We discuss the advantages and the limitations of comprehensive drug detection methods, and the quest for a truly universal method.
Contact us at toxpod@tiaft.org
Find out more about TIAFT at www.tiaft.org
The Toxpod is a production of The International Association of Forensic Toxicologists. The opinions expressed by the hosts are their own and do not necessarily reflect the views of TIAFT.
We discuss the advantages and the limitations of comprehensive drug detection methods, and the quest for a truly universal method.
Contact us at toxpod@tiaft.org
Find out more about TIAFT at www.tiaft.org
The Toxpod is a production of The International Association of Forensic Toxicologists. The opinions expressed by the hosts are their own and do not necessarily reflect the views of TIAFT.
Hello and welcome to The Toxpod. I'm Tim Scott.
Peter:And I'm Peter Stockham.
Tim:Today we're going to talk about the holy grail of forensic toxicology testing. When you're testing biological samples for drugs and poisons, there are a huge amount of things that you can be testing for.
Peter:Thousands.
Tim:Thousands. So surely the best way to cover as many drugs as possible in the most efficient way possible would be to have one method which looks for everything. Right, Pete?
Peter:Yeah. I'm sure that's what a lot of people who aren't involved in the field think we can do.
Tim:Well, I think people do try to do it.
Peter:To different degrees of success.
Tim:There's quite a few methods published which have hundreds of drugs in them. Even some which have up to a thousand compounds. Have you seen any publications with methods with a thousand drugs?
Peter:Oh, ones that aren't really validated. I don't know, I like a big screening method to have a known detection limit for all of your drugs.
Tim:Yeah, because you can have a database which has thousands of compounds in it.
Peter:Yeah, like GC/MS screening or...
Tim:Yeah, if you're using a GC/MS library, I mean some of those have tens of thousands of compounds.
Peter:Well hundreds of thousands. But can you reliably say that you can detect all of those compounds?
Tim:Yeah, that's the issue, isn't it.
Peter:We're talking here about large methods, not necessarily ones that are designed for a specific purpose like say, for example, if the legislation says to look for specific drugs, you don't want to screen for everything, do you?
Tim:No, in some cases it's completely appropriate to just have a small targeted method.
Peter:Yeah, you might have a two minute method that does three drugs and you can do thousands of samples a day.
Tim:Perfect.
Peter:Yep, but here we're talking about methods for systematic toxicological analysis. STA methods. Not your small methods which are made for a specific purpose, like maybe drugs in drivers or, where you're looking for three or four analytes that are in your legislation or drugs that are part of a standard.
Tim:No, I think it's where, where you want to do broad screening, but the question is do you have one method that does that broad screening or do you have lots of smaller, more targeted methods to do that same job?
Peter:So are we talking about, you're meaning a whole range of chemically diverse, you don't just mean 100 neutral and acidic drugs or a thousand basic and neutral drugs. You mean everything, don't you?
Tim:Yeah, I mean everything.
Peter:From paracetamol to zolpidem.
Tim:Yeah. I thought you were going to do like an A-Z thing there for a minute.
Peter:Um, acetaminophen to zolpidem.
Tim:Yeah, nice.
Peter:Yeah, so not just a thousand basic drugs. You want to do 10,000 everything.
Tim:Well, a method that only covers drugs of one pKa for example, isn't really a comprehensive method. Because there's a lot of other things that you're not going to see. The major advantage of having one big method is efficiency.
Peter:Yeah provided the method doesn't take two days to process.
Tim:[laughing] Oh, that's a huge problem, isn't it? Because the amount of data that you can get out of one method is phenomenal.
Peter:Yeah, you can spend a lot of time looking for common compounds that you find in hairspray.
Tim:But operationally if it is reasonably efficient in terms of the processing, the amount of staff time that you spend is less with one method than with many methods. You're using less sample probably, with consumables, you're relying on the same consumables, so you're getting economies of scale there.
Peter:Training. You can have many people doing the same method.
Tim:Yep. Instrument maintenance, things like that. If you're relying on the same instrument or maybe a couple of identical instruments to duplicate. Preparation of working solutions, all of these things. If you've only got one method, you can have a really efficient process. But the efficiencies aren't just in terms of the daily operations as well, it's also in terms of the initial validation. It's simpler to validate one method than to validate many methods, right?
