Prof Roy Gerona UCSF monitoring for New Psychoactive Substances (NPS)

Hello, my name is Anna Najafi. I'm the director of Strategic Pharmaceutical Development here at Emory Pharma. We're excited to welcome back to Emory Pharma Dr. Roy Girona. Dr. Roy Girona is an expert in bioanalysis and the development of mass spec techniques to monitor environmental chemicals. He also develops techniques to analyze new psychoactive substances or drugs of abuse. A lot of his work is done in collaboration with poison centers, medical examiner's offices, government agencies such as the DEA or the cdc. We're excited to hear his presentation today. I think it will be very exciting and educational. With that, I'm going to hand it over to Dr. Girona.

All right, thank you so much for the introduction, Anna. All right. I thought I'd start with telling you a little bit about myself. I think there's a story I wanted to share, and I'm not sure whether when I was here, I shared the story of how I started because I actually start. My trajectory in running a lab is a little bit different from the usual academic. I'm currently an associate professor at the center of Reproductive Sciences at the Department of Obstetrics, Gynecology and Reproductive Sciences at ucsf, but I'm a mass spectrometrist. So why am I in an ob GYN department? And that's always a question when I give talks about designer drugs. Wait, you're an ob GYN and you're getting talks in nps? When I started my postdoctoral, I think my aim is really to move away from academia. And the reason is when I did my PhD at University of Wisconsin, I became friends with assistant professors, and I saw how they lost their personal lives trying to get tenure in academia. And I told myself towards the end of my career, I'm not going to do that to myself. I'm going to move to translational research. And so when I told my principal investigator at the University of Wisconsin that I'm going to clinical chemistry instead of biochemistry and molecular neuroscience, he was very disappointed. However, when I started at UCSF doing clinical chemistry fellowship, my principal investigator, Alan Wu, introduced me to high resolution mass spectrometry. And I chose UCSF as a clinical chemistry fellowship because UCSF has a very strong emphasis on drug addiction. So I thought, I can still connect drug addiction with molecular neuroscience. We started this program with the Poison Senders, where we're doing drug sleuthing on cases. Like, you know, there is a mysterious case in the emergency department. They don't know what caused intoxication. We will try to solve it. And I thought that was Very, very interest. It's a new drug, different drugs every time we have cases. And so towards the end of my postdoc, my TI wanted to keep me. However, for clinical chemists, usually you do your two year postdoctoral fellowship and then immediately after that, it's actually a really great deal because you become an associate director in a clinical lab, basically overseeing testing, a lot of the testing that you normally would get when you visit the doctor. However, at the time that I graduated from my fellowship, there were two other postdocs before me who are already associate directors at SFGH and ucsf. So there is no available position, but my PI wanted me to stay at ucsf. So he had this grand idea of, why don't you start your own lab? It's as simple as that. Start your own lab. Where will I get the funding? Well, at the time that I was finishing my postdoc, my PI is Chinese, but he actually was born and raised here. His father is a Chinese engineer who moved to Illinois and became a farmer, but raised two postdocs and two medical doctors, so very successful children. So he died around the time that I was actually graduating, and he left my PI some money and basically said, why don't you use this to support young Chinese scholars in academia? And so he wanted to keep me. One afternoon, he called me to his office and he asked me, do you have some Chinese heritage? And I said, as a matter of fact, yeah, my great grandfather is pure Chinese. He's like, why are you asking me this question? It's like, okay, that's the rule. If you have 12.5% or more Chinese heritage with you, you can get this money and basically start your own lab. So he gave me $75,000 to start a mass spec lab. You know, $75,000 is nothing, but he is very smart. He knew who to connect me with and basically start collaborating. And from that $75,000 in 2012, we were able to raise more than $15 million in 12 years of research money. And the key is really just knowing what niche you can actually go to and contacting the right people that can give you money. It's not always fun because in science, typically the thing that you really want to do, there's no money in there. So you kind of have to have this backup projects where you get the money and then support the things that you wanted to do. So now that is the reason why I have my own lab. I did not really want to have my own lab, as I said, but here I am writing grants, so I run and direct the UCSF Clinical Toxicology and Environmental biomonitoring Lab. It's a mass spectrometry lab. It's very probably very similar to what you do. We develop methods for PIs. We develop methods for researchers that they can use in their own research. From a research standpoint, my area is really new psychoactive substances analysis and surveillance. So those are designer drugs, like analogs of fentanyl, for example. But my source of income is mostly collaborations on other areas like endocrine disrupting chemicals. Because there's a lot of money. Well, last decade there was money in endocrine reception chemicals from the National Institute of Environmental Health Sciences. There's still some money, not as much as what we used to have. And also other researchers who are trying to monitor drugs in several of their studies. So the lab is now doing three large areas. NPS analysis in edc, biomonitoring, and drug adherence monitoring. All of this, of course, is based on mass spectrometry. We do work with other PIs. We do PK studies as well, depending on what the PI would ask us to do. So we are mass spec lab. The two workhorses in the lab are obviously the liquid chromatography tandem mass spectrometry. And I think in this audience, I don't really have to explain much. We exploit this for. For really finding transitions that are unique to our small molecule target so that we can use those transitions to identify and quantify them. So in LC Ms. Ms. You have the quadrupole that selects the parent ions, the collision cell which fragments the parent ion, and then another quadruple that selects the fragment ion. So here's an example of, say, a molecule that we usually work with. Methamphetamine is, and we look at the fragments that methamphetamine would form based on the defined collision energy. In our method, the ratio of the peak areas of these different fragments is a constant in a given collision energy. We use that to identify the compound as well as the retention time that was defined by the chromatography. Of course, for compound identification or quantification, rather, we use isotope dilution method. The other workhorse in the lab is high resolution mass spec. And so I do have an LCQ TOTH Ms. And here really the only difference is that instead of a second quadruple, you have the TOF analyzer, the time of flight analyzer, coupled to a quadruple and a collision cell. So the difference is that we can collect exact or accurate mass information about ions from compounds. Either the parent ions or if you want some structural data, we will fragment them and collect fragment ion data. What HRMS allow us to do really is to have a number of different types of analysis that we can do in the lab. So in hrms you have data acquisition separate obviously from data analysis. You can either do a targeted or a non targeted data acquisition. If you do targeted data acquisition, of course, the only type of analysis that you will do is also targeted. So you define a priori what you're looking for. You can only analyze data of those fragment ions that you were looking for that you collected in your acquisition. The more exciting ability of Q2FMS or HRMS is its ability to actually get non targeted data. So here you're basically collecting all potential ions from a given sample in a non targeted manner, using their accurate mass. And the only reason why we're able to do that is because the advances in the engineering and optics of most QTOF Ms. Has allowed mass accuracies of about 2 parts per million. Or it's fair to say that for small molecules you can potentially accurately determine the mass to the 0.001 atomic mass unit, where when you get that level of accuracy, you're able to do an ambiguous formula assignment to your masses. When you do non targeted data acquisition, you can do either targeted analysis or non targeted analysis. When you do targeted data analysis, you can do them in two ways. If you have the reference