S. Courts.

                That's all I've got on updates right now.

 

Agenda Item:  DOT UPDATE

 

MR. EDGELL (DOT):  We have not received the name of a new director for our office.  I am the Acting Director.  We do have our nominations down to the Deputy Secretary, but no confirmation there.  Believe it or not, we are a couple of rows down from the Deputy Secretary, but we do believe that we will have a director named in the near future.

                Our previous Director, Mary Bernstein, did leave her legacy.  For any of you that might have been in a coma over the last year or so, we published on December 19, 2000, our 49 CFR Part 40, 118 pages of Federal Register text.  Not all of it is the rule.  The first 60 pages covers the preamble. We had a tremendous amount of comments in making this rule change.  I think that many of the commenters can find their suggestions directly in the rule text, and we hope that those whose suggestions didn't make it to the rule, we've explained adequately in the preamble as to why they were not included.

                The full effect of the rule goes in place on August 1 of this year.  There were a couple of sections dealing with specimen validity testing and the public interest exclusion that were implemented earlier, 30 days after publication, the 30-day period would have been on the 18th of January.

                We held a public meeting.  It was a one-day meeting that we repeated on the second day in February, 21st and 22nd.  We had about 400 people attending the two days and it provided an overview of Part 40.  The comments that we received on the meeting were favorable, and I must say that Part 40 held up quite well.  At the end of each day we had a number of questions and answers, and all but about three out of probably 250 or 300 questions that folks had written on three by five cards we actually were able to find, go directly to Part 40 and find the answers.  So we were very pleased with that.

                Along with Part 40, there are conforming rules that apply to each mode.  Part 40 tried to pull many things into the workplace testing procedures for drug and alcohol for transportation, and to try to do the same thing for all modes of transportation where it was possible to do so.  For example, references to the substance abuse professional were consolidated in Part 40, and there are corresponding conforming motor rules that will edit their rule to.  In this particular example, since Part 40 included the substance abuse professional, the motor rule would strike the reference to the substance abuse professional as far as the duties and simply point to Part 40 for the substance abuse requirements.

                Our plan was to have Part 40 and the conforming motor rules out at the same time.  That did not happen and the conforming motor rules were caught in the new administration's regulatory review plan and were held up.  They were not printed in the Federal Register.  We actually had to draw them back.  Just this week, the Secretary signed those rules and they will now go back to OMB to re-verify that they are non-significant.  It's the first step with the Secretary's signature, but nevertheless it is a first step in that direction.  I'm hoping that will go fairly quickly and we can see something in the Federal Register on these conforming rules in the month of March.

                We also received word of an intent to file suit against the Department and specifically Part 40 from the Air Line Pilots Association and one individual that had been affected by the specimen validity test.  We do not have the actual brief at DOT.  It was just a very general statement that actually included the document.  It was filed in the Ninth Circuit. 

 

MR. STEPHENSON:  What is the date for the MDMA meeting?

 

MS. MILLIARAKIS (ONDCP):  We had an MDMA Inter-Agency Demand Reduction Working Group meeting, and we brought together about ten agencies to talk about the issue of what everybody's doing, what needs to be done, what are the gaps, and who can do what.  Donna Bush will talk about her needs down the road.  There is a need for an assay.  The fact that it's not being tested, that MDMA is not being picked up on the test, is a problem for us.  We are just assembling the information of who's doing what and trying to figure out where to go.

                We also don't have a director.

 

DR. BUSH (HHS):  Just a note on the MDMA issue.  At the risk of putting the target on my back, I will say that I am trying to gather information in the context of our center, Center for Substance Abuse Prevention, any and all information that we can get concerning multiple populations, to include youth, and certainly to include workplace.

                How do you get that information?  Well, we've been surveying the literature.  We're listening well.  We're trying to communicate with those who handle populations like this, such as schools, looking for any kind of survey information.  We're going to be working closely with local universities, high schools, trying to gather what they know about student involvement with MDMA.

                It comes back to the fact that we don't have necessarily good screening tests for focused MDMA analysis.  I have a small amount of money to gather information and to ensure that we can get good information through assay development, through surveying any kind of population that we have concerning the use of this.  We are aware, through our colleagues at DEA, and just read Time Magazine, concerning how much ecstasy is getting into this country.  Someone is consuming this, and yet what we find when we ask focused populations who are doing amphetamine testing:  Well, we don't have an MDMA problem, because we are not seeing it.

                Somehow we are missing this.  We've got to figure out all those answers.  That's my job, and I'm putting an awful lot of energy into it these days.

                Then we find out incidents in the workplace in such a safety-sensitive, public health-sensitive area as airline pilots, in a tragic case as what has been reported in the newspaper.  All the more reason I need answers.  Any and all answers are greatly appreciated.

                I had to follow up Kate's discussion on the MDMA issue.  This is very high on the horizon with all Federal agencies.  We are all concerned. 

 

 

Agenda Item:  GC/MS/MS

                               

MR. CROUCH (DTAB):  First, I want to start with introducing the committee members, because I think there's a broad background here and a lot of expertise.  Mike Baylor, who most of you know, is with RTI and has a long history with this program.  Larry Bowers, you may not know, but he was the director of the former IOC certified laboratory.  He's now with USADA, which is the new U.S. Anti-Doping Association, and he's their key scientist.  Yale Caplan, who you know.  Ken Cole, who's with the military.  Roger Foltz, who probably has added as much mass spectrometry experience to forensic toxicology as anyone.  John Hughes, who's with Agilent, formerly Hewlett Packard, who makes the primary instrument that's currently used in the program.  Aaron Jacobs.  Ian Jardine, who's with Thermoquest Finnegan, and they make instruments, LC's, GC's, ion traps, single stage instruments, tandem instruments, which is what we're going to be talking about.  Christine Moore, who has developed a number of MS assays for testing of hair, primarily chemical ionization.  Lance Presley, who is the RP at LabOne.  They do a lot of saliva or oral fluids testing.  They also use MS/MS.  Barry Sample, who you know.  Mike Schaffer, at Psychemedics, who does a lot of hair testing, and they're using MS/MS.  Matt Slawson, who is the youngest person in the group, but probably has as much experience with these different techniques and testing of these matrices as anyone.  And Donna Bush and Charles LoDico. 

                (Screen.)

                We have communicated by email, etcetera, but we've only had one meeting and that was in September following the last DTAB meeting.  Our agenda basically was to look at chromatography, what's currently there for GC, and to develop criteria for LC.  Then to look at the mass spectrometry techniques which were out there, which include the ion trap, single stage quadrupole, and the MS/MS, and then to look at the ionization modes:  electron ionization -- why are we getting feedback?

                The ionization mode, EI, which is currently being used, electron ionization; chemical ionization, which is currently in the program, but which is seldom used.  With chemical ionization, of course, you can produce positive or negative ions.  The two HPLC ionization modes that are most common, which are electrospray and atmospheric pressure chemical ionization; again, you can form positive or negative ions.

                Then there are various acquisition modes for data collection for these different ionization and MS techniques.  There's selected ion monitoring for some form of monitoring a limited scan.  With MS/MS, or tandem MS, there area number of scan modes, including product ion, precursor ion, neutral loss, etcetera.

                MS to the n, which means doing, in the terminology or the definition that we use, it means doing at least three mass spectral analyses on the drug.  Then of course, full scan.

                (Screen.)

                We came up with a lot of recommendations.  I put them all in this slide.  I'll just give you a minute to go through that and then if there's any discussion we can have it.  In general, we used three or four basic documents to come up with our recommendations.  One was the current regulation or guidelines as they're published in the Federal Register; the current checklist for inspectors; the current guidance document, which explains the questions to the laboratory and to the inspectors; and what's called the NCCLS document, which is the document that Larry Bowers was very instrumental in developing, and it is sort of a similar model in terms of setting criteria for MS confirmation, MS/MS confirmation of drugs of abuse.

                It sort of boiled down to there were three or four questions that were modified.  These are current checklist questions:  Does the laboratory have objective criteria for acceptable chromatography?  The current criteria are that they be objective, there must be some way of measuring peak fronting, peak tailing, that the peak be gaussian, which means essentially that it's symmetrical, that there is an acceptable signal to noise, and that there's reasonable resolution between potentially co-eluting peaks.

                (Screen.)

                Here are the proposed criteria for GC.  Again, these are proposed criteria from the committee to SAMHSA, and obviously what we are here to do today is present what the committee thought should be incorporated and to get people's comments about what's realistic and what can be done.

                (Screen.)

                This is not a criteria, but just general comment.  Generally, the peak shapes are very narrow, no more than 10 seconds at baseline. That some sort of a criteria be set up for minimum resolution between potentially co-eluting peaks, such as peak to valley or valley to peak or an alternate calculation.  Have a method of determining peak symmetry, such as calculating tailing factor, or there are other ways of doing this.  Have a minimum signal to noise of three to one for peak detection.  Have a retention time of plus or minus two percent of a predetermined calibrator or control.  It was suggested that we use the NCCLS standard, which is one percent or plus or minus an absolute 0.2 minutes.

 

DR. JACOBS (AFIP):  Did we put numbers on the tailing factor?

 

MR. CROUCH:  There are some numbers explained in the document itself, and those are acceptable based on the literature and based on those calculations.

 

DR. JACOBS:  Do you know what they are?

 

MR. CROUCH:  I don't remember them offhand.  They're straight out of the literature.  We didn't modify what was in there.

                HPLC criteria.  There were none previously.  Please remember, we can expect broad peaks, more like 20 seconds at baseline.  That there be some way of determining resolution between potentially co-eluting peaks, such as peak to valley, valley to peak, or some reasonable calculation.  Again, that there be a determination of peak symmetry, a minimum signal to noise of three to one.

                (Screen.)

                The same retention time criteria.  There may be some additional needs with HPLC.  We're more likely to be working with areas than heights because of the rawness of the peaks.  That's an area that might need some more discussion, and we may need -- we suggested that we look at retention time, relative retention time criteria, because retention times vary more with LC than they do with GC.

                (Screen.)

                This was another question that was affected, and this was a primary question that was affected.  Are procedures for identifying each analyte and determining its concentration acceptable?  We currently have criteria for electron ionization, so these are criteria for chemical ionization, either SIM with a single-stage instrument or a similar scan function with the ion trap.

The current criteria is that there be at least one or two ions that are monitored.  These are the proposed criteria for positive ion, for positive chemical ionization and for negative ion:  whatever ion you're monitoring, it be diagnostic of the molecule of interest, whether M plus H, M minus H, or M minus HF, or similar.  It was suggested that whenever possible to monitor two ions, and that the ion ratios be calculated and those be within 25 percent of a target in that box, whether that be a quality control sample or a calibrator.

                These are the procedures for identifying an analyte by tandem mass spectrometry, or sometimes it could be called multiple reaction monitoring.  Currently, there are no criteria because this is a new concept, and it was broken down into two conditions.  One is, what is the resolution of Q1, in other words, what are we passing through to the collision cell, and what are we monitoring, going to be able to monitor, in Q3.  If the resolution is less than one, less than unit resolution, the precursor -- I'm sorry, greater than one, greater than one -- that the precursor ion be present in a signal to noise of at least three to one, two products and one ratio be monitored, and the ratio be within 25 percent of a predetermined calibrator control in the batch, and these ions that are monitored, these product ions that are monitored, not be from the derivatizing agent.

 

DR. SMITH (DTAB):  Do you have that same criteria in just regular mass spectrometry or the same analysis?

 

MR. CROUCH:  No, I'm not sure if it's in there now, but it's a good thing to include.  This comment would have avoided a lot of problems with some analysis in the past.

                (Screen.)

                What if the resolution is less than one on Q1?  Daily, there should be some sort of demonstration that the resolution on the first quadrupole is actually less than one.  I think we also made the criteria that resolution on Q1 never be greater than ten.  That was a really bad idea.  The precursor then is not required in the product ion scan, but it was suggested that there be two products monitored and at least one ratio, and that that ratio be within 25 percent of a predesignated calibrator control, and again the ions that are formed are not diagnostic of the derivatizing agent, but are diagnostic of the analyte itself.

                (Screen.)

                This is the same question.  I know from the hair group they had proposed product ion ratios of 30 percent.  What's suggested in NCCLS and what the committee suggested was 25 percent.  Again, none of this is cut in stone and I really would solicit your input, because what we don't want to do is write something that's not doable or suggest something that's not doable.

Currently there are no criteria for MS to the n, and this would be more than two mass spectral analyses.  About the only instrument that I'm aware of that does this now is the ion trap.  That's commercially available.  The proposed criteria would be that a single product would be okay, one product ion would be sufficient, the precursor need not be present in the product ion scan.

                What about HPLC?  Well, the proposed criteria are very similar to those of CI, EI, the ion trap, MS and MS to the n, except for these differences.  That there be a rationale for whatever is selected, M plus H or M minus H or an adduct such as sodium formate or an acetate.  Must demonstrate that the adduct is formed consistently and predictably, because if it isn't then everything is going to change throughout the run.  If ion ratios are formed, they should be within plus or minus 25 percent, and that the laboratory have some sort of written procedure to assess ion suppression.  Ion suppression is a phenomenon that occurs usually with electrospray ionization, where there is a limited capacity for the ion source to make ions.  If there's a large background in the analysis, it can threshold out the drug you're looking for or for the internal standard.  I'm sure that's clear to everyone.

                (Screen.)

                Does the laboratory ensure that the internal standard response for each calibrator, control, and donor sample is appropriate?  All calibrators, controls, and donor samples should have an internal standard intensity of at least 50 percent of that of a predetermined calibrator control in the sample.  Again, this is to look at things like ion suppression, poor extraction or recovery, poor injection, lack of derivatization, something like that.   The method for determining the minimum recovery of the internal standard needs to be written, obviously, in the SOP.

 

DR. SAMPLE (DTAB):  Again, we're looking at this for all modes?

 

MR. CROUCH:  Right.

 

DR. SAMPLE:  Not just for the new techniques, but even existing GC/MS?

 

MR. CROUCH:  That's right. That would be across the board for confirmation, regardless of what technique is used.