Peter:Yeah, it is.
Tim:Although having validated some large methods, the amount of data that you have to process in that primary validation is huge. That is a massive undertaking.
Peter:Yeah, if you're going to thoroughly validate every compound, say if you've got a thousand compounds, you can never do that. So most people would tend to validate model compounds perhaps. Especially for qualitative methods.
Tim:Yeah. What are your thoughts on that Pete? Using model compounds in validation?
Peter:Sometimes I think it's unavoidable if you've got a massive method like that, but having said that, I would really think that if you were analyzing a sample, you would want to be positive that a compound wasn't present. So I think your client, well many clients, would like to have the confidence that you know what detection limit you're able to get down to for that particular compound.
Tim:Especially in a forensic setting, right?
Peter:Yes, yes. But validation for a thousand compounds is virtually impossible I think.
Tim:And sometimes it's hard to get reference standards. You might have a drug that you really need to include in your method, but you just can't get a reference standard for it, but you might be able to find something that's analogous to it.
Peter:That's true, but is it valid, do you think?
Tim:I do think that with compounds that are really similar in structure, that you would think that they're going to behave the same way. Sometimes they don't. I mean we know that from experience.
Peter:You're gonna talk about the benzos aren't you Tim?
Tim:[laughing] Yeah, look, even the benzos, even though they all have the same backbone to them, they have quite a range of different functional groups on them, so you can't really expect that one benzodiazepine is going to behave the same way as another one at all points throughout the extraction and on the instrument...
Peter:And the instrument as well, yeah.
Tim:And that's the same for lots of other classes of drugs as well, so I think that that is a downside of having one massive method if you can't validate all the compounds that are in that method because model compounds, although sometimes necessary, are not really the ideal situation. But another advantage of having one major method is that you can have duplicate systems in place, you know, duplicate instruments, duplicate setups in the lab or maybe even three or four different instruments depending on your throughput. So then if anything goes wrong, you can still keep going with that method. It's harder to duplicate lots of smaller methods.
Peter:Yeah, and you can also export your method to other sites if you're a multisite facility.
Tim:Yeah, that's a good point.
Peter:I think in the environmental field it's not uncommon to have standard methods that cover many thousands of compounds, but I don't know to what extend the validation goes to.
Tim:I mean, it would be nice in a way if that could be duplicated in the forensic field because what seems to happen is everyone has their own methods, but it's tough because everyone's got slightly different instruments and, and people that are looking for slightly different things as well, so...
Peter:And every lab seems to have their own historical extraction method developed by a certain person who likes that and then wants it to carry on forever!
Tim:[laughing] Or it's probably that's just what they had in the laboratory at the time when they developed the method and so that's what they're using now forever.
Peter:So isn't the root of all this the extraction method, Tim, or is it the instrument as well?
Tim:You mean in terms of detecting a whole range of different compounds at once?
Peter:There's currently no instrument that can detect every compound, is there?
Tim:No, but I think the extraction is as much of a problem as the instrument. One of the benefits of having a large method with hundreds and hundreds of compounds in it is that sometimes you'll find things that you weren't expecting to find, these unusual compounds which are in your method but which you don't find very often. If you have a lot of smaller methods, you're not necessarily going to use all of those methods on every case. You'd be relying more on information that's given to you from your client, the police, the coroner, whoever...
Peter:Yep.
Tim:And you might not do the particular test that looks for that compound. Maybe a pesticide for example, or even just like an anticonvulsant drug, something that's a little bit unusual, not that common, and so you might not be targeting it in every case if you've got these smaller targeted methods.
Peter:It would be good just to replace all that and have one method, wouldn't it. But if you're going to have a humongous method, any humongous method you're going to have to compromise for some analytes. Some analytes may not have as good an extraction efficiency or maybe they have a poor detectability on the instrument you're using, might have different matrix effects.
Tim:Yeah, Matrix effects are a huge issue I think in large methods, because with smaller methods, you can really optimize the extraction for that class of drugs. You're going to try and extract out only those drugs and nothing else. I mean you can't get nothing else out, but you're going to try and get out as little else as possible, both endogenous compounds and other drugs, but in a larger method because you have to go for a more generic extraction procedure, because you have a variety of different functional groups, oolarities, PKas, the compounds that you're looking for are so different to each other. You've really just got to try and get everything out, which is why a lot of people use things like protein precipitation, if you're talking about blood or serum...