standards of your target, that's no different from a targeted analysis. But if you only have a database, but you don't have the reference standard, you still can do targeted analysis using the accurate mask. That's what you call suspect screening. Or you can also do purely non targeted analysis. So the ability to do suspect screening and non targeted analysis is really one powerful tool that have enabled us to actually work in the NPS space. Because one key requirement in doing analysis in PS is to be able to identify previously unreported substances. And you cannot do that. Of course, with targeted analysis. This is just a repetition of the previous slide. I was telling you a while ago that for targeted analysis we have the reference standard of the targets or the analytes. Because we have the reference standard, we have information, the retention time, the high resolution mass spec, and the Ms. Ms. Data. For suspect screening, you can have a database available, but not the reference standards. And this is actually very important for us because the universe of new psychoactive substances now is in the tens of thousands. It will be impractical to buy the reference standards for all those. So what we do is we create databases of what we know is potentially available and what is predicted to come out in the future. We use those databases to actually triage what potential reference standards to buy. And here you would have data on HRMs and MSMs. Obviously for non targeted analysis, you have no prior information available and all you have to rely on Is HRMs and MSMs. So for this afternoon, what I will be doing is really focus on three areas where we use mass spectrometry to come up with what our clients actually are asking us to do or to deliver to them. So one big project that I'm currently doing is the Drug Enforcement Administration Toxicology Testing program. And this Test DEA Talks program is really a surveillance program on new psychoactive substances. So it began in 2019 as a program aimed at detecting new psychoactive substances. Only in the United States though. So the goal is really to gather intelligence for the DA on current and emerging nps, while also educating local physicians and law enforcement of the specific drugs in their area of responsibility. So really this is more for intelligence gathering for the dea, but also to help local physicians on what to expect at the emergency department. Here's the simple workflow. So the unique feature of this program is that the DEA pays for the testing, right? So if they say you are a medical examiner office or a clinical lab in a hospital and you have a unique case that you think might involve nps, you can refer that to the DEA by emailing the DEA and saying, we have this weird case. This provide a two to three sentence description of the case and the DA approves that. Once the DA approves that particular case, then what the DA does is to inform us that, hey, there's a case coming from Pennsylvania. This is the description of the case. The lab or the referring institution is also asked to submit the sample. The sample that we analyze are usually blood. So whole blood, plasma, serum. We also accept urine. We also accept tissues from cadavers. We also accept drug products and paraphernalia. Now, once we have the sample, we would start working on it. Sometimes we would communicate with the referring institution because some of the data that they have regarding the case can actually be helpful in our sleuthing of what is in the case or in the sample. We are given three weeks to report back the data to the dea, and once the report is approved by the dea, it's sent back to the referring institution. Now, aside from this, we also issue quarterly reports and annual reports to all stakeholders so that they know what's going on in the United States, what's going on in the different regions of the United States. As far as designer drugs that are either killing people or designer drugs that are sending people to the emergency department. One unique feature of our program is our collaboration with University of Sydney. Well, this kind of stopped now, but at the start we were collaborating with University of Sydney because one of the difficulties with the NPS space is the availability of reference standard and you need that in order to confirm what's in the case. So Dr. Samuel Bannister at University of Sydney, who actually was, I met him here in the Bay area when he was a postdoc at Stanford. He has a unique interest on NPS as well. And one of the things that, you know, back in 2014, 2015, one of the things that we struggle with, how will we beat this illegal companies or research chemical companies in Asia that are synthesizing new drugs and causing this public health issue. And the problem is that the turnover of designer drugs back then is maybe two or three months. Every two or three months there will be half a dozen or a dozen new drugs that are coming out. If you don't have the ability to do non targeted analysis or suspect screening, you won't be able to determine what's actually in most of the cases. So one of the things that we did is, well, if they are searching the literature for new targets, we should be able to predict what they would be synthesizing next. And that's what we did. So we created what's called a prophetic cannabinoid standards library. He is a very talented synthetic organic chemist and medicinal organic chemist. He actually is now the CSO of silo who's probably somebody that can give talk also for this group. They are exploring the therapeutic potential of a lot of psychedelics. And so we have this library that no other lab has. And so what our program has is that if any of the synthetic cannabinoids in our library comes up, we should be able to confirm them right away. And it has happened several times. And I think that's the only reason why the DA wants to collaborate with us. So the method that we use obviously is the LCQ TOFMS that we use for all this work on nps. The acquisition is non targeted, but using data dependent acquisition. And we do targeted suspect screening and non targeted analysis. So we have a very comprehensive library of drugs. So right now we have about 1300 reference standards that includes about 970 new psychoactive substances, some of which are not available commercially because a lot of these are polypharmacy. Our Library also contains traditional illicit drugs, your common drugs of abuse, prescription drugs, dietary supplement, stimulants. We also have precursors, additive and impurities. So here's the workflow. So each sample that is sent to us undergoes three different analysis. First is the screening where we only use stuff Ms. So we collect parent ion data from the sample. And really the aim is to see what's in the sample. Everything that's in the sample. Right. The criteria for identification is mass error should be less than 10 parts per million. Retention time difference should be within 0.15 minutes. And the target score, which is an indication of how the mass spec of the parent ion actually matches the unknown using isotope abundances and isotope matching or isotope spacing matching is about 70. For confirmation, we do an Ms. Ms. Scan and the additional criteria is Ms. Library or mass spectral library matching. So here we actually look at the score of the spectral match. Right. In the literature you can have this as low as 50, but we've been doing this for several years now. The real matches actually are typically about 85 or higher. In terms of mass spectral match. We also impose at least two common fragment ions between the standard and your match in your unknown. And then of course, the analysis is quantitative and it's through isotope dilution. We're only required to quantify new psychoactive substances, traditional illicit drug and precursor additives and impurities. If there's prescription, we actually would just report them qualitatively. So here's what happens in the screening. This is the initial screening that was done. Sorry, I don't think you can read it, but my poor RA would have to go through. So he would compare the TOF data of the sample with the blank and all of the runs that we do. He would examine this line by line. Right. We typically would have about 100. And even though he'd say, well, you know, why is it that you have a targeted database, you still generate a lot of this. Well, actually, you know what happens is that there's a lot of endogenous compounds that are sometimes very similar in mass to drugs of abuse. Right. And so there are also sometimes the signal to noise ratio is not really passing based on our criteria. So you would either have ubiquitous or remember, because we're doing suspect screening as well, we have suspect databases. So sometimes some of them do not have plausible retention time. Right. It could be something that's polar, but the match is coming out at eight minutes. And so it's like, no, this can't be true. So he examines that. Yeah.