                (Screen.)

                Does the laboratory monitor and evaluate the performance of all analytical instruments each day of use?  Many of the newer instruments, particularly the LC/MS's, the manufacturers do not suggest or require that they be tuned every day.  In lieu of a tune, many laboratories will perform what's called an optimization, which basically sets the instrument up and optimizes the performance for that particular analyte analysis.  That there be some way of each day monitoring the system use or system capability for handling this particular assay.  This must be either through optimization or through analyzing an internal -- excuse me -- an unextracted standard that demonstrates the instrument's professional, and if an optimization is done that it should be traceable back to a tune.

                (Screen.)

                Does the laboratory use full scan?  This is the other major question that there had to be some discussion about for each of the different modes of analysis.  Currently with EI, either with the ion trap or single stage instrument, it's required that they have a fit of 950 or greater, although this is somewhat grey, and that a minimum number of ions be monitored.  What's proposed is that a scan be run from 40 to at least 100 mass to charge above the expected molecular mass, all ions with a relative intensity of greater than 20 percent should be present in the donor sample, that there be no ions greater than the molecular ion cluster, not a bunch of ions that are high molecular weight that don't belong to this analyte.  No ions that are present at a relative intensity of greater than 50 percent, that are not found in the reference, of course.  That identification be by match or fit, and the criteria be greater than or equal to 950.  I believe we proposed quantitation beyond a preselected ion.  I don't know about TIC. 

                In the NCCLS document it suggests that if the match criteria, the fit is less than 950, but greater than 750, that someone, such as a certifying scientist or a responsible person, could review those data and determine whether they're acceptable or not.  I don't know if this is something we want to put in here or we want to leave out, or no one cares.

 

DR. SAMPLE:  Weren't some of those criteria predicated on the algorithm that was used as well?

 

MR. CROUCH:  That's a good comment, Barry, because what's very difficult here is there are different algorithms to calculate this fit or match, and on most of those algorithms 950 would be a good match between this unknown drug and one in the reference library.  In some libraries, a match of 0.8 is pretty good.  Do you have a comment on that?

 

DR. COSTANTINO (AML):  That was my comment also, the difference in the algorithms and the instrument.  In addition, I think the laboratories need to have a set objective criteria for evaluating the EIC, instead of leaving it to the discretion.

 

DR. ISENSCHMID (DTAB):  I was concerned about the algorithms, because some instruments, a 50 percent match is great between what it is trying to match.

 

DR. MITCHELL (RTI):  Did you address the type of library that's going to be used for the match?  Has that criteria been established?

 

MR. CROUCH:  We didn't as such.  We said it was a reference library that's in the instrument or one that's generated based on a calibrator or control predetermined in that batch.

 

DR. SAMPLE:  I thought that we had to update the library each day in order to achieve those numbers.  That was one of the criteria for doing it.

 

MR. CROUCH:  I don't remember that.  I'm sure that's not in what I turned in, but if you want to make that up and propose it.

 

DR. MITCHELL:  I mean, is a library of one sufficient?

 

DR. SMITH:  That was my question.  Are you creating this library by using pure drugs?  Are you creating it yourself or is this an historical library?

MR. CROUCH:  I'm not going to do this.  Let me just make this clear.  I'm just developing these.  I'm not going to be responsible for them.  The thought was that you could do it either way.  You had a preexisting library and you matched a calibrator or control in your batch to that library, or you base it on the calibrator control in your batch.  That was the thought.  Obviously, that's why we're here, was to get comments.

 

DR. MITCHELL:  If you have a library that consists of only one compound which happens to be in your library and that's the drug you are looking for, you would naturally expect very high resolution.

 

DR. SAMPLE:  Not necessarily.

 

DR. MITCHELL:  Normally you do, though I don't think that it has the same strength of the larger library of similar compounds.

 

MR. CROUCH:  It certainly wouldn't give you very much qualitative information about potential interference.

 

DR. MITCHELL:  That's correct.

 

MR. CROUCH:  Because you either get a hit for the compound of interest or you'll get no hit.

 

DR. SAMPLE:  But it does not affect the match quality.  Whether it's one compound or a thousand compounds, the match quality is the match quality.  I think if you're going to target 950, unless you have an identically tuned instrument, you will be very hard-pressed to achieve that type of match quality unless you are creating your laboratory every day.  Not that it's impossible, but you could be very hard-pressed.

 

MR. CROUCH:  Are you suggesting we're going to have to review all of them?

 

DR. SAMPLE:  If you're going to use an historical library, yes.

 

MR. CROUCH:  I guess we're open for comments in terms of a reasonable number, or do we want to say this is going to vary based on the match criteria, whether it's a forward search, backward search, or whatever it is?

 

DR. ISENSCHMID:   I think you really have to have very good criteria of how you're going to determine whether something's a match, and to simply put a number on it is probably not the best way to go because it is going to change from tune to tune, background effects, all sorts of different things that full scan will create that you don't normally see in a SIM. It is going to need some very well-defined points that you would decide whether or not that's a match or not.

 

MR. LoDICO (HHS):  Could you look at that as relative to 100 percent for that day?  If you establish this range, it would be like plus or minus 20 percent of that, typically that you do with any calibrator control for that particular day relative to 100.

 

DR. SAMPLE:  Relative to a relative number.

 

MR. LoDICO:  Exactly.

 

DR. SAMPLE:  That would be another approach that I think we discussed at the meeting.

 

MS. EHORN (Varian):  Barry's right.  If you are running a daily continuing calibration checkpoint, that should go in your reference library.  That way you will be able to maintain that 950 or better test.

 

MR. CROUCH:  Do you think 950 is a reasonable number with most instruments, given that there's ion trap here?

 

MS. EHORN:  As long as it's put in the reference library daily when you do your continuing calibration checks daily.

 

MR. CROUCH:  Don't you think that that somehow has to link back to a library standard, though?  Because if you put the wrong analyte in today, you could get a great match, you could get great precision, but qualitatively it's wrong.

 

MS. EHORN:  Initially, when you build your libraries they must be matched to known standards, and when you're doing a calibration checkpoint you know that that's a standard following the same extraction procedures and prep procedures.

 

MR. CROUCH:  You would have a reference library, if I understand this right, and then each day you would match to that reference library, but you would use that match to set up the daily criteria?

 

MS. EHORN:  Correct, if you're going to make it the 950 or better.

 

MR. CROUCH:  Right.

 

DR. SAMPLE:  Yes, I think so, because I'm trying to ask questions at the same time.  The criteria we had for absence of ions above the molecular cluster when we were doing the higher mass search, was that an absolute absence or was it an absence of ions that would explain ions or clusters at lower mass through logical losses?

 

MR. CROUCH:  If that was a logical question, I think it's the latter.  I think the absence of ions that you can't explain came from this analyte or a derivative of this analyte.

 

DR. SAMPLE:  But you could have ions higher than the molecular ion cluster?

 

MR. CROUCH:  Right.

 

DR. SAMPLE:  As the ions that you're contributing to the compound would not logically result from those ions that you're seeing at a higher mass.  For example, if you're doing a TMS derivative and you see another cluster 15 mass units higher.

 

MR. CROUCH:  I think what you're proposing is very logical with CI, where you might get an ammonia adduct, or with LC, if I'm thinking along the same lines, where you might get a sodium adduct or something.  The molecular ion is 23 less than what you're seeing.  Is that what you're saying?

 

DR. SAMPLE:  No.  What I'm saying is, if you see higher mass ions and a normal loss, an N minus 15 or an N minus 90, if you're seeing something that's 90 mass units higher or 15 mass units higher than what your proposed molecular ion cluster is, then you might be concerned that that molecular ion cluster really is not the molecular ion cluster for the compound of interest; it's due to the loss, potential loss from a higher mass compound that's confusing the spectrum.

 

MR. CROUCH:  It certainly is confusing.  I'm not sure I follow you.  I think one of the statements was that there be a rational explanation for the ions that you see.

 

DR. SAMPLE:  No.  I guess what I'm saying is you don't want a reason for those ions that you see if you see a higher mass, because you might have an interference.  There might be something underlying.

 

MR. CROUCH:  Wouldn't that be ions above what you would expect to see from the molecular ion cluster?

 

DR. SAMPLE:  Maybe, maybe not.

 

MR. CROUCH:  I guess maybe we need to discuss this.  I'm not sure I understand it.

 

DR. SAMPLE:  We can talk about it later.

 

MR. CROUCH:  We can certainly make a paragraph about that.  All right, full-scale CI with single stage instrument or with an ion trap.  Essentially, 1 to 7 that we discussed are the same.  I think what's in the document now is that we start at mass 40, mass and charge 40, but I'm not sure that makes a lot of sense with chemical ionization.  We're going to have to start higher.  I don't know; is 75 reasonable?

 

MS. EHORN:  Actually, it does make sense if you're looking at complete unknowns, if you were trying to run an analysis on a compound that you have no clue what's there.  If you have a targeted analysis, you're looking for specific compounds, then it would make sense to start your acquisition a little bit higher within that range of that targeted analyte.

 

MR. CROUCH:  Yes, but even if you go to 40 you stand some chance of running into rating gas.  I think, wherever we start, we need to review this.  Wherever we do this, it has to be above the rating gas background, and 40 doesn't necessarily do that.

                (Screen.)

                And that as many ions as possible be monitored.

                (Screen.)

                I don't really know quantitatively what that means.

                (Screen.)

                What about full scan using tandem MS or MS to the n?  Currently, there are no criteria.  Obviously, you can use these instruments, scanning a single quadrupole, either Q1 or Q3.  If it's a tandem instrument, then the criteria that we just discussed would be applicable.  If you're doing a product ion scan, again scan from a minimum of 40 to 100 mass to charge units above the expected molecular mass.  All ions of relative abundance greater than or equal to 20 percent should be present, no ions greater than the molecular ion cluster.

                (Screen.)

                Again, the 50 percent criterion that you do not have interference at or above 50 percent.  A fit criteria; the precursor be present at a signal to noise of at least three to one.  If you're doing MS to the n, the precursor need not be present.  Quantitation beyond a preselected ion, and then the same match criteria.

                (Screen.)

                What about HPLC MS, ion trap, or tandem MS, or MS to the n?  Obviously, there currently aren't any criteria.  What we would propose is similar criteria to what's used for GC, with probably the following changes.  If it's an EI or some kind of fragment or voltages used to fragment the molecule, that it be the same as the electron ionization criteria.  If it's APCI or electrospray on LC/MS, that it be similar to chemical ionization.  If it's tandem MS or MS to the n, that it be similar to what we just talked about for GC.

                I just picked out those three or four questions that I thought were most impacted by the proposed changes.  We're willing to discuss any of the things we talked about, and certainly one of the reasons why we're here is to solicit input.

 

MS. EHORN:  Are there any criteria on the ion ratios of the MS/MS results if you're looking at two or three ions from the MS/MS?

 

MR. CROUCH:  Yes, the criteria was plus or minus 25 percent.

 

MR. IRVING (Psychemedics):  Was any thought given to, when you're looking at multiple compounds such as the drug plus its metabolites, if it has any effect?

 

MR. CROUCH:  I think John's question was, if there's a difference between looking at a single analyte and saying, it's here, we're going to report it, there's more confidence if you have the drug and its metabolite present, should the criteria be slightly different for MS acceptability.  We did not look at that issue.  Do you have any proposal?

 

MR. IRVING:  Right now we're waiting for your report to see how it would fit in with the alternate matrix of tests.  In the case of cocaine, for instance, the committee has recommended that with cocaine a minimum of one of the other metabolites has to be there in order to call the sample positive, either BE or CE or norcocaine.  In that case, you'd get two parents.

                In the case of LC tandem mass spec, you get two parents and two daughters, one product and one daughter from each of those entities; and how does that affect the criteria versus what you would propose?

                 One of the problems is, in order to get the metabolite you have to go to the lower levels.  We feel in the case of the alternate matrix the presence of the metabolite often eliminates the problems we have with environmental contamination, which is not a problem you have in the urine matrix.  We feel it's a necessity to address that.

 

MR. CROUCH:  I think John has a good point, because with urine testing primarily we're looking for one metabolite, sometimes we're looking for a couple of analytes, but on some of the alternate matrices we may find the parent drug and a metabolite.  That alone is some level of confidence that the drug is there, and should there be sort of another level of acceptance criteria based on the fact that you found two, the drug and metabolite, in the specimen that's diagnostic of the use of marijuana or cocaine or whatever the drug may be.  That's a good point.  Unfortunately, that's back to the drawing board.

 

MR. IRVING:   One of the things I don't know if it was discussed or not, was the possibility of looking at two different derivatives of the same compound, so you wind up in the case of tandem aspect getting two parents and the corresponding daughters, as opposed to trying to get two daughters of the same parent.  Again, it would not be a result of analytes, but derivatives.

 

MR. CROUCH:  Is this a genetic question?

 

MR. IRVING:  What I mean is parents, THC, but it could be a generic question.

 

MR. CROUCH:  I understand what you're saying.  I think what John's saying is sometimes you do two analyses and you have one precursor and one product ion for one analysis and one precursor and one product ion from another analysis.  Is that equivalent to having a single analysis with one precursor and two products?  I guess that's a question we have to discuss.  It's certainly more diagnostic than a single chemical ionization analysis.

 

DR. SAMPLE:  How does one handle that with quantitative cutoffs?

 

MR. IRVING:  You'd do things -- one, you would have to be above the cutoff value, and two, you would ratio them and give it some kind of a ratio with the two daughters.

 

DR. SAMPLE:  To verify that there's some quantitative agreement between the two numbers.

 

MR. IRVING:  Right.  In order to get down to the numbers that we're trying to get down to in hair testing, we have to do some things that are really innovative, where there are much lower concentrations than in a urine sample.

MR. CROUCH:  Did I hear you right, you say you do a ratio of the two products from the different injections?

 

MR. IRVING:  You do a ratio of the -- you're doing the internal standard against the daughter, the daughter from the internal standard against the corresponding daughter from one derivative, and you're derivatizing the internal standard and the compound you're looking for in two different ways and the ratio of those two, the ratio, basically quantitation.