Peter:Salt assisted liquid, liquid extraction where they use saturated salt and a bit of polar solvent to get all of the drugs out of there. it's a pretty universal method.
Tim:Yeah. Quechers, as a lot of people call it these days. Well, some people call it Quechers, some people call it Ketchers, some people will call it Catches.
Peter:I think it's Catches.
Tim:Well, I say Quechers and I'm sticking with it![laughing]
Peter:And then you've got the problems, say for example, you're analyzing a urine for a low dose fentanyl and this person also has taken paracetamol, a high dose drug. And in the urine, the concentration of that drug could be in the thousands of milligrams per liter and then at the same time you're trying to look for a very low dose, down to nanograms per mil concentration, and if they coelute then I'm pretty sure you're going to get some massive on suppression.
Tim:Oh yeah.
Peter:And if your extraction method relies on say a solid phase extraction, there may be limited active sites for the analyte to be absorbed to. So I think that could have a big influence.
Tim:Yeah. So that can affect your recoveries, and I mean you can't have a deuterated internal standard for every compound in a large method., I mean that's ridiculous. And those kinds of methods, the Quechers, Ketchers, Catches, whatever...
Peter:[laughing] Catches! It's Catches!
Tim:They generate a lot dirtier extracts as well, which can cause problems on your instrument over time.
Peter:Sometimes in Quechers to help with matrix effects, they add the dispersive SPE component. It's been shown that that can remove a lot of your analyte as well as you matrix effects.
Tim:Yeah. The good news is you've got no matrix effects, the bad news is you've got no analyte! But on the other hand, talking about matrix effects, I mean some matrix effects are caused by endogenous compounds, but then a lot of matrix effects are caused by coeluting drugs, as you mentioned before. In a broad screening method, at least you can see those other compounds that are coming out if they're ones that you're monitoring for, which they probably are, if they're common compounds.
Peter:So you mean like in GC/MS for example, you'd normally run a method that's in full scan or if you're doing a large screening method.
Tim:Yeah. Really any full scan method, I guess
Peter:TOF-MS? LC or GC.
Tim:Yeah. Although some have suggested that this is something that you should be checking in your validation, that you should run a bunch of common compounds that you might expect to see just to check where they coming out and what possible matrix effects they might have with the drugs that they're coeluting with. But I mean, I, I don't think everyone does that.
Peter:I don't know if it's required to do matrix effects, I think it's more to see if it's detectable. It's more selectivity than matrix effects.
Tim:Yeah, when you say required, you're talking about like, um, you know, the SWGTOX guidelines or whatever.
Peter:Yep.
Tim:Yeah, I'm not so much talking about the guidelines. I'm just talking about some scientists who have done work in this are have suggested that this is something that should be done during the validation because if you don't and you've got a coeluting compound and it's a high dose compound and it's commonly seen, but you're not actually screening for it. You're not going to see it in that targeted method. But it's still there. It's still coming out and it still could be having effects on the drugs that it's coeluting with.
Peter:It's the elephant in the room, Tim...
Tim:Yeah, it is.
Peter:That no-one talks about.
Tim:But how do you even go about validating that? I mean, how many different combinations and permutations of compounds do you look at? How much is this one suppressing this one and how much is this one suppressing the other one?
Peter:Oh, it's impossible. If, you need a better detection method that doesn't have any influence from coeluting compounds, and at the moment ESI and APCI are the only ones we've got for LC/MS that we use commonly.
Tim:Yeah. We really need a better mathematical modeling of how matrix effects work and so you can predict how various things are going to be affected in your method, but unfortunately no one really understands matrix effects, so...
Peter:That's funny, isn't it? No one understands... well, I think we've got general theories, but in, on any one day you never really know what the cause of matrix effects is, they always say it's phospholipids, things like that, but it could be any number of other compounds that are in your sample.
Tim:Yeah, but not even in terms of just the compounds, but even just how it actually works. Droplet formation and charge dispersion and things in the source. It's not completely understood.
Peter:Is Matrix effects happening in the droplets? Is it, are ions being stopped from exiting the droplet? Are ions being intercepted by other ions that are in the gas phase on the way into the source? No one really knows. So I don't know how you could use a mathematical model.