So does he go through all the chromatograms or just his data?

Yeah, just his data first.

And then so it's all acquired on the same, like a, same general method that you guys have or do you sort of change it based on the matrix?

So the serum and the plasma has the same method. The whole blood is a slightly different method. The urine has the same method. Well, you know, with non targeted data acquisition you can actually change the way you do data dependent acquisition. Right. So you can do fixed collision energy if you want. You can just basically let the abundances decide what would be. So those changes, those minor changes in the method comes after the initial screen. Like if we see. Sometimes we will have. So this is just screening after this. It's for something positive. Right. And then let's say we have a high scoring match, 98. But we can't find a spectral match. So. So we cannot confirm that. Right. But the match is consistent with. Hey, we think it's a stimulant. This is a stimulant. Is it possible that it's just very difficult to ionize the potential match? So that's when we will change the collision energy and say, okay, let's do different fixed collision energy and see what would happen to. So it's a case by case basis. And when I was starting to do this In San Francisco General in 2010, we naively thought that we can pass this to a laboratory specialist and say, okay, this is the protocol, just run it. We were so wrong. Right? A lot of the, I mean no disrespect to clinical lab scientists, they know what they're doing. But when it comes to high resolution mass spec, they came back to us asking, do you have a diagram of decision tree diagram? It's like, nope, you don't do it that way. So really for this I rely on chemists in the lab who has been doing this for like, I think for now it's about five years. So there are about 100 matches in the targeted analysis. Initially 13 over 100 withstood scrutiny and confirmation procedures. Five drugs were actually identified and reported out of the 102 drugs referred for quantitation. That is the reason why it takes a while to actually do each case. So one of the things that we struggle with a lot of times is a lot of the people that collaborate with us are not chemists. And so they're so used to CSI. Right. It's like in 15 minutes the results are there. I was like, nope. This alone can sometimes take us about two and a Half hours or three hours per case, Just looking at every case. And then we do a confirmation by mass spectral library matching. So here we look at the mass spectral match, right. You see here there's no score. And I was telling you a while ago that the mass spectral library match we tolerate 70, but really the actual matches there are very high. So this is an example of actual match. The mass spectrum from the standard basically mimics what we saw in the biological sample. This is an example of a really good target score. Well, it's a good target score, 35.5. And remember that ketamine has a chlorine. When it has a halogen, it will have that isotope abundance match, which is very convincing for us most of the time. Oh, you have the chlorine signature. It must be true.

Right.

But then if you look at the reference standard for ketamine and what is the MSMS or the unknown? It's not Right. And so that's when there's no score in the mass spectral matching and therefore that's not actually a match. Sometimes this is very interesting to us, right, because okay, it's not ketamine, but it looks like it's ketamine. Maybe it's an isomer of ketamine that they're using now. Right. And depending on how much time we have. Right. We, we can go on. Typically you would have a threshold. If it keeps on reappearing, let's say in a matter of months, we would have five cases like this. That's when we investigate. But now we are getting about between 50 to 100 case a month. And so I only have an RA doing this all by himself. So it really depends. We also get intel data from the.

DEA people that have, you know, effectively passed away or the actual live people.

Yeah. So it's becoming more and more like that. So we're actually transitioning now more to the medical examiners rather than just the poison center. And a lot of the reason why is because the DEA has become interested in our products. And so even though they're the ones collaborating with us, there are different divisions of the dea. So now the DEA censors a lot of cases and a lot of those cases typically goes to litigation. Right. And when it goes to litigation. Right. Almost always those case involves death. Right. And what you want to know is, you know, who is the dealer that sold this innocent 15 year old what he thought was Xanax, but it's actually. Yeah.

So in cases where you have like that sort of no match, what's sort of the process that you actually undergo to sort of determine what it actually is, it varies.

So the first thing that we would do is we would look at the case description. Right. What kind of drug is expected? Is it a psychedelic, is it associative, is it a stimulant? Right.

Are you always provided that background?

We would have about three to four sentence description. Yeah. Of the background that is actually the screener of whether the case is admitted or not. So we would, we would get that. So that's the first thing. Right. Then, then after that we would look at, we would do our own due diligence in terms of running it at various different. Using various strategies. Right. So we would do fixed collision energy changes, for example, or if the parent mass is known, what we would want to do is actually force. Sometimes, you know, even if you do fixed collision energy, you will not get an MSMS data. Right. And sometimes the reason for that is that at the time that that particular mass is coming out, there's just a host of all other ions that the, that the mass spec. But you can do an inclusion list. Right. So so small things like that, that we. And then after that we will look at what are the. So we would look at the fragmentation. Right. So that's when chemist real chemistry starts. Right. So this fragments, right. How will we assemble this molecule to see what, what particular parentine structure? And are there commonalities or are there fragments that are similar to specific classes? Like for example phentalols would always have the piperidinium ion, right. You will always look at that. It's like, okay, if this is this, most likely it's a phandalog, Right. So it, you know, it varies from case to case, but I think that's the most exciting part of our job. Yeah.

Do you ever do like any confirmatory like you.

Yes.

Okay.

Yeah, you need to. Right. Because especially if the D A would need this for a case, it has to be confirmed. Right.

Oh, so in the event you don't have a match, sometimes the DEA does actually want you to identify.

Yeah. Oh yeah, yeah, yeah. There are cases where. There are several cases now where, you know, the DA wants to know because. Sure. You know, it might be a new fentalog that's coming out. Right, Cool.

So do you use APCI as well? Is that useful?

You know, and that is actually, that is a matter of practicality. Right, right. Because I would say that we've been doing this for, you know, more than a decade now. I would say more than 98% of the drugs that you will be dealt with will be ESI ionizable. Right. And adding more like apci, for example. I think at one point we did APCI and it's not really non targeted. It's because we were tipped that this drug is potentially there. And we know for a fact that our ESI method will not be able to ionize it. That's the only time where we actually did use APCI. I would say maybe 99% of the time we use CSI.

Yeah, it's supposed to be better for like non polar.

Yes, of course, of course, for hydrophobic. And there are some synthetic cannabinoids that are like that. Right. You know, highly hydrophobic. And you know, with synthetic cannabinoids, sometimes especially the more potent ones, it's kind of difficult because, you know, the, the level that you, that you expect in the biological sample is already low.

Right.

Plus it's very difficult to ionize.

So when you get a sample to. Do you like get rid of the proteins first before you inject it?

Yep, yep, yep, of course. So, you know, we have protein precipitation, we have solid phase extraction where we take care of the phospholipids. All the.

You're just interested in small molecules, right?