 

MR. CROUCH:    If you have comments, they can go to Charlie or they can go to me.  I think it would be helpful if they're in writing, primarily so I don't get them confused, because some of these things are fairly esoteric, and if I hear them verbally and I change them then they become something that they weren't meant to be.

 

MR. LoDICO:  I agree with Denny in the sense that we're tentatively going to have another meeting after this Board meeting to at least consolidate the comments we receive in our office, discuss them, and then create that final draft.

 

MR. IRVING:  One of the things that I might suggest for each of the matrices, at least for hair testing, we're now sitting with a set of PT samples that we're analyzing, but maybe going into this coming meeting we might be able to ask each of the labs that's involved to give you their best effort, to give you the actual data, once we've done the stuff, to see if you have a different proposal, to use these same PT samples so everybody's on the same footing and you can see what these capabilities are for each of these methodologies.

 

MR. CROUCH:  That's probably a good idea.

 

MR. IRVING:  You get more samples, John, in case people need it.

 

MR. CROUCH:  What I would need is the method to accompany the PT result, to see who's compliant or who's using these sets of criteria or how they vary from what's proposed here.  That would be true of all the matrices, really, the oral fluids and the sweat testing, because if it's not similar to this, if it's vastly different, then we probably need to know about that right now so it can be discussed whether it's okay.

 

MR. STEPHENSON:  For purposes of this public meeting, I want to make sure that anyone who's in attendance here recognizes that public comments are being requested by Denny at this point in time.  You don't have to wait for a public comment period at the end of the session.  If there are any other public comments that represent a different approach or issues that you'd like to raise on this topic, we can certainly do it at this time.  I want to make sure that there's an understanding for everybody's who's in attendance that this is not a closed discussion.

                Second, I think it would be helpful for purposes of the documentation that comes out of this session to make sure that this goes up on our web site as early as possible.  I'd like to make sure you really fine-tune this section of the comments to capture this.  I'd like to see if we can get that somehow up on our web site, even though this is a preliminary document and the responses, so it invites a broader input from other parties.  Or you can reference it to people who might not have been able to attend the meeting, or you might want to have some input just by saying, go to our web site and pick up this document and look.

 

MR. CROUCH:  Actually, Charlie and I talked about that this morning and what we propose to do is to put together an outline, not all the detail that's in the current document, but an outline with these sorts of bullets, include what's been discussed today, and I guess Charlie would get that on the web site for comments.  The current document went from about 18 pages to 35 and it looks like that recommendations slide.  I don't think you want all of that, because this is the important thing, just the criteria.

 

MR. STEPHENSON:  Denny, I'd like to compliment you and your committee for taking a look at not only the division of what you think needs to be there, but to help move us in a direction that's scientifically challenging and the kind of thing that we need to be working towards.

                I think you're seeing a practical interface with the hair testing working group needs.  It's my understanding that one of the hair testing working group members has boarded their proprietary method to the hair testing working group for review, which is certainly an opening of the process.  It's the kind of thing we're going to need to do more of, and it will be necessary before we actually implement any of these procedures to make sure that the applications and how they would work in any kind of an inspection process is well understood and well documented in what we're looking at.  Thank you for taking the leadership on that.

 

 

Agenda Item: Military Testing for MDMA

 

DR. JACOBS (AFIP): (Music.) I have a little music here for the rave party.  This group has been around for about 20 years.  It's called Underworld and they have actually been doing this for 20 years for rave parties.  It started in England.  I think someone told me they have about 15 albums out, and it just goes on and on and on.  You want to drive people crazy?  Some of these go on for 11 or 12 minutes.

                I'll turn it down.

                These are some numbers that we have had for the last three years.  You can see on the top, amphetamine positives and MDMA positives.  You can see real fast on the second line, 93, 432, 1,070.  I'll go into some projected numbers soon.  You can see what percent it was of the positive distribution in the middle.  That's of the total positives we had.

                MDA went from 0.6 to 2.6 to 5.6 as a percent of the total positives in the military.  At the bottom you can see that it is approaching about 0.1 percent in FY '00.

                (Screen.)

                This is another way of showing it; same numbers, no different.  The same data here, but the white you can see is going up.

 

MR. CROUCH:  What sort of concentration cutoffs are you using with the MDMA?

 

DR. JACOBS:  500 screen, 500 confirm.  If it does not screen with the methamphetamine or agent that we use, of course we don't go.  Every sample that screens positive in the military is then tested for methamphetamine, MDA, MDMA, and MDEA.

 

DR. WALSH (Walsh Group):  What's your screening reagent?

 

DR. SMITH (DTAB):  It's a Roche methamphetamine.  Can I make a few comments about that, if I can interrupt Aaron for just a moment?  The percent cross-reactivity is low on that reagent, but it's a little deceptive.  If there's MDMA present you get a synergistic effect.  I think through this we don't know what the cross-reactivity is, but it is low.  We don't know.

 

DR. WALSH:  I think the package insert says 2,000 for MDMA.

 

DR. JACOBS:  I think that's right.  Again, we are talking pure MDMA.  Every time we have it there's something else there.

 

DR. BUSH:   Isn't there also a criteria for a second initial test, a second screen, which is done using a different reagent in some labs that do that?

 

DR. SMITH:  I don't think so.  I think the second test that you're talking about is the same reagent and there's an adjunct test that they can do to rule out the alpha hydroxy compound, which doesn't affect this analysis.  This is positive, and the second test using the same reagent.

 

DR. JACOBS:  Many of the labs were doing 3 screens.

 

DR. BUSH:  I understand, but I believe there's an EMIT reagent that's being used by some labs also.  There's several immunoassays that may be used initially as kind of a gating thing, where positive, then positive, then positive again.  It moves on.  It's to screen out a lot of the cold medications and the pseudoephedrine type stuff.  But being unaware, as we are necessarily, of the cross-reactivity of MDMA, when you do that you're layering on gates that we don't necessarily know.

 

DR. JACOBS:  You can see here when we plot the concentrations of the positives and the number, the number at that range, you can see the 500 to 1,000, 1,000 to1500. You see the MDMA, you would probably expect this range in here are some we would be missing.  You would expect them to go up, just like methamphetamine.

                (Screen.)

                Here's another way of plotting it.  This is just the log frequency.  You can see that MDMA is above all the way down here, but it drops off at about 6,000, 5,000, somewhere in there.

 

DR. SAMPLE:  Isn't that a function of your screening cutoff in this case?

 

DR. JACOBS:  Yes.  The screening reagent there -- you're exactly right.  It shows that it doesn't look like there's any relationship at all between the methamphetamine that is in the sample and the MDMA that's in the same sample.  These are only samples that had both in it, plotted methamphetamine from highest to lowest, and put the MDMA that was in each sample in there.  Anybody wants to tell me what that means, I'll listen.  It doesn't look like there's any relationship between the methamphetamine and the MDMA that are in the same sample.

                (Screen.)

                Another way of looking and seeing that it looks like the dropoff here, the screening reagent isn't picking them up, again taking the methamphetamine and the MDMA out.

                (Screen.)

                This is projected, extrapolated MDMA values, taking the difference between in this range right here, the MDMA, and the methamphetamine and projecting them all the way out.  That would be the projected number.  We might have been missing as many as 1,500 samples that could be positive if this thought pattern works.

 

MR. CROUCH:  I'm not sure that's absolutely true, because the MDMA doses are five to ten times what a methamphetamine dose would be.  You would expect much higher concentrations in the urine.  You might decrease the detection time by lowering the immunoassay cutoff, but I'm not so sure that it's going to reflect the frequency of use. 

 

DR. JACOBS:   I would also think that we're probably missing some in there.  Even though they took a higher dose, they're still done.  You might be right, it might be the window we would be extending.

 

MR. IRVING:  What we're seeing in hair samples, because the antibody has a little better than one to one cross-reactivity to MDMA, concentrations of MDMA parallel the concentrations of methamphetamine, at least within the hair.  There may be some question.

 

DR. JACOBS:  What reagent are you using?

 

MR. IRVING:  It's a homemade reagent.   A home brew.  It's about a 1.2 cross-reactivity.

 

DR. JACOBS:  (Screen)

                These are the numbers we have seen this year.  The things of interest are, if you look at Fort Meade, which is on the East Coast, these are East Coast samples and Europe, you can see the MDMA positives so far through the end of February compared to the methamphetamine.  If you compare that to Tripler, which is West Coast and the Pacific, you can see their methamphetamine positives compared to their MDMA positives.  That is the two Army sites.

                If you look at Jacksonville, which is Navy, you can see its MDMA compared to its methamphetamine, they are East Coast.  The San Diego lab, West Coast methamphetamine compared to its MDMA.  Again, if we take these numbers and project it out through the end of the year, guessing as best we can, we may have, instead of a thousand, as many as 1,800 this year.  If we look at that, and then we add on the numbers of those we missed using the same formula for last year, we are up around 3,000.

 

DR. HUESTIS:  What types of pills have been seized in military investigations?  Are there particulars?

 

DR. JACOBS:  I don't know the particulars.  I think it's all over.

 

DR. HUESTIS:  Is it MDMA alone or mixed?

 

DR. JACOBS:  Just about including everything, as we heard last week, all the reagents, anything else they can put in that's going to give them an effect. 

                (Screen.)

                Of course, we can buy the Ecstasy test kits over the Internet now, for those of you who want to test your drugs before you take them.  I don't know how accurate they are or whether it really tells you, is it Ecstasy, or does it tell you is there anything else in there, but you can feel better when you take it.

 

MR. STEPHENSON:  Would you, from the DOD-armed services perspective, be willing to explore with us how we can do some further work, whether it's with new screening assay reagent kits or with human study subjects, and so on?

 

DR. JACOBS:  What we're going to do is try and get some of the samples that were depressed but were not positive, by screening from Fort Meade, because it closed, take them to AFIP and look at some other reagents.  We have another reagent that may be much better at detecting.  We will rescreen.  We will also send them to confirm anyway, say 500 or so, and see what we can see in those that did not screen positive, but were depressed.

 

MR. STEPHENSON:  That would be very informative, very helpful for us.  I think the idea is that we all recognize this is something that we can work on together.  We don't have to come up with independent answers or different ones.  We need to come up with something that works, that works soon.

 

MR. IRVING:  Have you looked at EPC's reagent?  I think the cross-reactivity there is much greater, is much closer to meth.

 

DR. JACOBS:  I will write that down and look.  I think we were first going to look at the reagent that the British are using for their military, that they have had very good results, they say, for finding MDMA.  Is it Cedia?

 

DR. SMITH:  Cedia.

 

DR. JACOBS:  But we'll probably be looking at a lot of reagents.

 

DR WALSH:   I just wanted to mention, I was involved in Project Rosita, which was an EU-funded eight-nation study of drugs and driving, where they were looking at different matrices with onsite testing for police officers.  We saw a lot of MDMA, and I would suggest you make contact with the typical suspects, Manfred Muller in Germany and Nils Sammon in Belgium and Kerl Lusen in Finland, Bjorg Morland in Norway, to see what they're doing, because they picked up a lot of MDMA.

                An interesting sidelight in terms of looking at different matrices: Because Ecstasy is a problem in Europe, the Belgians specifically were doing a number of saliva assays and it became very evident what people were on Ecstasy, because they could not get a saliva sample out of people; they were bone dry.  It sort of points out the need for having different types of assays available.

 

DR. JACOBS:  We are starting to have more court cases also on Ecstasy.  For some reason, for the first few years there were none.  Apparently, they were in such a loving, partying mood, they didn't really care.  It seems now that we are having more of those cases coming.

 

MR. STEPHENSON:   I know this was one of those things of sharing that sometimes doesn't happen.  We are very appreciative that you were willing to share your experience with us.  We know that that often is, no good deed will go unpunished, but in this case I'm glad that it's out in the open.  I think from an ONDCP perspective this could be helpful, and maybe even a comment that Mike Walsh had made, something that we could look at to see how we might be able to do an outreach with some of our European counterparts.

 

 

DR. HUESTIS (NIDA):  We have lots of contacts with the Europeans.  They have been asking for more than four to five years for manufacturers who produce MDMA kits, and the European marketplace was considered so much smaller than the U.S., they couldn't get the manufacturers to even address their issues and problems.  Unfortunately, now that we're having problems, they are hopeful that they can get the kits there, so they have the same availability that we have other than the Cedia that does have the assay, I don't think as yet approved in this country.  They have the same mix available, the same problems.

 

DR. BUSH:  I'll tell you candidly in the open session, many times when you do speak about new test kits and new test ventures, people do want to know what kind of payback they will get for the development.  Manufacturers definitely want to know what kind of payback is coming, and that's a fact.

 

Agenda Item: Alternative Specimen Testing Industry Group Presentations

 

Hair Testing Working Group

 

MR. IRVING:  We had a meeting of the hair group that was bigger than the first two meetings we had.  We had much bigger attendance, maybe because we had it in Vegas.  That may have been part of the situation.  It also may have helped the cooperation between the groups, but all but one of the laboratories that was involved in the PT program, at least in this country, attended.  We had people from RTI.   For an overwhelming number of issues, we had very good consensus.  It was a very good working meeting, much better than we've had in the past, and with one or two exceptions we came away with a pretty good consensus.

                If you can go to page 3.  There are 16 of 15 pages in this document; don't ask me why. 

                (Screen.)

                We were asked to address eight questions from DTAB.  The first question was based on how much hair needed to be collected.  In the past we vacillated back and forth between some different quantities, and the last draft that we sent in was 100 milligrams, and they wanted to know whether this was sufficient to perform the initial testing, the confirmation testing, and repeat analysis, and also if you had multiple positives.

                We needed to determine if this was adequate and how we would describe the amount of hair to a collector.  We first decided that for routine collections head hair is the primary sample of choice.  One of the other questions we were asked to address later on was whether we could collect hair from another source in the body.  We were talking mostly about hair testing from the head to start off with.