Tim:No, and we're going off on a bit of a tangent there, but my point is that in a large method where you're monitoring a whole range of different compounds, at least you can see those other compounds coming out.
Peter:As long as they ionise, I guess.
Tim:Yeah, I guess I'm talking about compounds that are in your method. You can't find everything, but at least if you've got several hundred compounds that you're actively looking for, you're going to see that there's a large ketamine peak that's coeluting with your fentanyl peak, for example.
Peter:Or your, a large ketamine peak coeluting with your deuterated ketamine that you actually use as an internal standard for another drug.
Tim:[laughing] Yeah, that is a whole other issue.
Peter:Maybe we should't talk about that Tim. I don't think people like talking about it.
Tim:Maybe we don't have time to talk about that in this episode.
Peter:It would be interesting to do a whole Toxpod on matrix effects because there's a whole lot of different issues that we could explore there.
Tim:Mmm. Another advantage of having one large method is that you can really focus all your efforts on it to get the most out of it, because with any method, it doesn't matter whether it's a big method or a small method, there's always just those little things that just really take a lot of time, whether it's the instrument report, getting it just how you want it...
Peter:So you're talking in developing a new method?
Tim:Not just developing, but actually using it in operation as well, because often you'll develop a method and then you'll improve it over time, you know?
Peter:Yeah. It gets refined. Through natural quality incidents that occur.
Tim:[laughing] Yeah, every time something goes wrong, you refine it. Yeah. But with one method, staff become very familiar with it so they know how to deal with any issues that come up.
Peter:So you've got all your scientists who only know one method. Is that what you're saying, Tim?
Tim:[laughing] Yeah, I guess that's the downside of it. It doesn't make for the most rounded scientists, if you do just have one method.
Peter:It'd be a bit of a boring life I think.
Tim:Yeah, and I mean the good thing about having lots of smaller methods is that your staff do get to use a lot of different instrument techniques, extraction techniques, and they just become better scientists over time because they're exposed to different things, fixing different types of problems.
Peter:I think any scientist would look a bit of variety in their working life, wouldn't they Tim? You don't want to be stuck doing the same job, I mean, I know it's not always possible, but surely most responsible bosses would try not to kill their staff's enthusiasm, by making them do one thing for 40 years.
Tim:Well, I think, I know in my case, I think I'm a better scientist for having been exposed to lots of different things and trying lots of different new things as well.
Peter:Unfortunately, often there's a narrow range of employment opportunities for specialized people, people who specialize in our field, so yeah, if they've got a job, they just stuck there...[laughing]
Tim:What?
Peter:In a boring, boring life.
Tim:Oh, I see, if...
Peter:All they ever do is...
Tim:One method.
Peter:They're chained to that GC. We should try and release these people Tim. Release them![laughing]
Tim:Yeah, so I guess there are some advantages to having more than one method. It does help to develop the skill level of your scientists.
Peter:And it also, moving from one method to another, you can transfer knowledge and improve other methods at the same time.
Tim:Yeah, that's true. You never know what's gonna, you know, you learn something in one area and it can pay off in another area. You don't know that. It's hard to predict that. And talking about method refinements, if you have one big method, every small refinement that you make to the process of that method has a really big payoff because you're putting everything through that method. Whereas if you have a lot of smaller methods... I remember I developed a method for hypoglycaemic drugs, uh, antidiabetic drugs, and the aim of developing that method, my aim in developing it was that whenever the pathologist, for example, wanted to screen, if someone had diabetes, they wanted to screen for antidiabetic drugs, then we'd be able to use this method. And so I developed it, validated it, did some refinements over the next couple of years, but it never got used as much as I had thought or hoped that it would be used. And so I think with smaller targeted methods, you put a lot of energy into them to make them as good as they can be because you want your methods to be robust and to be efficient operationally, but there's not always the payoff that you expect or want because sometimes they just don't get used very often for a number of reasons.
Peter:Yeah. So for your investment, it's better to do a larger group, a more diverse group of drugs, rather than just a targeted therapeutic class.
Tim:And every method that you have requires that peripheral support as well, you know, IT support, maintenance, troubleshooting...
Peter:Linking to your LIMS and that sort of thing.