Yeah, unfortunately I know you're a large molecule guy. We're very interested in only a small molecule. We have always been a small molecule. And so for quantification, I don't think I have to belabor this point. We do isotope dilution method. We have obviously the calibrators and internal standards that we use. One thing I would say though is that the quantitation for very large comprehensive databases is a little bit different from your typical LCMS Ms. That you have here. Right. So strictly speaking, if you want to quantify, you want an isotopolog of your target.

Right?

Right. Well, what will you do if you have 1300 targets? Would you, would you be buying the internal standard for each of those 1300 targets? The answer is no, because that's not practical. So we have a set of 15 internal standards that brackets the retention time of the analytes. And we use those, of course, we investigate whether that is reproducible. Right. Because some of the issues that you will have there is like, well, you know, your internal standard is one minute away from your, from your analytic pictures. I think the, the longest retention time difference that we tolerate doing that is probably a minute. The retention times of our internal standard are spread so that it would actually there should be at least one that's very, very close to each of the target drugs that we have.

So how about nanospray? Mostly nanospray.

We don't need nanospray. Yeah, we are not as sophisticated as yours.

No, we don't do it.

We don't. So here, here. So the last time I was here, it was 2022. So I presented data for 2021. So now I'm presenting the data for 2022. Right. Here are the 370 cases. 2022 is really slow for us. The 100 cases a month that I coded is for this year. So this year between 50 to 100. So for last year, this is the distribution. As you can see, there are states that are. We're getting more like Pennsylvania, Kentucky, Tennessee, Alabama and Georgia. And the reason is we have developed relationships with them. So we've already had, you know, a medical examiner that is continuously giving us cases. And so Nebraska is another one. But now those relationships are expanding. So now we have a relationship that's just started with Maryland. Right. So we're getting more Maryland cases. And so obviously the data would be lopsided in terms of where the cases are coming from. So we have a good coverage of the South. Right. And unfortunately we don't have much coverage of California. So the reason why we don't have much of California is that, remember that a lot of these cases also that are referred to us are sometimes referred to us because they cannot do NPS analysis. And fortunately for San Francisco, the office of the medical examiner here has the ability to actually do NPS analysis. So they have a good lab. So those cases, nps, new psychography, designer drug analysis. Yeah, new drugs. This is the number of cases and I'll go to the number of drugs.

Why is Kentucky has so many cases?

Because the poison center in Kentucky is actually sending us all their potential NPS cases. So we have built relationships with some of this we call sentinel sites. Right. Like Tennessee also has a lot. Right. Because the forensic center there actually sends us the out of scope finding from nms. So instead of sending it back to nms, they send it to us. All right. So for sample types, you see, you can almost always correlate whole blood with a medical examiner case. Right? If it's whole blood that's coming from a dead person, right. For overdoses, typically serum, in 2021, we have more cases. 564 that year. You can see the difference here, right? Like, you know, we are getting more whole blood now, right? Because we really have a push for really the death cases. The DA is more interested now in the Death cases. And that's why a lot of the cases that we actually admit to the program are mostly whole blood in terms of case types. Right. So it's almost. If you look at the percentage, it's almost. There's an equivalency. Whole blood, 42%. That's 42%. Right. And the overdoses, we are discouraging this. Now. Kentucky has sent us a lot of urine from overdose cases, but you really can't use urine data to directly associate what you find in urine to the overdose. It's really the serum. So now we're switching again to, you know, even in overdose cases, we encourage serum or plasma. And then again, you see here, you know, the Deaths were only 18% in 2021. Now it's about 42%. You see this push towards the more deadly cases. We were delayed by Covid. This project should have started 2019, but, you know, the first full year was actually 2021. Because of COVID we thought the program will die because, you know, we don't have referrals for 2020. So number of drug detection. So this is the detection frequency in those 360 cases or so. And you can see, you know, NPS, we have 12%, 40% for TIDS or traditional illicit drugs. That's your drugs of abuse, a lot of prescription drugs, dietary supplements, stimulants, and then precursors, additives and impurities. You might ask. Well, you're an NPS surveillance. It's only 12% in your cases. Well, the reason to that is that these are mostly polypharmacy. Right. In almost all of this, you will find a combination of a tid, an nps, and almost always a prescription drug. Right. And so the probability of getting a prescription drug is actually higher. Now, if you look at it differently in terms of the number of drugs detected, right? 32 different NPS, 17 different TIDS, 83. So a lot of different prescription drugs. This 32 is actually much less than in 2021. Well, aside from the fact that this is higher number of cases, what is actually really happening in the NPS space is now we're working with fewer nps. And, you know, it used to be that they have a new NPS almost every other month. That's not the case anymore. So this is something that's, I think, is the evolution of this phenomenon. What they're doing is they're changing the additive, right? You. You've heard about trunk, right? Silazine, right? You've heard about silazine, right? And silazine, they are adding it to fentanyl. Because the report in the field is that it actually lengthens the euphoria that you get from fentanyl. I mean, you do not take fentanyl for respiratory depression. It's really the euphoria that you. And supposedly alpha agonists like psilosine can do that. Well, now, you know, because psilosine has become so popular, they are using. And psilosine is an animal drug, Right. It's a vet drug. Right. Now they're using another drug, medetomidine. Right. Which is another alpha agonist. And this one is actually prescribed in humans or administered in humans. That's what's happening. Yes. Are you seeing any modifications of psilocyne like these veterinary drugs, or are people just kind of cutting it straight? So. Meaning that the silosine itself becomes an analog of silosine. People are purposely manipulating it to try to get around. Not yet. But, you know, I won't be surprised if what they're doing now is they're marching through different alpha agonists. Right. So there's already alpha agonists in the market and they're using. So they're transitioning from thylacine to medetomidine. Right, but your point is. Okay, can we manipulate the structure of silazine so that it skirts the law? But I don't think psilazine is not scheduled yet. And it's a vet drug. It's not a human drug. But of course, I mean, that is actually a debate right now. Right. Do you schedule psilazine? And if you schedule psilazine, it would potentially trigger the proliferation of psilazine analogs. Right. When we first saw psilocene, we actually saw psilosine in 2021. We first saw it in Pennsylvania from our. And it just became popular, I think, towards the early part of this year. Right. I'm hearing news reports about people in San Francisco originating and dying from cells. Yeah. And that was just like early this year. Right. And so people were, like, asking me about it, and it's like, oh, that's old news to me. So drug class distribution. So here I think this is a different way of looking at it. Like, you know, you saw 12% only NPS in detection frequency. Now I am presenting here, how many cases have nps? Right. So that's a different way of looking at it. So about 41% have NPS. Right. About 82% with NPS and TID. So here, look at this. This is very important data. Right. Look at what happens when you only look at deaths, 67% in overdoses, 15%. So it's like that's empirical data or observation that the NPS is actually pushing for death cases. You have higher percentage of NPS in death cases. The overall is about 82%. This is in 2021. I think what happened in 2022 is that in 2021 the NPS cases accounted for about 38%. There's a slight increase in NPS, but the overall for NPS and TID is the same, about 82%. A lot of this is polypharmacy. So you can see here, this is the number of drugs in a case. There is a case with nine different drugs, right? In one case. So about 62% of our cases are polypharmacy. Right. And despite the comprehensive drug panel that we have, we didn't find NPS or TID in 19% of our cases. Most of those are just, they thought it's an NPS or tid, but it's a prescription drug overdose and we will find it mostly in the everdoses in deaths. The highest number that we found is about nine. The average number is three drugs for overdose is seven, about two. Here's an example of a case that was sent to us. 50 year old white male called in as a probable overdose. Suspected meth and pills were found on the scene. They have a history of intravenous drug abuse, hypertension, tobacco use and chronic pain. A controlled substance monitoring database report was positive for clonazepam, buprenorphine, naloxone, alprazolam and oxycodone. The decedent, so he died, was transported to our facility for a full autopsy and toxicology screen. The Tox returned with 8aminoclonazolam and bromazolam as out of scope finding. So remember what I told you about out of scope finding. So this is probably there, but they are not able to confirm it because they did not give NMS more business. So when we analyzed this case, here's what we found. 22 different substances, 18 different drugs, right? Three different NPS. We confirmed the clonazolam and the brumazolam. Maybe there's a difference in the LLOQ and LOD of our method. With their method. They did not report parafluorofentanil, but we found parafluorofentanil. And look at the. So this is the problem now, right? It's always a combination of tids and nps. And look at all the prescription that this person has. And this is the precursor for fentanyl, we quantify all TIDs and NPS and the precursor, additive and impurities. But we qualitatively report the prescription. Except when we were asked to, we were asked to quantify alprazolam. In this case, what are the number of drugs in each NPS class? So for 2022 is a story of depressants. So the top classes are opioids and benzodiazepines. We used to be a cannabinoid lab. In 2015, if there's a cannabinoid case in the United States, most likely they would refer that to us. But it changed. In fact, if you look at just 2021, you can see the change already. In 2021, you still have cannabinoids and opioids in most of our cases. Where are cannabinoids now? Third. But we used to have about 15 different cathinones is also. So, you know, what happened here is that the benzos have overtaken cannabinoids and cathinones. Right. So benzos is also very popular. Designer benzos. Also there's a limited number of classes. I think I referred to this earlier. There's less number of drugs. They're focusing on specific drugs.