                We agreed that the length had to be equal to or greater than one centimeter -- trying to cut anything less than that, trying to do shavings, would not work -- and that the sample that we would test in the laboratory would be cut down to a length of 3.9 centimeters.  That averages about -- that's approximately a three-month growth.  An individual would have his hair collected in the field.

                There was some discussion of having the hair cut to this length at the collection site, but it was decided that the hair would be collected close to the scalp, put into a collection device to show which is the root, and the entire amount sent to the laboratory; the laboratory would be the one what actually cut it down to 3.9 centimeters.

                This information, the length (indicating) -- that looks like I scalped somebody.

                We were asked to provide some information to the Board as to what samples need to look like, how we would tell a collector how to collect hair, and giving an example of a 3.9 centimeter length of hair wasn't sufficient.  We decided that we really needed men's hair around 2 centimeters, so we're going to have to give them samples of, if a hair is this long, this is the approximate diameter of the hair that needs to be collected, if it's longer than 3.9 this is the amount of hair that needs to be collected.

                There was some talk about -- one of the questions was how many strands does this involve.  You'd be surprised how many strands there are in just 10 centimeters of hair, somewhere between 30 to 80 strands of hair within a 10 milligram length of hair, amount of hair.

                We decided we could devise materials such as this -- and I'll pass this around -- that we can pass to the collectors to show them what it actually looks like.  We could also provide them pictures of what it looks like if it was spread out, so that as part of the collection drive to the inspectors this would be pretty well defined.

                We talked about collecting from other sites on the body.  We decided that the head was the obvious place of choice.  However, a lot of people will shave their heads and some people don't have a head of hair.  In that case, we decided that you could go to other sites of the body, such as underarms, chest, back, leg, or arm.  We also decided that pubic hairs should not be collected for the federal program.

                (Screen.)

                I think that's really about it.  Once the guide for the collection sites was developed, we would have the material to send out to them.

 

DR. HUESTIS:  Did you specify where on the head?  The posterior vertex?

 

MR. IRVING:  The posterior vertex was the first choice, but it doesn't have to be from there.  But the length of hair that's being tested by the laboratory needs to be specified back to the medical review officer so that he can evaluate that as part of the evaluation process, knowing if it's a one-month period, two-month period, three-month period of time.

 

DR. SAMPLE:  If the 3.9 centimeter length approximates 90 days, what period of time does the minimal one-centimeter length approximates?

 

MR. IRVING:  About a month.  You're actually taking about a week to ten days to grow out from the scalp, out above the scalp level, to a cutting section where you can actually cut it.  You're actually talking about a month period, but add another week from the time of the actual last drug use.

 

QUESTION:  John, usually underarm hair, it's very difficult to estimate how much time has elapsed on normal length.

 

MR. IRVING:  Underarm, actually with any of the body hair, you're now talking about a longer dormancy period, so that we as a rule of thumb say somewhere between a 5 to 6-month time period that you're actually evaluating, I mean when you're talking about any sort of body hair.  There's no attempt to try to cut body hair to any length.  There's no way of lining it up and cutting it to a particular length.  Most body hair tends to be somewhat curly also.

 

DR. HUESTIS:  Why did you decide on underarm hair rather than regular arm hair, where it would be much less invasive?

 

MR. IRVING:  It's a little easier to harvest underarm hair than arm hair.  Arm hair, when you take a look at the quantity, you're talking about a lot.  Some people need to grow a third arm to give us enough hair.

 

DR. SAMPLE:  Getting back again to the different lengths, have there ever been any issues with fairness raised on the part of employees because of the different lengths of hair being harvested, a three-month window versus a one-month window, and if so how has that been addressed?

 

MR. IRVING:  We have had various challenges in court and nothing has -- we won all the cases as far as the length.

 

MR. THISTLE (Psychemedics):  I just want to address that.  I would think that that issue would be no different than a fairness issue with regard to a guy that's doing heroin with a urine test as opposed to someone doing marijuana with a urine test.  You have detection time differences of different ratios that are far different than what he is describing here.  You have those detection time issues.  It hasn't been an issue because you're talking about people that are doing the drugs and they have an explanation for what happens in between:  I was doing drugs and last month I was rehabilitated.  They have the opportunity to bring in documentation to that effect.  It hasn't really been an issue.

 

MR. EDGELL:   Hair on other parts of the body, underarm hair, does that grow at the same rate?

 

MR. IRVING:  It grows at the same rate, but it's maintained on the body for a longer period of time.  They have a higher proportion of dormant hair, so that it represents drug use further back in history.

 

MR. EDGELL:  So a 3.9 centimeter length of underarm hair could give you --

 

MR. IRVING:  Five to six months.

 

MR. EDGELL:  A longer window?

 

MR. IRVING:  Yes, that's correct.

 

MR. EDGELL:  If somebody defined current drug use as within 90 days and someone no longer engaged in current drug use, such as an ADA issue as over 90 days -- I'm just picking a number here.

 

MR. IRVING:  As far as the ADA issues, we've had successful prosecutions in federal court -- no.  State court?  Not court.

 

MR. THISTLE:  It hasn't been an ADA issue, for the same reason it's not an ADA issue with hair.  If someone brings up the issue that they have been rehabilitated, they bring in documentation to that effect.  We think if it's a head hair sample we can screen the sample; if it's a body hair sample -- I mean, sorry, we can segment the sample.  If it's a body hair sample and the time frame is unknown, you're talking about a person who's admitting drug use at that point and just the issue becomes a question whether or not, first of all, they were covered under the ADA to begin with, where they were substantially limited in a major life activity.  That's not casual drug user, that's not a moderate drug user.  They may not even be a heavy drug user.  It is someone who's substantially limited:  Can they walk, can they see, can they breathe, can they talk?

                That's a high standard to meet.  Basically, you're not employable at that point.  I realize it's a Catch-22 for the user, but that's the way the law is written; difficult standards to come under the ADA as a drug user unless you are an addict or perhaps having a record of being substantially impaired, a very difficult standard to meet.  We've done millions of these tests without an ADA issue.

 

MR. EDGELL:  Did the issues address the specific amount of time?  I mean, the ADA doesn't address an individual no longer engaging in current drug use, and I'm not sure that that has a specific time on it.  But the court cases, the two court cases, did they examine a particular period of time?

MR. THISTLE:   No.  The ADA, in fact, in the technical assistance manual it says it's to be determined on a case by case basis.  There have been court decisions that have had long time periods.  For instance, if there was an undercover agent who monitored someone doing drugs and the investigation went on for months, you could take an employment action at the end of that time and not run into any ADA issues because you're taking an action based on something that occurred earlier.  It does not necessarily put you under the ADA.

 

MR. EDGELL:  If you could say that a particular length of hair on average represents 90 days and it's given, arguable, that if you used drugs within the last 90 days, you are engaged in current drug use.  But let's say that you have the same length of hair that goes back six months and you're saying, well, this is what I did six months ago, however I have changed and this does not represent what I am doing today.  That's the only point I'm trying to make.

 

MR. THISTLE:  There is no 90-day magic number there.  Hair testing goes back for months.  In fact, if you have a urine testing program, you have a pre-employment urine testing program, most companies have a six month or one year no-reapply rule.  You can't go in, take a urine test, fail it, and then come back the next day or two days later and take another urine test.  Most companies will have at least a six-month, I would say the majority maybe one year, no-reapply rule for urine testing.

 

MR. EDGELL:  That's not what I'm talking about.

 

MR. THISTLE:  Well, they've established a "currently using" window.  Basically, what companies are saying is that we believe that "current" for the purpose of ADA is this one-year window because you're denying a person the opportunity to reapply.  On that basis they're saying, we're determining that a one-year period or a six-month period is necessary for you to properly rehabilitate yourself.

                Again, if someone comes in with documentation that they have been rehabilitated, that's something the employer can take into account.  I imagine you could do the same thing, just in a shorter time frame, with a urine test that might be positive for marijuana, that if the person claimed they were a chronic user and it was from a while ago, that in that intervening weeks or whatever they went to rehabilitation.  It's just that you shorten the time period, but the issue is still there. We've had millions of tests.  It's not an issue.

 

MR. STEPHENSON:  I think one of the things it demonstrates is that the discussions that you have in developing a proposed standard is no substitute for the actual court case that's going to define what you really do, and those will come. 

 

MR. IRVING:  We didn't actually address that issue.  Back to question 2.

                (Screen.)

                This addresses the recommended reasons for testing.  The last draft of the guidelines eliminated random testing for hair collection as applicable to use in a random testing program.  We got information roundabout that part of the reason for that was that if you were in a random program and you went back three months, it may predate the time of employment by the company.

                We thought if that was the reason then you could go ahead and say:  Okay, you can't do hair testing for a three-month period after the individual's first brought into the company.  But we thought it was sort of inappropriate to eliminate hair testing under the random program because it does give you a wider window.  It gives the companies a very good tool to go ahead and check their employees, and there is no real reason to eliminate this tool from the program.  If there's a procedure problem during the first 90 days, eliminate it as a possibility for the first 90 days.  But you should also understand that when a person applies they're also saying, I haven't used drugs for a certain period of time.  If they say they haven't used for a 90-day period when they apply, they get screened two weeks later and it shows there they did, then it's a fraudulent application.

                You may not even have to eliminate it from that standpoint.  That's a legal issue.  But from a strictly scientific issue, it seemed foolish to eliminate hair testing as a tool when it really fits very well into a random testing program.

                (Screen.)

                Question 3 was:  Should cocaethylene and norcocaine be included to screen and confirm?  Well, under the screening procedure, since cocaine is overwhelmingly the drug that's seen in hair, the amount of cocaethylene -- you're really screening for cocaine.  In all the assays across the board, you're screening for cocaine.  However, in the case of the confirmation procedure, in all the recommendations we've sent back from the previous meetings we have included cocaethylene and norcocaine as part of the testing scheme.  As a test -- this is a little different.  This goes outside of any of the normal ways we've proceeded looking at testing under urine and I suspect saliva and the sweat patch, too.  We feel that in the case of hair testing you need to have cocaine there in addition to one of the other metabolites in order to prove that the individual actually used cocaine.  This, used in conjunction with good decontamination procedures, whether it's our wash procedure or somebody else's wash procedure, however, you want to go about demonstrating decontamination, we feel is necessary for good identification of cocaine.

                This is where it's a lot different from how you view any other positive.  In this case, there's different ways of calling an individual positive for cocaine.  First of all, in all cases the individual has to have 0.5 nanograms per milligram of cocaine detected in the confirmation procedure.  But that's not enough to call him positive.  After that, he has to fit one of the other criteria.  He has to have benzoylecgonine there equal to or greater than 0.05 nanograms per milligram and has to have a BE-cocaine ratio of 5 percent or greater.  That's to eliminate any benzoylecgonine that might be available in the drug due to the environment.  Absent that, he could have cocaine at 0.5 and cocaethylene at .05 nanograms.  Cocaethylene is an extremely good metabolite to look at.  It only occurs when you simultaneously use cocaine and ethanol at the same time.

                The other possibility is to use norcocaine, again using a 0.05 nanogram cutoff.  The norcocaine shows up in a much lower concentration than any of the others and only be looked at if you can't meet it by the first two criteria.

                This is a lot different from anything that we've done in any of the other test schemes, but we feel it's necessary for proper confirmation of cocaine.  Any questions about that?

 

DR. VOGL (HHS):  Was this complete agreement from all the attendees?

 

MR. IRVING:  Yes.  One of the questions is, does it have to be -- I think everybody when faced with the question, if you don't have cocaethylene there, if you don't have benzoylecgonine there, and you don't have norcocaine there, doesn't that sort of raise a red flag that you have something else as a problem?  We may miss some true cocaine users this way, but we thought it was safer.  The biggest question was can we get down to these levels of the metabolites, and there was general consensus that yes, we can get down to these levels and it is a necessity.  For some reason or another, it's been left off, cocaethylene and norcocaine have been left off the last two drafts, even though it's been in there from the beginning.

 

DR. SAMPLE:  One question about the benzoylecgonine.  Are there any issues about hydrolysis?

 

MR. IRVING:  One of the things you have to demonstrate is that in your process you're not hydrolyzing a disproportionate amount of cocaine to benzoylecgonine.  If you show that you're doing less, you have to demonstrate that you're doing less than one percent.  That would be one of the things charged to the PT program.  The other thing that we talked about in this question, but it applies across the board, we didn't specify the decontamination procedures.  We thought that, instead of saying that, that the PT program is going to be the area that's tasked with providing samples that are contaminated versus samples that are not contaminated, also positive samples that are contaminated, so that the labs can demonstrate that they can clearly decontaminate the samples and can identify a user versus a sample that's strictly contaminated. The question of hydrolysis would also be part of the PT program.

 

DR. SAMPLE:  So PT and ongoing QC is part of the assay?

 

MR. IRVING:  Yes.  You'd have to check the hydrolysis of each batch to make sure that you haven't hydrolyzed to meet this criteria.

 

DR. SAMPLE:  Okay.

 

MR. IRVING: (Screen.)

                Question 5:  Can methamphetamine alone be the result of external exposure or use?  Again, there are two questions.  One, you still have to demonstrate the ability to have decontamination of the sample to make sure that it wasn't external contamination.  The big question was whether we need to apply the same type of methamphetamine rule that applies in urine.  We thought for consistency that needs to be, that they require that you demonstrate the presence of it, amphetamine, along with methamphetamine, the same type of rule that is in urine right now.  But the decontamination really is a separate issue and again to be addressed through the PT program.

                Question 6:  Should only head hair be allowed?  We decided that really if you want to give people a perfect way of getting out of being hair tested, say you only allow head hair, because the person will shave their head.  We've seen that already.  You've read a lot of places where people will shave their head to avoid hair testing.  You really don't want to give them that out.  That's why we left the body hair in.  There was a minority group within the program that said for simplicity they would like to go ahead and just do head hair.  But that's strictly -- that really does limit the use of hair.

                Question 7:  Do we need to widen the range for the laboratory open controls from 50 percent to 200 percent for screening and minus 40 to plus 150 for confirmations.  We decided that the recommendation from DTAB is fine the way it was written.  There wasn't any reason to change that.