Tim:The more methods you've got, the more of this stuff you have to do because all of this stuff really is unique to each method.
Peter:There's also your quality manual as well, you have one method.
Tim:Oh,the number of procedures you might have if you've got a lot of smaller methods.
Peter:I remember we used to do benzodiazepines on a GC/ECD and basic drugs on a GC/NPD and anti psychotics on an LC/MS. That seems absurd now.
Tim:It does seem funny now, doesn't it? To think how things used to be done. Everything was done on a very specific instrument with a specific detector.
Peter:But LC/MS has really changed the way we're looking at analyses now.,I think.
Tim:Yeah, and it's obviously better because otherwise people wouldn't be moving towards it. I mean, there's clearly advantages in it and you can have more comprehensive methods using LC/MS.
Peter:But there's also those dangers that we keep talking about.
Tim:Yeah, which, maybe we're still, you know, if you think of the history of forensic toxicology, we're probably still in the infancy stage of these massive comprehensive methods.
Peter:Yep, for sure.
Tim:We couldn't really do it until fairly recently, and so people are doing it more and more, but we're learning about some of the limitations, I suppose.
Peter:You mean in terms of the ionization variability and things like that?
Tim:Oh yeah, so there's technical limitations, but there's also just operational limitations like, uh, preparing working solutions.
Peter:So you mean you've got hundreds of compounds in the same working solution. We never have trouble with those do we Tim? I've asked many labs, I say do you ever have trouble with the stability of, you've got 100 drugs and one of them's gone off. What do you do? Oh, we never have that trouble.
Tim:Even just the logistics of how you prepare those solutions, you know, you've got all your reference materials, maybe there's hundreds of drugs, so you've brought in all these reference materials from a supplier and so you want to make up a mixed working solution that you can then use in your method because maybe you're quantifying 100 or more compounds and so you need calibrators and so on, for all those, QCs.
Peter:Every run, yeah.
Tim:So you've got to prepare a mixed working solution. Do you prepare it directly from the reference material because that's obviously the quickest way to do it, but if you make a mistake with any one of those 200 drugs, that's going to be a mistake that lasts for quite a long time until the working solution's expired. Or used up. And when I say a mistake, I don't mean a mistake that you're unaware of, I mean a mistake that you know about, but you can't fix it at that point. Short of making up the whole thing again.
Peter:So the concentration of a particular drug is not what it was supposed to be in the certified material, perhaps? That can happen. You put your mix together and all of a sudden you find out that this drug is incorrect. Has that working solution been a waste of time then?
Tim:Well, I guess there's a range of ways you can deal with that. You can just remove that drug from your quant method for the time being and quantify it some other way, but you know, what a lot of people do is they'll make intermediate solutions. So then you might have half a dozen or maybe more of these intermediate solutions. So if there's a problem in any one of those, at least you can isolate it and you don't have to go right back to the start and start the whole thing again.
Peter:So you have to remake that solution of 50 drugs instead of 300.
Tim:Yeah
Peter:And so what about stability of these compounds?
Tim:Yeah, we talked about stability issues in a previous episode.
Peter:Some drugs have a very limited life in these solutions, like maybe zopiclone in some solutions goes off,. promethazine. Often if you've got a single solution that you may use for 6 or maybe even 12 months, if you're using it every day, taking it in and out of the freezer, I think there can be an issue. I mean, who has never left a solution out on the bench overnight.
Tim:I certainly have never done that Pete.
Peter:[laughing] So sometimes it's a good idea to separate them into little tiny aliquots that you may use for one analysis, so GC vials or something like that. So it's just got the required volume that you need for that particular calibration run.
Tim:Yeah.
Peter:I remember a study we did where we were looking at, it was a matrix effects study and we couldn't work out why we could not get two solutions to match and the reason was that one of them was prepared in GC vials at a very low concentration and some drugs were just sticking to the glass, so I don't know whether in some cases you can actually exacerbate problems by having them split up into smaller solutions because there's a larger surface area of glass there. So that could be something to watch out for perhaps. Then we silanized the glass, but then different drugs stuck to the silanized glass, and the other drugs were happy. We couldn't win.