Roy, speaking of cannabinoids, are you seeing a lot more abuse and overuse and overdose and does that cause death? I mean, people generally think cannabinoids, you know, they can smoke it, nobody gets hurt.

I'm talking about synthetic cannabinoids here, right? Not your regular Delta 9 derivatives of THC derivatives. These are typically the cannabinoids that were synthesized to be cannabinoid receptor 2 agonists without CB1 affinity. But that rarely happens in the pharmaceutical industry. What happens is you have these candidates that have overly high CB1 affinity. It didn't get authorization, but this chemical companies have unearthed that, oh, this one has a very high CB1 affinity. Let's add it to our herbal incense. What is happening in the cannabinoid space though is they are exploiting. So the main component of the main cannabinoid of marijuana is Delta 9, right. For states that do not legalize recreational use of marijuana, what they have exploited is Delta 8 and Delta 10, THC. And this is weird, right? You know, if Delta 8 and Delta 10 are components of your regular marijuana, right. That's illegal. But if you synthesize Delta 8 and put it on a gummy, that's in a gray area because it's the marijuana that's actually illegal, not the Delta 8, right. So this has Been exploited actually. So we published reports of Delta 8 THC. A lot of this are really, you would hate these cases. These are like three year olds, two years old. The dad has his Delta 8 cannabinoid Gummy, right. He left it somewhere, the poor kid. It's a gummy. It's sweet. Took it, right. One of the cases actually ended up they did not die. So they were. Fortunately, they did not die. They were at the emergency department and they were intubated and you know, they were there for like a few days before they can be revived. I think the dad will die from the wife.

Delta kid was like. I, I think I remember seeing this metam a few years ago that they talked about it as being like a hangover cure, like it was a good way to transition off of Delta 9 because it wasn't.

As anything can be a cure, it's about the dose, right? So the doses that they put in these gummies are high. And the reason also for that is they know for a fact that the activity of the delta 8 is less than the delta 9. So they're compensating for it. So how will you compensate?

Where do they add?

Oh, a lot. You know, don't quote me on this, but I think in that three year old case it's equivalent to smoking 10 joints that this kid took.

Edibles are very dangerous.

Yeah.

Is it more harmful, the Delta 8.

And Delta 10 versus the Delta 9?

No. Well, again it's the dose, right.

So.