                (Screen.)

                Question 8, and this was the only place that we got into some controversy.  Part of the controversy was put off because we had to wait and see what the MS group said.  For the drugs there really wasn't any controversy.  The cutoff levels for amphetamine, cocaine, opiates, and PCP were pretty much agreed to and there was a very good consensus on that both for screening and confirmation.  The only place that we got into a controversy was THC.  We decided that, obviously, the drug of choice as a target of confirmation was THCA, not THC, that you need to have the THCA there in order to call it positive.  In here we put a couple different possible cutoffs for the confirmation.  Again, it's going to be dependent upon what is the final draft of what the MS group, what the Board is going to decide the MS group requirements are going to be, and that's still somewhat fluid.  Obviously, we want to make sure that the confirmations that we do are scientifically sound, that once the minimum requirements for mass spec are agreed to you can't lower them, nor do they go down to a lower confirmation, but at the same time you can't be -- I guess you have to think outside the box to make sure that you haven't eliminated also scientifically valid criteria.

                I think probably if you take a look at some of the things that may be proposed for your next meeting.  Once those draft guidelines are set, then this group can come up with a criteria for what a confirmation rate for THCA should be.

                The other thing -- and this is where the controversy actually came up in all the e-mail that flashed back and forth once it came out -- was, what should the screening level be?  Here we know what the cutoff level for confirmation is since the screening is in support of the cutoff, again that has to be put off until we determine what the criteria of confirmation is going to be.  But for the other four drugs we had a very good consensus and basically total agreement.

 

DR. CHILDS:  Does that mean that you're going to allow the actual performance to determine the cutoffs, and as technology improves the cutoffs may go down?

 

MR. IRVING:  Yes, and I think that's part of what happened with the urine program.  When we first started, 6-MAM was not put in because the capability of the laboratories wasn't there.  As the capability of the laboratories increased, then that could be put in.  But I think as the PT program shows, the levels can go down.  Also, you have to look at what the rates are, and part of the documentation that we provided for the group was there's a breakdown of what the positive rates would be at different cutoff values.  That has to be balanced in also.

 

DR. BUSH:  Paula, also cutoffs are determined based on what can you interpret, what can you clearly say is a response to a particular analytical result you get from any specimen analysis.  One thing that we were looking at in urine and have over the entire life of the program is taking a look at passive exposure and that possible likely contribution to a positive test result.  In that, we are very conservative in our cutoffs for a variety of reasons, and that's based on a lot of peer-reviewed scientific information.  Many times, depending on the matrix, are you looking at chronic exposure or are you looking at acute exposure, what can you draw from those cutoffs as they're established for interpretation purposes.  The capability in a laboratory is only one side of it.  Clearly, that's true for the 6-acetylmorphine.

 

MR. CROUCH:  If you use the proposed MS/MS-tandem MS criteria, how many laboratories could achieve one picogram per milligram THC and 0.05 picograms per mL?

 

MR. IRVING:  None.  If you have to produce the two daughter ions, about -- different terminology, 0.3 picograms per milligram of hair is what's achievable at this point.

 

MR. CROUCH:  If you use the current criteria, the THC would be one picogram?

 

MR. IRVING:  The THC would be, if you're requiring two daughter, we're talking about 0.3.

 

MR. CROUCH:  What about for THC?

 

MR. IRVING:   We tried to stay away from THC because THC is going to produce a major environmental problem.  It gets absorbed on the hair.  We can go to a lot -- you don't have to go that low because there's a lot of THC sitting on the hair and it doesn't wash off that well.  We decided you really need to show the THCA.  Showing the THC made no difference.  I think APL has experience or they have a lot of samples with high THC, but the THCA is at a lot lower concentration.

 

DR. SAMPLE:  Do these cutoffs take into account the 40 percent retest type of criteria?

 

MR. IRVING:  In this particular case, yes.  The 0.3 would count 40 percent.  The cutoffs that we looked at then was for the base compound.  For cocaine, we didn't apply it to the BE or CE, but that's very similar to the requirement for amphetamine.  When you use amphetamine as a decision maker for methamphetamine, you're not applying a 40 percent rule to that cutoff.  We didn't apply it to the decision makers on cocaine, but we did apply it to cocaine itself, because the lowest level you challenge methamphetamine at is 180, even though you have a 200 decision making cutoff.

 

DR. GUARNIERI (DTAB):   I just wanted to make a comment.  It's a related subject and it's a medical issue.  As we get further down the road -- and I realize that we're talking about cutoffs.  But one of the questions that you've asked here is should we only use head hair, and I just wanted to be sensitive and talk about ADA issues, etcetera, but there are some medical conditions alopecia being one of them, where people will not have head hair and very little body hair.  There will be people who will be on chemotherapeutic agents for breast cancer, as an example, who will lose their head hair and a lot of other body hair.  When we set up our programs, please be sensitive to the fact that there will be some medical issues that will preclude that people may or may not have enough hair for such a sample.

 

MR. IRVING:  I think that we have companies that do that, that if you can't provide a hair sample for a medical reason then you go to an alternate matrix.

 

MR. STEPHENSON:  I think the bottom line here is that this is exactly why we look at the complementary nature of some of these other testing technologies that will give you a better set of alternatives to look at as tools.

 

 

Oral Fluids Working Group

 

DR. NIEDBALA (STC):  Rather than go through each question, what I'm proposing to do, and the Board will direct me differently, is go to the items that either were the most controversial or sort of the heart of it, because some of these are very repetitive. 

                Back around Christmas we were given a series of 13 questions to which this working group was to respond.  We had a series of three teleconferences with everybody calling in from around the country.  The list of folks has grown compared to the first meetings.  Generally, multiple representatives from already participating organizations was the end result.  The dynamics were good in that there is more and more information that's becoming available right now on oral fluids.  There are publications which the Board has been given as part of the response, package inserts from now products that have been reviewed by the FDA.  In addition to that, there was some research information that was provided by certain individuals or organizations.

                As a general rule, let me just recap a few of the most important issues.  One, everybody involved believes that as much as possible manufacturers should be held responsible for quality control and ongoing assurance of performance for any one of these products that are being proposed.

                Secondly, in all cases we should be normalizing back to oral fluids, the whole oral fluids, rather than individual devices which may have a dilution factor.  There's really nothing new in that and that still continues to be consistent.

                More where we had our disagreements, or I should say discussions, were on some of the QC criteria and also on the cutoff.  Again, anyone want to guess which analyte is the most controversial?  It's the same for this fluid as well.

                In general, there was agreement among the group for most of the recommendations that are shown.  Any dissenting minorities, as we called it in our conference calls, are noted to you, and people certainly were encouraged during public phases to submit any additional information, any disagreements.  I know that is the way that we try to work with things.  It's an open dialogue that way.  In addition, a second bullet point here is that people generally have lots of studies that are ongoing right now, although many times I said, well, we're almost there.  Our rule of thumb is, unless we can include it in the package you need to submit that on your own and so be prepared to do so in the future.

                (Screen.)

                Let's go to the bottom line, which is the suggested cutoffs.  You'll notice a couple of changes here.  The methamphetamine was lowered from 160 to 50.  The cocaine is kept at 20.  And there was much discussion about whether we should be targeting BE and cocaine or BE alone.  Where the group ended up is that they targeted BE.  However, just to make note, there are manufacturers that will have antibodies that may cross-react to a high level with cocaine, but ultimately the confirmation is where we separate out, and 8 nanograms is the cutoff confirmation, which may target BE and/or cocaine with a minimum of 8 each.  This is not combination.  We wanted to be as specific as possible for the Board.  It's 8 nanograms each.

                THC, again, there was much groaning and weeping and gnashing of teeth over THC.  The reason is because the cutoff is fairly low.  Every technology is challenged to do this.  There are already products on the market that are targeting these levels, but as time goes on we would anticipate technologies making improvements and being able to achieve this.

                The cutoff for the screen was kept at 4 nanograms of THC and for the confirmation 2 again of THC.  I would say that you'll see additional information coming from people who participate and others looking at potentially the kind of the noise rather than THC alone.  But again, no one stepped forward during the Board meeting or during the working group meeting to present that to the Board.

                Opiates was pretty straightforward.  There's really no changes there.

                PCP, the recommendation was to raise the cutoff a little bit, from the 4 nanograms up to 10.  This is probably the weakest in terms of rationale.  There is one product that's already on the market that's a collector that's lab-based, but none of the onsite devices really had data on this.

                The bottom line was there was some tweaking that went on, there was some new information presented, but not a whole lot of change, and again the most controversial was the THC.  In addition, I should be sure that we make note that these cutoffs and all of the information presented includes discussions about onsite devices as well as lab-based tests.  As you may or may not remember, there are devices which collect an oral fluid sample and send it back to a laboratory for analysis and then there are devices that will use on-site technologies that analyze oral fluids.

                The same cutoffs are recommended for both and the same target analytes or parent, depending on the situation, are also targeted in both.  That was where we had ended up in terms of the recommendations.

 

MR. CROUCH:   I believe the last cutoff for stimulants was 160.

 

DR. NIEDBALA:  Correct.

 

MR. CROUCH:  What's the rationale for reducing that threefold?

 

DR. NIEDBALA:  There was additional data from some of the working group participants.  This was I think primarily spearheaded by Christine Moore.  She had presented a packet of information which supported the lowering of that cutoff and the working group discussed that, put the data into the package for you all to look at later, and had gone with that.

                Again, the principle has been to present scientific evidence to support any of the cutoffs, generally based on ROC analysis.  As you know, in the early phases of any sort of new technology area, especially in drug testing, our cutoffs in part may be a reflection of the prevalence in the particular population studied.  I think the suggestion was to lower it, which was consistent now with more information coming to light.

DR. SAMPLE:  Did you say that we are being provided with documentation, a handful of papers?

 

MR. LoDICO:  Yes.  I was the conduit by which all the timelines and all the information was sent to my attention.  I collected and put it into the spiral folder. 

 

DR. NIEDBALA:  For the public comment now, I tried to just bullet point some of these responses and the ones that we spent most of our time talking about.  There's two other areas that I'd like to at least mention.  One was the minimum quantity or volume to be collected.  What we did here, because I think Charlie was mostly driving this on these particular conference calls and Yale to a certain extent, you got to quantitate.  You have to give us concrete answers.

                What we did here is we had each manufacturer list out for the screen, a rescreen, and a confirmation the volume that they need for their particular device or approach to work.  Without naming names, we went across the board on the conference call and listed these materials out.  The end result of this was that by the time you got to the confirmation a 0.5 mil was the minimum recommendation. 

 

DR. SAMPLE:  Milliliters and microliters, is that correct or is that a typographical error on the screen?

 

DR. NIEDBALA:  Typographical.  In your notes it's all correct, it was in my notes in getting the bullet points.

 

MR. CROUCH:  Is this for one confirmation or is this an adequate volume to do two confirmations or a confirmation and a retest?

 

DR. NIEDBALA:  Let's see.  On the particular discussions, we talked about it in confirmation.  I could tell you, my assumption was that there would be enough there to reshoot if you had to do it again.

 

MR. CROUCH:  The reshoot or re-extract?

 

DR. NIEDBALA:  No, reshoot, not re-extract.  I can't say that we went through that as a detailed point of discussion.

 

DR. SAMPLE:  Would that then mean only one analyte for confirmation, and if you had multiple analytes one would be in trouble in trying to confirm?

 

DR. NIEDBALA:  No.  The expectation was that you would be able to do that.  Part of the criteria of the answer was what do you need to do all of them.

 

DR. SAMPLE:  So on the basis they all screen positive?

 

DR. NIEDBALA:  Yes, what would you need in order to go back.

DR. SAMPLE:  To confirm at least one time every analyte.

 

DR. NIEDBALA:  Right.

 

DR. SAMPLE:  I guess in response to Denny's question, then, if there was only one analyte that screened positive, only one confirmation you had to do, you had sufficient volume to re-extract four or five times.

 

DR. NIEDBALA:  Based on his question, I guess what you would say is you can shoot and then if you went back and had to do a second you probably couldn't do them all again if you were just bringing it back up in volume.  That's my opinion, though.  I'd have to caveat, we did not discuss that kind of detail around this question.  The other point to this which we wanted to make to the Board is just that these are the volumes that exist for technologies today.  There's clearly this move to nano-technologies.  There's some discomfort with putting something like this down, because as things develop with all of the impetus to push on the technologies I wouldn't be surprised if this goes to a tenth in the not too distant future of the volume that you're seeing now.  I think that will be because of all of the new technologies coming down the technology pipeline.

 

MS. CLARK (Alpha Pro Solutions):  Mike Walsh raised the issue earlier regarding the dry mouth with the MDMA user.   These volumes are not a problem with the drugs that we're discussing here, but wouldn't that be appropriate for potential for an MDMA user?

 

DR. NIEDBALA:  After he said it, I was sitting there thinking we need to go get some of these people.  And one brought that up in the discussion, although dry mouth is one of the things that I think in prior presentations we've looked at data from Mike Pete's lab, LabOne, where we compared urines where we couldn't collect oral fluids, where there's a great deal of that testing going on today.  What we saw was a very small percentage of people that we could not collect from.  Now, MDMA-specific studies I don't know about right now with devices.  It's a good question.  It just has to be answered with the scientific data.

 

DR. GUARNIERI:  Just of interest, blood, gastric contents - do any of those contaminants throw off your work?

 

DR. NIEDBALA:  That really depends on the device, and people have to answer.  I know in my own experience, we've tested everything from people drinking Coke and orange juice to peanut butter sandwiches to see what would be the effect.  The position of the working group from a long time ago has been for people conducting the test they wait ten minutes before they actually collect the sample.  That gives the mouth time to clear.  We salivate between a liter and a liter and a half a day.  That's supported by other studies.  I think we had previously addressed that one.

 

DR. HUESTIS:  For THC, you're assuming you have a screen that targets the parent compound and not the carboxy-THC metbolite?