Tim:With multi-class screening methods, you're often going to be looking at several orders of magnitude in terms of the concentrations of drugs that you expect to find. You're going to have some drugs which are very potent, which have very tiny concentrations, but then other drugs are going to be present in huge concentrations which are going to saturate the instrument potentially.
Peter:So it's quite difficult to mix those two different groups of compounds together.
Tim:Yeah, so you might need to look at creative ways of quantifying like uh, using the isotope ion, or having segments throughout the run where, in a particular segment, you're deoptimizing certain parameters for different compounds.
Peter:Yeah, or if you're using a triple quad or a targeted method with lots of drugs in it, you can deoptimize the transitions so that they have a lower response at the detector, but still adequate.
Tim:Because the problem is if you've got one method to cover all of these different things, you've got to choose the volume of sample and the final volume of the extract and the injection volume on the instrument, and those are all constants across, across the method. So then there's very little scope to, I guess you're making the instrument do a lot of the hard work then, the detector, in terms of reducing the signal of some and increasing the signal of others.
Peter:So you might have to do multiple injections instead in some methods because there's just no way you can detect everything at once.
Tim:Yeah, which is pretty common. A lot of people do multiple injections.
Peter:Sometimes positive mode, sometimes negative mode if they can't do them both at the same time.
Tim:Yeah, or even just injecting a lower volume and then a higher volume. Or even, some methods will split the extract halfway through, so you're sort of getting to the end of your extraction and then you split the extract in two and you'll concentrate one and dilute the other one, so then you've got these two extracts and maybe you're running them on different instruments or maybe you're running them on the same instrument, but you're just going to get different responses.
Peter:You're going to have a long run time, if you have in the same instrument, I mean.
Tim:Yeah, it's not ideal. I guess we keep coming back to that ideal situation. It does really seem to be one method with all drugs in it. That's the ideal situation. One method to rule them all. Lord of the Rings references aren't outdated are they Pete?
Peter:I don't know they might be copyrighted though.
Tim:[laughing] Okay, I'll cut that bit.
Peter:So you can't do it Tim, is that what we're saying? There's just too many problems. That's why it hasn't been done yet.
Tim:Yeah. One day, one day you'll be able to do it. Or maybe not. I don't know.
Peter:You're dreaming.
Tim:Someone has to invent, I mean, I really think the current LC/MS instruments, I mean they're the best thing we've got at the moment in terms of these comprehensive methods, for forensic toxicology anyway. I'm not sure that improvements to LC/MS are going to be the final solution to this problem though. If there is a fully comprehensive method that can be developed at some point in the future, and maybe there's not, I think it's going to be something other than LC/MS.
Peter:Well, that's interesting. There's other methods around. There's um, what about supercritical fluid MS? Sort of a cross between LC and GC. There's commercial applications of that these days. I'm not quite sure how it goes with, whether there's more or less ion suppression. So compressed carbon dioxide goes through a special column that's got a special phase on it, uh I don't know much about this obviously, but the advantage of it is at the source it's very easy to get rid of the mobile phase, so you just theoretically dealt with hopefully ions in the gas phase rather than salts floating around.
Tim:Yeah, because that's half the problem with LC/MS, it's just that ionisation issue.
Peter:And it's a pretty low ionisation efficiency too, I think, in LC/MS.
Tim:So, who knows. Who knows what'll be invented next. But you'll hear about it first on The Toxpod![laughing] No, probably not.
Peter:So ideally you'd have one method to do everything. I mean if your extraction method's good enough, and your equipment's got the capable range from low concentrations to high concentrations, and it's got the linearity...
Tim:But it's just a question of trying to overcome some of the limitations, and I really enjoy reading all the literature about new and creative ways that people are trying to overcome those limitations. And speaking of great research, FACTA is coming to Adelaide in 2019, that's the Forensic and Clinical Toxicology Association, which is the Australian regional association, and we have a conference every couple of years. And it's coming to our hometown Pete, in Adelaide.
Peter:Yep, it's going to be good.
Tim:FACTA President Dimitri Gerostamoulos will be there. TIAFT president Marc LeBeau will be there. Pete, you and I will be there. If you're anywhere in the region, come to the FACTA conference in Adelaide, June 16 to 19 in 2019.
Peter:And if you want to get in contact with us, just email us on thetoxpod@SsadotGov.ae you. Thanks for listening. We'll see you next time.