But the activity of the delta 8 and the delta 10, if you look at their CB1, CB2 affinity binding is less than the delta 9. So it's a dose, the detection frequency for each class. As I've said, it's a story of depressants, right? Opioids and, and benzos. So in 2021 we have cannabinoids, second to opioids. But now it's like opioids and benzos. Here's a unique feature of our surveillance. A lot of the surveillance just give you data on what they found. We actually give you the reference ranges and in various different matrices. This is very useful because in the medical community when they publish papers and they would say, we found 2.5 nanograms per milliliter of metonetazine. But what does that mean? So what we're trying to do here is to give the medical community and the forensic community what are the ranges where we're finding this. One other thing that we should do here is to differentiate the death versus the overdose. The problem with that with the opioid is that it's not always the case that the higher dose of the opioid is the one that kills. Why? Because of tolerance. Right. If you have a 15 year old that it's his first time, he can die from like very low concentrations of fentanyl. And you will see later, we have people surviving at 100 nanogram per milliliter of fentanyl. The reference dose for fentanyl, 1 to 2 nanogram per. These are overdoses, these are not deaths. So you see parafluorofendanil, which is a fentolog, metronidazine. You're aware of Kratom or Kratom, right? So a lot of people also like using that. Mitragynine is the drug that's in there. These are the more common ones. We do this for all the different classes. So we have all these tables for the different classes. What I will do now is focus on two drugs, right? Fentanyl and parafluorofentanil. Because a lot of our cases are actually about fentanyl and parafluorofentanil. So one is a traditional illicit drug and the other is an NPS. So 47% of our cases have fentanyl. Right. And as you can see again here, you know this is because of our sentinel sites, right? Mostly from Kentucky and Tennessee. So about 50% are deaths, 44% are ODs. In terms of gender, usually have more males, about two thirds to about 75%. In terms of age, it's older. Right. Strida is a monitoring surveillance in Sweden back in 2012-2016. And the average age usually of their cases is typically about 20 to 25. In the United States, most of the cases that we have are typically above 35. That's the average. So here's the comparison of the average concentration we find in the death and the overdose. This is expected. You have higher concentration on the average, but you can see the spread is the same. And here in 2022 you have an overdose that's. This is logarithmic, by the way. So you have something close to about 80 nanogram per milliliter survived. If you compare what we found in 2021, there's something that would strike you right away. Look at the average concentration in 2021. 2.5, 7.5 in 2022. For deaths, it's the same. This is the observation that we have. Overdose cases are becoming higher and higher and higher in concentration. A lot of these people are developing tolerance that they're Surviving. What are the blood fentanyl concentration ranges that we're seeing for the state? And here, what we're trying to see here, is there a difference? Like, you know, would the concentrations of deaths in Nebraska higher than in California? And so what we're seeing here, of course the death cases would have relatively higher. But there's some interesting observations here. Obviously, Nebraska here is higher than Georgia and Tennessee, but look at Pennsylvania. This level is supposedly death, right? These are overdoses. This is one mass intoxication event, though, where the heroin was contaminated with fentanyl and the levels are high. Most people survive because, you know, I think Philadelphia is the mecca of opioid crisis in the Midwest, northeast area. Drug co occurrences. You've heard this all the time, right? There is fentanyl in my meth or there's fentanyl in my cocaine. And that's true, right? The highest CO occurrence actually that we find in fentanyl cases, stimulants, methamphetamine and cocaine, right. There are also co occurrences of fentalog and other opioids and benzodiazepine and fentanyl. And here's the difference between how much are deaths and how much are overdoses. Now, the next slide is a little bit complicated. What I want you to focus on are the red and the blue first on the pie chart. The red is the proportion of deaths, the blue is the proportion of overdoses. So if you have an opioid like parafluorofentanil, metonadocine, morphine, bromazolam and clonazolam, the general observation is that adding another depressant, more deaths, right? Because you can compare that with stimulants, right? Methamphetamine and cocaine. And I think what is happening here is that the two opposite forces trying to balance each other out, the stimulant and the depressant, so that not a lot of these cases are actually dying, right? The other. So the next data I want you to look at are the fentanyl concentration. The average fentanyl concentration for the CO occurrence, typically around the 17. There's one outlier, though, morphine. See this? 17, 19, 17, 16, 17.5, 21.5, 20.1. But the morphine is 8.4. This is average concentration. If you want geometric mean still morphine is an outlier. 3.5 versus all of this other ones. We do think that maybe there is more synergy between heroin. We're thinking that this is mostly heroin and fentanyl than the other depressants. Here's another data. And this is where you know the balancing effect. Right. If you have fentanyl with meth. Right. The average is higher for fentanyl. Right. If you have fentanyl with no meth, you know, you would die or the average. These are death cases even at lower fentanyl concentration. That is not true, though, for parafluorofentanil, which is a fentolog, there's no difference between parafluorofentanyl and fentanyl. Switching gear now to parafluorofentanil. Right. The only difference here is one fluorine atom. 20% of our cases are parafluorofentanil. In 72% of those parafluorofentanyl cases, we also saw fentanyl. Only two of the 74 cases have nitosine. So nitosine is a new synthetic opioid that they're trying to replace fentalog in fentanyl combination. Right. So when you find a fentalog, almost exclusively, they're mutually exclusive. Right. If there's a nitazine and a fentanyl, you won't find a parafluorofentanil. But if parafluorofentanil is already there, you won't find a nitazine. Nitazine are also powerful drugs. So 82% are dead for parafluorofentanil compared to 50% for fentanyl. The average age is older. The reason why there's no overdose here is we didn't have cases with blood for overdoses in parafluorofentanil, mostly urine. But it seems like that, you know, there's really no increase in the concentration of parafluorofentanil. So again, so different states. Again, NE, parafluorofentanil, the average concentration is about 21.4 compared to this other one between deaths and overdose in drug occurrences. Okay, so pie charts again. What do you see this time? Almost all red. Right. Almost all deaths for like combination with depressants and even this stimulants. Right. It has become about 75% compared to about 50% the parafluorofentanil concentration. Quick question. These are just the cases that you received? Yeah, of course, of course, of course, of course, of course. As I've said, it's lopsided with wherever we get these cases from. And so with fentanyl or no fentanyl. Well, this makes sense. Right? So with no fentanyl, you would probably require a higher powerful fentanyl concentration. With fentanyl, you have two Opioids working. These are death cases. Okay, so these are fentanyl cases on the left and parafluorofentanyl cases. Just look at the pie charts. When you have parafluorofentanil, right. You're pushing it to more red. Right. In almost all drug currencies. But it's also difficult to say that it's just parafluorofentanil doing that because in almost all cases that there's parafluorofentanil, you will also find fentanyl. So it's just that there's so much co occurrence of these drugs with both fentanyl and parafluorofentil. But in a lot of these cases, you know, you're pushing it towards more death.

So the parafluoro is stronger.

No, actually, well, a little bit stronger than fentanyl. But I think it's the additive or synergistic effect between fentanyl and parafluorofentanil that causes the deaths. And this is a shameless plug. If you want to learn more about designer drugs, I have a book coming out. October is the release date. So I'll move to biomonitoring because I think this is a. I'd like to discuss this because this is a powerful demonstration that we at the mass spec world can actually make a difference in huge epidemiological studies. Right. We can have our $0.01 and maybe $1 because there's power in the technique. Right. And not because it's developed by some big agency that they're necessarily correct. You can challenge it, but it will make your life hell. So biomonitoring is really the measurement of the body burden of toxic environmental chemicals measured in biological samples, typically urine and blood. An example of a biomonitoring program is the United States National Health and Nutrition examination survey. So NHANES collect survey data from 7000 to about 9000Americans. And every other year. Right. Of those 3000, around 3000 biological samples are actually used to measure environmental chemicals. So every other year this happens. So biomonitoring is an integral part in the assessment of the risk for chemicals. And risk assessment is a complex process. It has at least three components to it. 4. First, you need to know the hazard that that chemical causes. So what are the effects that the chemical have in our bodies? Right. These are typically obtained from animal studies or in some cases epidemiological studies. If the chemical is hazardous, you need to know the dose, where dose effects are being caused, especially if there are adverse effects. And the third important Component is exposure. Right. You need to be able to measure the exposure level of the population. And if the hazards of the chemical and the dose by which adverse effects happen are known, then you can decide, based on the human exposure level, whether you need to clamp down and maybe pass some regulation in trying to limit the exposure of humans that specific chemical. Right. So the biomonitoring is actually your only way to work on exposure assessment. And so biomonitoring can provide data for hazard identification, dose response, and risk characterization. So in risk assessment, biomonitoring provides all data points, all the different factors that are actually considered in assessing the risk of chemical. So one of the most popular chemical that is being biomonitored are endocrine disrupting chemicals. They are molecular mimics of naturally occurring hormones. Just like the example here where you can see the molecular structure similarity between isononal phenol and environmental phenol and estradiol. Right? Now, because EDCs affect hormone synthesis in it affects secretion transport, it affects receptor binding, it affects the action and even the elimination of hormones, because hormones are critical in fetal development and in almost all the physiology of organ systems. Endocrine disruptors can disrupt sexual development. It can have an effect on cognitive brain development. Sometimes it can cause anatomic deformation. Cryptorchidism, for example, undescended testicle or hypospadias, the bad development of the penis. Those are potentially caused by EDCs or malformations of. And you know, it can have a role in a host of different pathophysiologies like diabetes, obesity, cancer and metabolic syndrome.