 

DR. NIEDBALA:  No.  I think just like with the limitation or benefit of all antibodies, they're going to cross-react with multiple things.  The question is whether or not the antibody picks up sufficient, or has sufficient, sensitivity to the parent.  It's highly likely if we went back and examined every one of these antibodies that we're going to have cross-reactivities to the other metabolites.  I wouldn't be surprised if there is a huge cross-reactivity with almost every one of them to the carboxy-THC metabolite, as an example.  That's just the nature when you develop these antibodies.

 

DR. HUESTIS:  For instance, some of them that target the carboxy-THC have very low cross-reactivity to the parents.

 

DR. NIEDBALA:  Yes, agreed.  Someone would not collect that one to use in a test.

                (Screen.)

                The last overhead I was going to go to unless there's a specific question, was QC criteria.  This was also a place where we ran into discussion.  Here's where you separate lab-based testing from on-site testing in a certain respect.  For testing in general, what we followed were the recommendations for the point of collection urine tests.  What you see then is that for the sites that are performing oral fluid on site that they test each day a negative, a specimen 50 percent above the cutoff, and when a presumptive positive is found that they run a positive control following the positive test, then one out of every 20 negative samples is sent to a laboratory for confirmation.  We just kept the same model and are making that recommendation to the Board. 

 

DR. VOGL:  If you're recommending that everything goes back to neat oral fluid, then how accurately can the different manufacturers establish what's actually collected in their devices?  Some go into a diluent.  How are they going to manage to say how accurately they can collect that amount?

 

DR. NIEDBALA:  That is going to be one of the primary questions for all of the manufacturers.  They're going to have to present the data that says, on average we get this much fluid and therefore when we normalize back this is how much we see as whole oral fluid.  Each technology that comes up in the future will bear the burden of proof to show that, and therefore inform potential users of the technology exactly what they would expect to see.  That's basically the point of proving safety and effectiveness, that you're telling them this is what it does and this is the data to support those assumptions.

                It's very dry today.  There's nothing controversial here.  We had a little bit of that on the conference calls, but these were good discussions.

 

MR. EDGELL:  This is not controversial, just a question.  On number 11, are there special storage requirements, you answered that each manufacturer would need to verify those particular conditions required for storage and transport.  Could you just list some?  Are there any special storage requirements that you're aware of?

 

DR. NIEDBALA:  No, I think the intent here was that folks have -- and again, remember we're answering on-site and lab-based.  There are a number of very clever ways that manufacturers are addressing this, like a pad being sent back to the laboratory has a storage buffer, does the buffer have stability well-documented for each of the target analytes that we want to see?

                If we've now set the bar and said these are the things you're going to measure for screen and confirmation, then it's incumbent on the manufacturer to provide the data or technology behind it to make sure that those things show up at the lab within the stability time frame that they have.  That's why you couldn't have sort of a generic general answer -- or, I'm sorry, you couldn't have the specific answer necessarily that covered each and every device, but rather put the burden so that they meet the requirement of how the testing algorithm will be conducted.

 

MR. STEPHENSON:  As you've gone through this process, we had one orphan that sits out there that's been pre-validated in the context of alcohol.  In any of the efforts that you've gone through now, are you finding that the alcohol-complying product issue might be an outlier in any of these standards you're developing?  Are you bringing it along at the same time and updating that, so we don't wind up with two different sets of standards?

 

DR. NIEDBALA:  You're talking about one of the participants who developed a combination alcohol and drug test.

 

MR. STEPHENSON:  Yes.

 

DR. NIEDBALA:  Happy to consider it.  There was no data presented.

 

MR. STEPHENSON:  That's the issue, is that you have a complying product issued that's alcohol-specific now for DOT screen and that as we look through these kinds of issues we are several steps ahead now in terms of the whole industry, and I just want to make sure that we do in fact be inclusive in this, if there's anything that we need to assist to take a look with NHTSA in that perspective that we do.

 

DR. NIEDBALA:  I was involved in the early days with the alcohol and lived through that.  I thought that was a really good series of events in terms of how we now conduct.  A long time ago when our working group first met, that's exactly what we pulled out to say, here's how you test.  I think the model has worked out very well.  Years later, doing on-site alcohol tests ourselves, it works well.

 

MR. LoDICO:  Can you comment on question 14?

 

 

DR. NIEDBALA:  This question relates to the usage situation for oral fluid.  In the last draft there was a change in the usages assigned for oral fluids, meaning from everywhere to just a limited section.  Now, the working group itself had presented some data on comparing oral fluid to urine tests.  Those come in the form of new publications, some pending publications, some internal data.  Again, you'll look at some of that later today.

                What you'll see is that if you look at comparison rates the agreement between oral fluid and urine is about the same.  What is a pharmacologic fact, however, is that for oral fluids you'll find some people who will be positive in oral fluid and may be negative in urine.  Conversely, it can be positive in urine and negative in oral fluids.  Those are a small percentage of what you find.

But what it really reflects is the time of dose, in simple language.  There's more and more data to support that now that's coming out in the peer-reviewed scientific literature.  But for the purposes of deterrence, which is the primary purpose here, it works comparably to the urine in the way that it's used, in the same way with the added technology challenges that come along with using this fluid.

 

DR. SAMPLE:  But that statement holds with the cutoffs as presented here, at least within the working group's opinion?

 

DR. NIEDBALA:  Right.

 

DR. SAMPLE:  And that is supported then by the documentation, the additional documentation that's been provided?

 

DR. NIEDBALA:  Correct.

 

DR. SAMPLE:  But if the cutoff were to be different, then that statement may or may not still hold.

 

DR. NIEDBALA:  Right, and that was the point of all of our discussions, that in some respects there were folks who would love to just push up, as an example, the THC cutoff.  Why?  Because it's analytically easier to hit.  Clinically or in practice, it wouldn't be very practical at that point.

What we did is we held the line and we said the utility is ultimately what we have to fit, and so technologies, just like in the early days of urine, are challenged to step up and meet it.

 

MR. STEPHENSON:  Utility is also a function of the general value or perception of what you're getting with the test, what is your detection window that you'd be looking at, some of these, where are you in terms of errors or multiple errors, and so on, do you have a sense of that.

 

DR. NIEDBALA:  Yes.  You're going to look at some of the information later that corresponds to this.  For those materials that there were dosing studies for, marijuana generally was 24, and there's one study that will show positive rates in comparison to a single dose out to 72 hours.

                For cocaine, a single dose was fairly short.  I think the longest was about 24 hours.  For opiates, there's not a time-dose study that's in there.  There are broader-based, where you have a generic population that you are collecting.  PCP, it just doesn't exist.  We can't hold people still long enough to do something like that and we've never had an RRB give us approval.

                For methamphetamine, it's more population-based than it is time of dose.  Again, remember the criteria as a medical device is not necessarily time of dose.  Nobody claims windows of detection when they develop these kits.  What they do is show comparison to a predicate, which is analytically based and clinically based only in the sense that, is it here by GC mass spec, as an example, and immunoassay in both fluids.

                The basis for these products being used as medical devices is not window of detection.

MR. STEPHENSON:  But maybe that's the difference between where we are and some of the others.

 

DR. NIEDBALA:  Yes.

 

MR. STEPHENSON:  I agree with you that for medical detection that's one issue.  For us, though, I think the utility is going to be the window of detection as one of the primary things that we have to understand.  That's something we also have to communicate and educate a lot of other people about when they start to use these.

 

DR. NIEDBALA:  Sure.  Hopefully Charlie will hand those out later and we can look through that and come back with questions.

 

Sweat Testing Working Group

 

MR. FORTNER (PharmChem): 

                (Screen.)

                We had five questions that were submitted to the working group.  The working group did not meet, partly based on the nature of the questions.  Some of this information has already been submitted to DTAB and we weren't quite clear if it was just a clarification of that.  We did have some conference calls and discussions.

                The first question actually became an all-encompassing question as it answered two or three of the overall questions.  2.4 asked about the minimum quantity of sweat to be collected.  Now, I think if you go back and look at this, and there was a long discussion in previous meetings as to what constitutes a sample, certainly sweat is the biological matrix, but you're collecting that and it dissipates off of the sweat pad.  Really, the issue was are you going to have enough sample to perform initial confirmation and potential re-screening and-or reconfirmations of that.  Then this links back to the size of the pad and how do you express it in terms of volume.

                I think the answer to A is yes, under the current guidelines you have enough samples to do screening, repeat screening, and probably two to three confirmations, depending upon what drug that you're looking at.  Now, this ties into the other component of this question that asked about the physical dimensions of the absorption pad in the patch.  Currently, it is 4.8 centimeters by 3.1 centimeters by 0.698 millimeters in thickness.  It is the consensus of the group that this is the de facto size of the patch, at least of the absorption pad, as this is the device that was used in all of the 510k FDA studies.  This is the medically approved device.

                The group was not in favor of changing the size of the pad at this point.  Plus I think there's going to be an issue of whether anyone wants to manufacture one that's two to three times this size.  Certainly, in previous issues before this group there has been questions about split testing and the recommendation was to apply multiple patches, which would allow you to test one and not the other.

                (Screen.)

                The question continued to go on in terms of how do you reconstitute it, and we'll talk about that because it relates to I think question number 3.  We'll follow it along those lines. 

                The question as this relates to reconstitution of the patch, currently the pad is reconstituted with 2.5 mLs of acetate methanol solution.  It's been proposed that you could increase that to 3 or 4 because there was concern about having sufficient sample to do multiple tests.  In our experience, we rarely get poly-positives.  Typically, an individual is positive for one drug, not several drugs, although our experience is driven primarily out of drug court.  Certainly in other arenas, as in criminal justice, where poly-pharmacy is a different issue, you can have that occurring.

                As it relates to question number 3 is asked, how should it be expressed.  Currently, there are two approaches.  It's expressed as either nanograms per mL of the eluate or it's expressed as nanograms per patch, and that's simply taking the amount of solution that you used to reconstitute the patch, multiplying the drugs by that, reaching a nanograms per patch.  In one scenario, you'd have a cutoff of 10 nanograms per mL.  That same concentration would be 25 nanograms per patch.  Either way, there didn't seem to be a large concern about expressing it until we got up to the point of discussing how would you do a retest, and that if a retest is requested the ability to express it as nanograms per patch now becomes dependent upon what the original laboratory used to reconstitute that with.

                Based on the fact that some laboratories may want to use different amounts, it then adds a level of complexity that says you then need to have a total amount, you need to know how much they reconstituted the patch with originally in order to say, okay, I got half mL, but they used 5 mL to reconstitute.  To express it in total nanograms per patch, I have to know what that total volume was from the original outset.  In discussions within the group, we thought it would be easier and consistency within the program to just keep it at nanograms per mL.  Then we wouldn't have to worry about what did lab A reconstitute it with, what did lab B reconstitute it with, and having to have that information transferred, although it didn't seem to be a tremendous sticking issue within the group.  Is that clear?  There seems to be some questions or a puzzled look.

 

DR. SAMPLE:  Does that mean that you would specify the reconstitution volume?

 

MR. FORTNER:  Yes,  in essence you would fix the reconstitution volume to be able to express that consistently.  If you want to do a 3 or 4, you still want to fix that in there, so you don't reconstitute it with 10 mLs.

                (Screen.)

                Going back up to question number 2, which we sort of got out of sequence, but that's the order in which the questions were proposed to us, asked about the data to support the detection of heroin at 10 nanograms in this particular case, 10 nanograms per mL of eluate.  What we did is, if you go down, there's been several studies, Dr. Cone's, Dr. Kintz's in particular, who have done controlled dosing studies of heroin and individuals at those detection levels.  We provided the references to DTAB, and that's where that data comes to support the detection of that.

                Now, certainly there is one consideration in there that came up after we had submitted this in, in that long-term storage and retest you may have hydrolysis of diacetylmorphine into 6-acetylmorphine.  The retest has to take into consideration that you may be looking for both heroin and 6-acetylmorphine.

 

DR. HUESTIS:  In doing our research with sweat patches on individuals in treatment, the 6-acetylmorphine identified 75 percent of the people, where the heroin itself only 28 percent.  If you went for heroin -- I mean, you can see it in some cases, but in a lot of Pascal Kintz's work, which I don't believe controlled administration -- I'm sure they weren't -- they didn't find it.

 

MR. FORTNER:  That's certainly an issue.  The work that Dr. Cone has done is probably, at least in the U.S., the only controlled dose studies out there.  Once you reconstitute it, you do have an issue on the potential of hydrolysis going forward.  I think that's an issue certainly that has to be redressed in retest scenarios.

                Question number 4 went on to ask about, do you require the amphetamine confirmation rule?  Here I would probably divert and disagree with John Irving's statement, only from the standpoint of following the guidelines.  If you look at the guidelines, the guidelines are that you require the 200 rule as a result of potential sympathomimetic conversion to methamphetamine.

                Clearly, in the PT sets the laboratories have to demonstrate, even in urine-based programs, that you don't have that analytical problem occurring.  I think that's a different issue in terms of, it's a different issue in terms of imposing an amphetamine rule.  In this particular case, I don't believe it's the issue that needs to be addressed in terms of generation, although certainly that's a concern.

                We really had to go look at it and to address an external contamination issue:  Is my positive methamphetamine because, for some as yet unexplained reason, I have methamphetamine sitting on my skin and that's why I'm positive in this particular case?  The implementation of an amphetamine rule really addresses, I think, more an external contamination than it should an analytical laboratory issue, which is addressed throughout a PT or meth validation program.

                Nevertheless, there wasn't disagreement within the group to include some kind of requirement for amphetamine.  However, if you look at studies that have been done, you don't always see amphetamine.  It's a dose-response issue.  Certainly, amphetamine present at certainly limit of quantitation, which in our laboratory is about 3 nanograms per mL, would be a reasonable approach to use and address this issue that's really more directed towards external contamination.

 

MR. LoDICO:  Would you then recommend that the laboratories perform some type of validation study so that they don't generate this amphetamine and remove that as a condition?