The isonomyl phenol, is that. Where is that. Why is that in the environment?

Oh, it's used in industry. Right. As something or. No, it's not an herbicide. I think sometimes it's being used as an emulsifier.

Oh, I see, Right.

Because you have, you have this and then you have a. You have a hydrophobic and then a polar. There's a lot of different uses for EDCs. There's actually a lot of different types of EDCs. And fortunately, these are chemicals that make our modern life more convenient. So bisphenols, for example, these are components of polycarbonates and epoxy resins that line tin cans. So they can be in reusable plastic bottles. It could be in the epoxy resin that lines tiny cans to prevent rusting. It can be in thermal receipts, even in dental sealants. Right. They use bisphenol A in dental sealants. Dental sealants actually is very interesting because you either have a choice of mercury or bisphenol A.2 poisons, phenols. Other phenols like parabens and triclosan, they have bactericidal or fungicidal effects. So they are incorporated in personal care products in capsules. You can have phenols also. Also in non alcohol based, non alcohol based hand sanitizers. Colgate Total used to have triclosan. I think they already scrapped it, so I should take this out. Phthalates are plasticizers. So they are in toys, they are in medical products. They can also be in personal care products and household cleaning products for fluorinated alkyl substances. Our ingredients of Teflon. So they are used to coat nonstick pans. They are in stain resistant and water resistant fabrics. And so I don't think there's a human being that is free of any of this. Right. It's very difficult to find one. And this list is not exhaustive. You know, there are so many other endocrine disrupting chemicals. Polyaromatic hydrocarbons, flame retardants, polychlorinated biphenyls that dioxins, pesticides, fungicides and even pharmaceutical agents. Non steroidal anti inflammatory drug is actually endocrine disrupting. When I started working on this field, the question that I had because every single one has an effect, why are we still alive? If you think about it, the effects would add bisphenol A is the poster child for endocrine disrupting chemical. And so I chose this because I want to show how mass spec methods can actually influence the risk assessment of an environmental chemical. So according to the FDA and the plastic industry, BPA is not dangerous at the levels people are exposed to. Right. However, in 2008 the US National Toxicology Program had published some concern for adverse effects on the brain, behavior and prostate gland in fetuses, infants and children. So how do we determine what are the levels that are safe? So for every environmental chemical there is a level called the low adverse effect level or the lowell. So this is usually established in animal studies. And to translate that to the reference dose of safety. Right. The dose that is safe for us, you divide the Lowell with 1000. So a thousandth of the lowel will be your reference dose. So 50 milligrams per kilogram per day is the Lowell for BPA. So again, this is logarithmic. 50 microgram per kilogram per day is the reference dose for humans. So above that, that's not safe anymore. So according to estimates, the American population exposed to 1 to 15 microgram per kilogram per day. So it's safe. There are multiple studies, though, that suggest that BPA can alter rodent reproductive physiology and behavior at doses lower than the Lowell. So what do you do with that data? Right. Should you lower the Lowell? And how is this determined? This is determined by nhanes. So the next question is, how accurate is this data? So this can only be accurate if the method used to measure BPA in the human population is actually accurate. So that's what we would want to examine here. What is the accuracy of the CDC method for measuring bpa? So what I am going to be discussing in the next 10 minutes is from a paper that we published in Lancet in 2019 where we have gathered enough courage to question the method used by the CDC in NHANES. Okay. In order to emphasize the importance of measuring BPN's metabolites in biomonitoring, I'd like to point out first that BPA has a wide range of adverse effects. And almost all of these studies are done using rodents or animal studies. Some are actually done or backed up by epidemiological data, but a lot of these are rodent studies. So in rodents, the most critical part of BPA exposure is, is really in the fetal development stage. That's when BPA's effect is critical. And when the developing fetus is exposed to bpa, the gonads and the reproductive tract of both male and female fetus are sensitive to bpa. In addition, the developing brain is actually susceptible to the effects of BPA exposure. And the mammary glands and the prostate of female and male fetus are exquisitely sensitive to the effect of BPA during development. That subtle changes in these organs during development actually implies higher propensity towards developing cancer in later life. Now, you know, the changes that BPA induces in the fetus in the brain actually affects behavior when that particular individual is born throughout his life. And that is, you know, some of the BPA scientists would say that's why, you know, you have a lot of autism now, you have a lot of this ADD, adhd. It actually affects all other organs, like heart, the lungs, and other organs in the body. Well, the question here, though, is that all this data, this is a lot of adverse effects, right? The question here is, well, you gather that from rodent studies. How relevant is that to humans? Because you're doing your risk assessment, basing it from animal studies. So one thing that you. And this is quite relevant is rodent really a good model Right. There is a reason why people question whether rodent is a good model. The metabolism of BPA in humans and monkeys is different from the way it's metabolized in rats and mice. So BPA is actually rapidly metabolized to BPA glucoronide in humans, that goes to urine, mostly urine. BPA glucoronide in rats, a lot of it is actually excreted in the feces in the urine, plus you have enterohepatic recirculation in the mice. So our pharmacokinetic studies in mice, which becomes the basis for risk assessment, are they legitimate to apply to humans? So what we did to address that is we did some monkey studies and compare that with mouse study. So with monkey studies, you kind of have to compare oral and non oral exposure, contamination. That's a huge part of studies on bpa, because everything has bpa. So how sure are you that you're not just measuring contamination? Pregnancy has an effect on how BPA is distributed. And there is a passage to the fetal compartment, which is critical because that's what actually affects the fetus. Right? So BPA is metabolized to BPA glucuronide. It can also be metabolized to BPA sulfate. The current understanding is that BPA is endocrine disrupting, but these two are not or less endocrine disrupting. Right. So what did we find out? So for this study, because we're so afraid of contamination, we actually treated the animals with deuterated bpa. So you cannot contaminate it. Right. Because what's available? So there is a pharmacokinetics of a 400 microgram per kg of BPA in a monkey. In rats, you practically have the same kinetics, Right? So this is probably interiorhepatic recirculation, but it follows the same pharmacokinetics. Right. So this is one way to say that, okay, the rodent studies are legit. Right. You know, BPA has the same pharmacokinetic behavior in both mice and monkeys. Now, what happens in pregnant monkeys? So this is non pregnant. This is GD50, gestational day 50, and this is gestational day 95. Well, you can see there's a lowering of the free bpa. Where does all the BPA go? It goes to the fetal compartment. Right. So we did this study at San Francisco General where we collected samples from people who are undergoing elective pregnancy termination. We collected their maternal urine, maternal serum, cord serum. We collected fetal liver, fetal ovary, placenta, Amniotic fluid. What I'm showing here is maternal serum and maternal urine data. And there seem to be an association when if it's high in the serum, it's also high in the urine. This is an exception to the rule, and this can be easily explained. These two individuals probably touch a thermal receipt. That's why it's high in the blood. It did not go to the liver to be converted to BPA glucuronide. That's why you don't have BPA glucuronide in the urine. But what is concerning in these levels? Look at the x axis. According to nhanes, this should be the levels you're seeing. Right. According to our surveillance, these are the levels we're seeing in pregnant women. Well, given that our women are mostly from the lower socioeconomic status, which is usually, they usually have the higher exposure, but still, this is hundreds. This is like 0.53. Right. So why the difference? To answer that, one has to look at the method. Right. So there are two different methods for measuring bpa, indirect and direct. Right. The methods that were developed in the 90s were indirect. Why? Back then, there is no reference standard for bpa glucoronide and BPA sulfate. So the way you measure BPA is indirect. You have a split sample of the urine. One is measured for pre bpa, the other one you treat it with H pomacea glucuronidase. So you deconjugate the conjugated form of bpa, measure the total BPA, and then determine conjugated BPA as an indirect variable, the total minus the free BPA. While in 2010, the National Toxicology Program was commissioned to have BPA glucoronide and BPA sulfate synthesized. So 2010, we now have the reference standard. So there are four labs that develop a method that directly measure BPA, BPA glucoronide and BPA sulfate. Right. So why is this quite problematic? Well, in the indirect method, right, One question that you would ask is like, how sure are you that all the BPA glucuronide is converted to BPA if you treat it with H Promasia glucuronidase? Well, they use a surrogate. They add four methyl and biluurone glucoronide. And at the end of the analysis, if there's NO4 methyl and biliphyll glucoronide, they say the deconjugation is complete. We will be taking all the total bpa. But what did we learn from enzymology Enzyme substrate specificity. What may be true for four methyl and boliful and glucoronide may not necessarily be true for BPA glucoronide. Right. The other way of monitoring it now that we have BPA glucoronide is to monitor exactly how much BPA glucoronide is disappearing. Right. You can still, you know, do an indirect method, right? Treat it with H funacia glucuronidase and say, okay, if I use this method, am I really getting rid of all the BPA glucorid? Right. To basically validate your method, here's what we found out. The FDA uses a very low concentration of H pomacea glucuronidase at 2 hours incubation time. Their percent conversion is only 8.3. And remember, it is the FDA that sided with the plastic industry to say that the exposure level is below the reference dose. Right. The CDC, the conversion is 100. That's good. That's what's being used for the NHANES. Well, we tried going overboard. Let's increase the concentration of H formation glucuronidase. Because we are shortening the incubation time, we achieve the same thing, right? So this is well and good. You know, the indirect methods seem to be working. There's another way of monitoring this. If this is true, you look at the equivalent amount of BPA that should have formed from the BPA glucuronide right after the deconjugation, and this is what we found out. At low concentration of BPA, only about 70% is actually being recovered. The problem is the higher the concentration, the lower is the recovery of bpa. And remember, it's the BPA that you're measuring in an indirect method, not the BPA glucoronide. To make the long story short, here's what we found out. We looked at 29 different women. We arranged the levels of total BPA using the direct method, and these are what we found out. If we analyze the same sample using the indirect method, this is what you'll find. Suddenly, the high levels are now being measured at about the same level that the CDC is reporting. Right. So what does this do to the data? This is the actual data that we found. This is the data that you will generate using the indirect method. And this is where the risk assessment and exposure assessment is based upon. Now, if I change this, because when we publish this data, all the epidemiologists remember this has been measured since the late 1990s by the CDC. There has been tens and dozens of papers published on the effects of BPA on many different organ systems. And they are finding out effects. If you change this, you'll still see a difference. The effect is, is lower though. So all the epidemiological studies are still legit. What we're saying is that if you probably use a direct method, you would have seen a higher effect or the reference dose or the exposure level now is exceeding and that's what we're seeing. So we revisited this. This is the Lowell, the reference dose, this is the estimated exceed exposure from the CDC. This is only 29 though. But if you look at the actual exposure of these women, they're way over the reference dose. So this is where the direct method could potentially give you the more accurate data. This is baffling to us. The CDC has dozens of postdocs that can develop the direct method and just compare.