 

MR. FORTNER:  Absolutely.  I think your first question is yes.  The laboratories need to demonstrate, just as we currently need to demonstrate in urine testing, that we don't generate methamphetamine.  There has to be some similar type in all the analytical processes to say, I don't generate amphetamine in this particular case.  That's a different issue, though, I think in terms of addressing if you want an amphetamine rule from a metabolite perspective. You may want to have the rule, but for a different reason.

 

DR. CHILDS:  Maybe I'm missing something here.  It may not be detectable.  Have you seen a change in the rate of how many methamphetamines are being reported as positives since instituting this amphetamine LOD reporting rule?

 

MR. FORTNER:  No, because if you go back and look at the data, all of our methamphetamine positives typically are methamphetamine only, but there is amphetamine there.  We apply an administrative rule that says if it's not at least 10 nanograms it's reported as negative.  That doesn't mean that it isn't there in at least a 3 nanogram, which is our limit of quantitation, concentration.

 

MR. STEPHENSON:  You've raised this issue a couple of times.  Maybe it's me being a little redundant, but I want to remind everybody here that our primary objective in this group is to develop approaches that are forensically defensible, and that we talk about what's scientifically valid or an appropriate analytical procedure that's not enough for us, that our intended audience is hostile by nature, and that we have to prove our case literally, and that when we do this it's not a lesser charge to each of the working groups to come up with the full approved standards that show there is not an alternative explanation for a test result, and that there is in fact a forensic defensibility for all of our reports.  Somehow, this seems to get a little lost in some of these things and we need to strengthen this part of it whenever we have the opportunity to do it.

 

MR. FORTNER:  The last question that we were asked was, provide the reasons for application.  This is question 5.  Actually, question 5 really refers to question 2.2 of the guidelines in terms of how is this alternative matrix applicable.  I think from the start of this concept of alternative matrix it's been clear that sweat testing is really not applicable in a pre-employment situation.  We know off the controlled dose studies that it has to be worn for a minimum probably of 24 hours.  Most of the drug is excreted between 24 and 36 hours in those controlled dose studies.  It has to be worn for a minimum of two hours to detect drug, but that's at a fairly low concentration, around limit of detection or limit of quantitation, 2 to 3 nanograms per mL.

                Its applicability really doesn't address pre-employment.  It addresses perhaps follow-up, aftercare, after a positive.  I think it has that particular application and that's what's been proposed in the guidelines, and the group had no questions in terms of changing that.

 

MR. LoDICO:  In terms of the use of the patch for follow-up, is there a recommendation of the time that the patch has to be applied?

 

MR. FORTNER:  Yes.  Clearly it needs to be worn for at least 24 hours.  Most of the programs were for 7 to 14 days, and no longer than 14 days is the recommended.  We have a lot of data, but the data is most supportive of 14 days, because starting after 14 days you will begin to see some falloff as the epithelial cells begin to sluff and you can lose adhesion.  Certainly loss of adhesion from the polyurethane covering then presents a potential compromise in terms of exposing the absorption pad to the environment.  In that window, at least 24 hours, no more than 14 days.

 

MR. LoDICO:  To be fair to the general public, what is the absolute time?  Is it two hours or is it 14 days?

 

MR. FORTNER:  No, it's got to be at least 24 hours.

 

MR. LoDICO:  In terms of a rulemaking policy, if you need to say exactly that it has to be on one day, seven days, or two weeks, which is the one that you want us to aim for?

 

MR. FORTNER:  You ask a multiple part question.

 

MR. LoDICO:  No, it's simple.

 

MR. FORTNER:  A minimum of 24 hours.  Maximum of 14 days.  That's the window.

 

MR. CROUCH:  Shouldn't this be more generic, though, and say something like:  follow manufacturer's recommendations?  Because suppose somebody comes out with Patch 2, which, heaven forbid, is better than yours, and it only needs to be worn for 2 hours or 30 minutes or something?  Then we're locked into a guideline in the Federal Register that says it's supposed to be worn in this fashion, which really is peculiar or particular to the PharmChem patch.

 

MR. FORTNER:  Well, that's certainly true.  If you want to go put it generic just in terms of saying, following manufacturer recommendations, there are other devices that are out there that collect sweat, although currently in much smaller volumes, in a much shorter period of time.

                There's another device that is years from the market which is worn for 20 to 30 minutes, and it has advantages, disadvantages.  There aren't enough studies done on it at this point, but a general statement of follow the manufacturer recommendations would be a pretty broad statement.

 

Point of Collection Testing Working Group

                               

DR. CAPLAN:  The point of collection testing group is a little bit different than the others.  First of all, the group had not been meeting previously.  There was a one-day or several day meeting which we had a while ago where there were a number of presentations from all different sources, and there wasn't any follow-up until there was some questions developed from the guidelines.

                At that time, we created a group which seemed to be representative, and we had three conference calls and it kind of grew as we went.  Although I think it was not intended to have every person who ever manufactured a device represented, all the major manufacturers and anybody who asked to be on the call was included.

                The group generally -- you don't have this on what's here, but the group generally included me as the moderator, Bob Willette, who was the reporter and was supposed to give this talk today.  It's probably a little cumbersome because I didn't set up specific slides to do that since he couldn't make it.  Charles LoDico, David Evans, Sal Salamone, Jennifer Trivitts, Ken McNeil, Mike DeFao, Chuck Blake, Kae Jung Kim, Rafael Wong, Jennifer Collins, Murray Lappe, Jim Bayer, Leo Kadesian, and there were some others who might have been on one call or another.

                The other decision that was made early on was that we would be looking only this time at the application of urine.  As Sam already indicated, the oral fluid working group ultimately included both lab-based testing and point of collection type testing.  I guess at some point in the future that we will have to merge those activities for certain commonalities.  At this time, we were looking at the things which were characteristic about point of collection devices, particularly for urine.

                There were nine questions that were submitted to the group and, because of the nature of the questions, there was some overlap.  I'm going to just try to go over some of the concepts without going through each question one by one.  Some were very simple, straightforward things.  Several were more generalized and conceptual.

                Many of the questions dealt basically with quality control and the ability to do quality oversight and what that would mean with point of collection test as compared with the laboratory.  All things were cautioned that, whatever happens in this program, as you heard already in the various alternative specimens, there has to be certain uniformity across similarity.  We can't make lots of exceptions of one type and be different for another type, merely because it's practical.  There has to be some general commonality.  This is going to be somewhat difficult.  You will note also that in the original regulations that were circulated the point of collection test, when we submitted to the laboratories, essentially we started the process at the laboratories.  There was again screening and confirmation at the laboratory, unlike some of the other things.  Therefore, there appeared to be some possibility to do a few things somewhat differently, and there was this concept of maybe there would be some idea of thinking outside of the box, that we might not be able to make this entirely the same as the laboratory.  Yet, because of other elements that would be in the Prime program that included both the regulation of the manufacturers, on-site manufacturer inspections, as well as a qualified products list, that some of the quality control elements might be somewhat different.

                The first question was fairly simple.  There was a question about what should monitor these negatives that we were suggesting being sent and a decision or a recommendation was made that the organizations who certify the tester should be the one to monitor that this process occurs.

                The bigger questions around QC came along the lines of how many to run, what constitutes a batch, how frequently to run them, because there were a number of organizations which could be using this test from two to three tests a day up to 20 or 30 tests a day, and the quality control aspects that were recommended in the guidance -- there was two suggestions; I believe one was to run a negative control and a plus 25 percent and another one was to run a positive and a negative each day.  There was major discussion over these issues and there's a lot of -- in several of the questions, there's a lot of discussion and opinion shown here.

                There was not a clear conclusion or recommendation made at this time.  But what was reflected was generally that there was concern that the recommendation which had a plus 25 percent and a negative might skew the cutoff by determining these -- might skew the cutoff.  There was a clear consensus that plus or minus 25 percent, as has been the traditional quality control criteria that a laboratory could not readily be met by these devices, but that a plus or minus 50 percent criterion could readily be met and a plus or minus 50 percent criterion was favored, so that the cutoff would not be skewed, as was recommended in the original guidelines.

                There were other considerations as to how frequently these needed to be done.  There was again considerable discussion without a final conclusion.  However, what was generally agreed upon conceptually -- and this is why I think it was important that the nature of the fact that this was a screening test which is going to be retested, might be, and manufacturer controlled is reflected here, in that there is not a practical way in an onsite field test that the quality control test result could dictate the outcome of the actual test which is being done in the field at that time.

                It's going to be a quality control mechanism, but the quality control mechanism did not need to proceed or follow and have a decision on the base of that made on each one of the particular tests that were done in the field.  There are two types of controls that are associated with these devices. 

 

MR. CROUCH:   Does that mean that a quality control sample could fail, but the result is reported to the donor anyway?

 

DR. CAPLAN:  Because it's going to the laboratory, yes.  That was part of the discussion, that unless you did a positive and negative prior to each specimen then you will invariably, if you do them at the beginning of the day or the end of the day, you will invariably have failures of controls for which specimens will have been recorded.  But all positives would go to the laboratory.

                There's two types.  Let me finish this.  There are two types of controls.  One was a process control -- that's the built-in control, and there was agreement that the process control must work, if the process control didn't work in the device that that test would not be used.  However, in addition to the process control there was the built-in control, and again there was considerable discussion that this process does not test the immuno-reactivity of the antibodies.  Certainly it doesn't.  Some of them, while they may test a little bit of the reactivity, didn't necessarily test reactivity for all the drugs.

                The process control is a fundamental control and must work.  If that control didn't work, then the test really was not completed and can't be done.  The external quality control, which is the potentially positive and negative controls which are run periodically, there was discussion about whether they should be done every batch of kits, every day.  I guess this was a divergence of opinion from the original guidance, which was the fact that they had to be run essentially with every positive every day and that would negate the positive or negative result.

                From a practical sense, it was felt that if there was a failure these failures would have to be dealt with.  One of two reasons for failure in the field.  Either the control material was bad, but it was unlikely that people in the field doing the field tests would have new control material, and in the laboratory you would go ahead and test your new control material, but in the field that would not be practical.  The other thing is that the kits for some reason deteriorated.  There was discussion of the fact that some of the kits might be temperature-dependent and whether or not they continued to be viable in the field is something which could be tested periodically and therefore you might detect by doing something by batch, by week, by month, depending on what it is.  There was experience presented by several manufacturers who have experience in programs where they started out with daily QC testing and switched it to weekly QC testing, and essentially are reporting that there are no failures.

                This is not the same kind of control, the manufacturer control over this, and the fact that this is all on a list and the fact that manufacturers are expected and are going to have to maintain documentation about the performance of this device over a period of time was thought to somewhat outweigh the need for a clear control with each individual test.  The key thing about the control process, the whole control process, was that that amounted to ensure that there weren't an excessive number of negatives which were sent to the laboratories.  As long as the cutoff can be confirmed and established, then the laboratories are defining specimens which go to the laboratories and so they can then be completely rescreened.

          The key thing was balancing out whether or not the kit could reliably produce results which are consistent with what's found later in the laboratory, and the means to do this may or may not be effectively done by requiring a positive or a negative control with each particular specimen or each day.  Again, there was not a final conclusion reached on this, but this was the summary of the discussion.  This was the concept of being outside the box.  If you could really monitor and ensure that these devices worked, that the reaction, that the flow occurred at each individual thing, device, at a time, and that there was a control mechanism, whether it be weekly or monthly, that this control mechanism could be adequate notwithstanding the individual control at the time that each test is done.

 

MR. CROUCH:  If a QC fails, all of the donor samples are going to the laboratory?

 

DR. CAPLAN:  Yes.  The only thing that would happen with the QC sample the way it was discussed here is that you would stop using that batch.  There may have been some between the time period, which has to be determined.  That time period may be weekly, it could be daily.  We could go back to daily.  But if it's not daily, it is weekly.  There would have been some specimens which might have been negative which are not going to go to laboratory.  But at the time that there is such a failure, and the reports from all the participants there was that the instances of failures was enormously low.  The failure would cause a new batch of materials to be used, rather than a specific test to be required.  This is still something which will require more discussion.  It was more of a thing that was out of the box.  That covers a number of the things.

 

DR. HUESTIS: Another big source of error is analyst error.  The kits may be fine, but the analysts who do the tests can make errors, and we've seen that.  I'm confused about what's going to happen if the QC fails.  You're saying QC fails and the last test they did was a week ago, that nothing happens to the samples that were done during that week?

 

DR. CAPLAN:  No, because all positives or all non-negatives would have gone to the laboratories.  There might have been some negatives that were wrong, that's possible.

 

DR. SAMPLE:  There might have been some false negatives in that scenario, but no false positives, because it's being re-screened and reconfirmed when it goes to the laboratory.

 

MR. IRVING:  The only thing that this makes the final decision on is the negative sample.  If they're sending out a false negative sample, that seems to be the biggest safety issue.  You're not putting a positive individual in a truck.

 

MR. STEPHENSON:  I think everybody understands this.

 

MR. IRVING:  You're having your laboratory be your quality control.  It will handle it for the positives, but not for the negatives.  You could turn out a large number of negatives incorrectly and find out at the end of the week that they have all these negatives there.

 

MR. STEPHENSON:  This is pretty well understood.  This is the quest for the perfect negative, on site, immediately, for immediate responses, only dealing with the negatives.  If it becomes an issue where there's a liability for accepting that negative to the person what uses it, like an employer who hires a group of people and they go out and have accidents or something and then later test again, this time test positive, and they say, well, how on earth does this happen, the system will correct itself through lawsuits to the company, to the manufacturer, to the employer.

                The issue here is we have traditionally looked at assuring that we have no false positives that go out.  We have assured everybody when we look at point of collection tests we have a mechanism for addressing a confirmation step and the re-screening confirmation takes care of all of that for the non-negatives and for the presumptive positives that come from the field.

                The only area that you wind up with is what level of false negatives are you willing to live with.  I think you've said that you have pretty good evidence submitted from the manufacturers that you don't have that as an underlying condition or general vulnerability to this assay.

 

DR. CAPLAN:  That was reported by several groups, one of which was a large laboratory that has an ongoing program that is currently doing this.  That's something -- again, this was the presentation to the group.  This particular group was fairly new and it was clear that after this goes through the DTAB next time there will be some revised questions that come back for more definitive answers.

                It was a thinking process and it was a very useful exercise at getting us into the course of some of the things we're going to have to make decisions on.  There was a question generally about whether the point of collection devices could meet the lower cutoffs, and there was general agreement that they could.  There was discussion about the MDEA and MDMA and it was felt, well, it's the same thing; it's a matter of the same antibodies, as discussed before.  There would have to be at least two assays and there might be some analyte-specific issues, but no one had looked at that really yet.

                We talked about the quality control thing that came back.  We were talking about it on each of the calls and it crossed over a number of the questions.  What I just talked about is this question on page 6:  Does lower QC standards, plus or minus 50 percent, lower the gold standard from drug testing?  There was a strong feeling amongst manufacturers that these devices are much more stabilized in some ways than the liquid solutions which are used in laboratories.  Hence, the stability could be documented in the manufacturing process and the plus or minus 50 percent mechanism to demonstrate the cutoff being required of the manufacturer should not adversely affect the equivalence of this for a urine test.

                There was very little discussion.  We didn't get into the fact as to what the use of this urine should be, because obviously, while it got modified in this particular draft, urine is used for everything.  Whatever urine is used for otherwise, it could be used for here, and more than likely urine will have to be maintained in most of the modes of testing, at least for a long while, until the other things are established.

                There was a discussion about validity testing.  Again, it was clearly agreed that validity testing could be conducted, in fact it might be better to conduct it, on site.  There were devices out there now, there were papers and tests and things, there was no problem with doing the validity testing in conjunction with onsite drug testing.

                There was a question about whether there were devices that could reliably show the substitution at 5 milligrams per deciliter creatinine and it was agreed that there are devices to show you less than 20, but anything less than 20 would have to be defined in the laboratory.

                The last thing of a general nature there was, there was some discussion, although it wasn't one of the questions, about the use of a separate collection cup, because we had written in the regulations that you cannot use the collection device, the test device, as a collection device.  Obviously, there's one or two products in which the kit is built in.  There was some general discussion about that, and that was kind of another question left open.  If there was a way that the manufacturer could ensure that this was not tamperable, then I suppose this could be considered.  Again, we left that one without finality.  That's basically what happened with the on-site discussion.

                They were all conference calls.  It was all very congenial, and I think we made a lot of useful progress using this mode rather than trying to have the sit-down meetings.

 

MS. CLARK:   Going back to your question number 1 on the first page, and this may be a general question, maybe something that I misunderstood in the draft guidelines, your answer says "The organizations certified to train the testers are responsible for monitoring the overall QC results and providing periodic reports."  My question is, it seems to me then that there couldn't be a training organization that provides the training only that is separate from the collection site manager, supervisor, employer or whoever you want to call it, that would on a day to day basis be ensuring that the QC was being done correctly.  I guess the short of my question is I envisioned that there would be an independent training organization that would be involved in the onsite collection training process in general, but they wouldn't necessarily be involved in the day to day overseeing of the actual testing and the QC.

 

DR. CAPLAN:   Again, this was for this particular group.  This is going to be an administrative decision eventually.  We are requiring 5 percent.  Who's going to monitor?  It was something passed back to the group.  We didn't feel that it should be the MRO, but there has to be someone designated of an administrative or semi-administrative nature to collect that data to ensure that these things go to the laboratory and that a report comes back.  Whatever the oversight organization is, and it's not likely to be -- it might be the MRO, but we are not going to say that that's the duty of the MRO.

 

DR. SAMPLE:  As I recall, there's a requirement for trained testers.  I think it was sort of a division -- that the group responsible for training those testers very well could be or might be the original manufacturer, but I don't think the draft guidelines precluded the possibility of a separate independent organization taking responsibility for training and monitoring the performance of those trained testers.

 

MR. STEPHENSON:  Which should be open for discussion.

 

DR. CAPLAN:  Right.

 

MS. CLARK:  My comment would be that I see those things as being completely separate.  The training organization is not the same as the QC organization.

DR. CAPLAN:  It could be.  This is just something that this small group was asked because it's going to be an issue.  Someone is going to have to ensure the quality for this diverse product.  Someone's going to have to take that and it's likely not to be the MRO.  It's likely to be one of the third party type organizations that does management of these things, which may use training or may use other, so you could use independent training.  That's not saying that it has to be the trainer, but there's someone responsible to ensure that the process is working.

 

MR. STEPHENSON:  I think, whoever that turns out to be, you just want to make sure you have the same objective review of potential conflicts of interest, so that you're not winding up with a product manufacturer underwriting this kind of review, and then having the potential for someone saying maybe this could mask some problem, is there an economic gain to be had by having some particular group do this?  That's a problem we've dealt with for the last ten years with the MRO's on the issues of lab testing and so on.  I think we can bring that into this environment.

 

MR. LoDICO:  If you look at the draft of the guidelines, to answer that question, in both 4.4 and also in 12.8 we ask the question what the requirements are upon an HHS collector-approved certification program.  He states that a mechanism to track errors made by collectors that cause a test to be cancelled, in order to identify those that need to be retrained.  That could be a QC error or it could be a QC failure, but also in 12.8 he also asked the manufacturers who want to become HHS-approved that they have to also develop a mechanism to identify testers that need to be retrained because of errors associated with incorrect use of the POCT.  In both sections you have conditions which they both have to comply with.  What we probably need to do is to focus on a sort of oversight of who has the initial responsibilities for the oversight and then who has the responsibilities with HHS

 

DR. CHILDS:  Maybe I'm missing something here, the laboratory's report back to the tester that an error was made when the device is read.  The testers then report that back to their certifying agency.  Isn't that like asking the fox to guard the henhouse?

 

MR. STEPHENSON:  I think your point is well made and it speaks to the same issue that we talked about, that we need to look to make sure that, whatever we do, it's objective and it addresses the issue of potential conflict of interest.

 

DR. CHILDS:  I was suggesting there is that maybe they could report it, not only to the tester, but also to their agency.

 

MR. STEPHENSON:  It could be a range.

 

 

Agenda Item:  Public Comments

                                                               

MS. OCHS (Roche Diagnostics):  I'm here to ask you to consider allowing the flexibility to use certain integrated sample collection testing devices.  I'm here for Roche, as that's what we have with our TesT cup products.  As Dr. Caplan said, it was an item that was discussed briefly, but I came here today to show you the reasons that we want you to consider this.

                (Screen.)

                I have data to show you, a couple of graphs.  The urine sample is more stable in our TesT cup than in other plastic cups, and I think you want to be sure that the sample, whatever it is put in, is stable.  With the TesT cup, there is virtually no risk of device tampering that would give an invalid result.  It has been known for a long time that certain drug compounds bind to plastic surfaces.  In 1993, an article by Blank, et.al., in clinical chemistry showed that cannabinoids bind to certain plastics.  They found significant losses of cannabinoid when they used a plastic pipette.  They saw that the surface to volume ratio of solution in the cup was related to the amount of loss of the drug.  In glass, they felt like glass was the gold standard for storage of these samples and they showed a loss of about 2 percent, and in the plastic cups to upwards of 12 percent or so.

                (Screen.)

                These graphs show you the stability of the HHS-5 drugs stored in the TesT cup or glass containers.  You can't tell the glass from the TesT cup because they're exactly the same, except for right there where the TesT cup did just a little bit better.  These samples were stored for six months at either 4 degrees or at minus 20 degrees in the cups, and they were assayed by GC/MS.  We have a fresh, zero time point, and 6 month time point (indicating), and you can see there that the product did not degrade in the TesT cup significantly more than it did in the glass.

                (Screen.)

                In contrast to that, Roche published some data in 1995 on all the HHS-5 drugs and we repeated this just last year to see if the situation was still the same, where we put THC into a TesT cup -- that's the black diamonds that you see at the top -- and into two other cups.  These were popular plain sample cups that we purchased.  These are used all the time.

                You can see at room temperature, after two days in the cup on this blue cup, we had significant loss of THC, enough so that had that sample started out over the cutoff of 50 when you send it in for your second screening it very well could have dropped below the 50 and given you a negative.  The same with the yellow cup.

 

MR. STEPHENSON:  I don't want to intrude too much here, but has this information been shared with the point of collection testing working group?  Has it been considered as a part of that process?

 

MS. OCHS:  We had some discussion on it, but I don't believe this data was shared.

 

MR. STEPHENSON:  Are you participating in that working group?

 

MS. OCHS:  Yes, Sal is.

 

MR. STEPHENSON:  In that context, I want to make sure we set a standard for presentations in the public sessions and we have a presentation of various working groups, in fact, that is the primary mechanism for individual industry participants to have input and discussion and potential for consensus developed within those areas for product lines, and that we don't use this time in a public session as a marketing issue.

MS. OCHS:  I'm not trying to market.  I'm trying to draw to your attention that there's no regulation today over sample cups and you are putting a regulation in that forbids the use of my cup.

 

MR. STEPHENSON:  Wait a minute.  Who has done that?

 

MS. OCHS:  Well, it's not done yet, but there is language in the last draft that does not allow the use of these kinds of cups.

 

MR. STEPHENSON:  And you have responded to that?

 

MS. OCHS:  We have.

 

MR. STEPHENSON:  In writing?

 

MS. OCHS:  Yes, we have, every time there was a draft.

 

MR. STEPHENSON:  And the issue has been addressed in the point of collection testing?

 

MS. OCHS:  Not to our satisfaction.

 

DR. VOGL:  Your point is that you object to collecting urine in a typical specimen cup and then pouring it into the TesT cup, rather than the donor urinating directly into the TesT cup.  The issue is that the transfer is going to potentially lose 12 percent of any metabolite that could have been in the urine?

 

MS. OCHS:  That's right.  When you make your list of approved devices, maybe you better ask for data on the sample cups as well.

 

MR. STEPHENSON:  How did you address this particular issue and concern of yours to DOT when it came out with the revised Part 40?

 

MS. OCHS:  I don't know what DOT did.  We have responded to this issue for each one of the drafts of the HHS Mandatory Guidelines.

 

MR. STEPHENSON:  I'm thinking more specifically of the standard kit that DOT is now stipulating.

 

MS. OCHS:  I can't answer that question.

 

MR. STEPHENSON:   I'm not trying to stop you, but I want to focus you on what you need to communicate to this group at this time and hope to re-engage you in the process both of formal written responses to the draft guidelines, because these are just the informed interested parties right now.  The guidelines will be coming out as a formal public Federal Register notice at some time in the future and, as with all of these cases, we must address the issues that are raised.

We are not trying to exclude anyone, but for point of collection testing, what you are saying is that the point of collection testing group has not listened to Roche, has not addressed Roche, and  Roche feels that that process was not adequate, therefore, you needed to come to this group at this time to make this presentation.

 

MR. LoDICO:   I was part of some of the telephone conversation and Dr. Salamone did express his concern about requiring that the integrated device would not be used in the collection of the specimen.  After we had further discussion and stated that there is a DOT requirement for split specimen collection, this issue is moot because you still need to split that specimen before you can even do a test.  You can collect the sample in a non-integrated collection cup, and then pour half into your TesT cup and the other half in your split specimen cup which is sent to the lab.

 

MS. OCHS:  I agree with you, that is the case for these mandated guidelines.  It is my understanding that FDA is going to point all workplace testers toward the SAMHSA guidelines and we know, for a fact, that they are not all splitting their samples.  We would not like this guidance to read that these kinds of devices are unacceptable for use because we believe we have data that they are definitely safe and effective, acceptable for use, and we'd like to have that data considered.

 

MR. STEPHENSON:  I suggest that you make sure that it is provided in a written format, you can leave a set of slides for us.  For any other discussions, please make sure that you have full participation, as we do with all of our participants in this work group.  Are there any other things that you would like to summarize at this point?

 

MS. OCHS:  That's it, thank you.

 

DR. SAMPLE:   First, a question that I'm not sure that I heard answered.  It was for you on the oral fluid working group.  I guess this relates to question 7, but it's right above question 8.  It says:  "Address split specimen," and there's no comment there.  Does that mean that the oral fluid working group is still silent on how they are going to address the split specimen issue?

 

DR. NIEDBALA:  Well, I'll just say yes.  We didn't delve into that nearly as deeply as we should, but there's all kinds of devices that collect.  The ones that go back to the lab obviously have a fluid that's received at the lab.  Others, there'll be a cassette that may have a remnant of a specimen as a point of collection.  The split specimen, at least in the initial discussions, was incumbent on the manufacturer to describe the best way for their specimen to achieve this goal.

                               

DR. SAMPLE:  There are certainly many ways one may choose to address that.  I didn't really see any discussion on that here.

 

DR. NIEDBALA:  No.  This was one of those issues that was sort of everybody normalizing back to whole oral fluid.  There was no one answer that seemed to fit.  I think in part we've gone on to the other issues that we felt were most important to answer, namely how much do you collect, what are the cutoffs, what are the analytes, and that's as far as we got.

 

DR. SAMPLE:  Maybe I could infer that the participants feel that they can address this issue and that it's not really an issue any more, that they have solutions of some sort?

 

DR. NIEDBALA:  Yes.  I would say, though, that the solutions vary.

 

DR. CAPLAN:   One of the things that did come up in those discussions was the fact that, unlike urine, in some of these other specimens there are likely to be many more device and product-oriented requirements that are different, that you have to have 30 plus 15 of urine, which was very clear.  There was resistance to setting an amount which had to be collected.  If, for example, a manufacturer could use a much smaller amount for the screening and there was an adequate multiple time testing, that this might be 0.3 mLs for one, it might be 0.6, or it might be 1.0.  If you want to cover everybody, you might have to collect a much larger specimen of a mL or more.  There was just resistance to making the decision based on things at the time, but everybody felt that they could do it.

 

DR. NIEDBALA:  Yes, absolutely.

MR. STEPHENSON:   At this time, we will end the open session.

The meeting adjourned at 12:49 p.m.