So regarding the pfas present, you know, what regulations are there? I mean, it sounds like there's anybody can do anything, any amount in anything. Is there any kind of a.

So they have regulated some of the PFAs like PFOA and PFOS. And that is actually the reason why, you know, there are new poly halogenated substances. But you see, there is some theme in the work that I'm doing. Like in nps you have this synthesis of new drugs, right, to replace the old one with environmental chemicals. It's the same. You know, you'd release a regular. So for example, one example is mbp, dibutyl phthalate and diethylhexyl phthalate. George Bush actually passed a law limiting the use of that, especially in children's toys. That was around 2008. And you would see in the NHANES data the level of these chemicals actually went down in the population, which is a win. Right. But what did the industry do? They replaced it with DiNP, right? Diisononyl phthalate. And you know, the industry has actually a very tricky way of defining what diisononial is. For us chemists, we know that isononal is a chemical entity with a one methyl chain branch at the penultimate carbon. That's ISO. You talk to any chemist. That's always the definition to industry. It's not if a 9 carbon chain has a branch point, that's ISO to them. So what does this. This is a huge implication by monitoring. Because now, you know, there are two CAS numbers for ISO, diisunonyl phthalate, right. For biomonitoring, we look at what is the chemical to biomonitor. Well, if you vary the branching in that chemical, the metabolite that you will analyze will also vary. And depending on the metabolite you'll target, you know, you may not be able to see the chemical in the population. So for example, there's a difference in the kinds of diacal phthalates that are being used in the United States and in Europe. Right. And you know, not only that, when they say diisononyl phthalate, it's not one chemical entity, it's a mixture of branch isonials. Each one of those can produce many different metabolites. So the question is what is the one that you would biomonitor? So I mean the same is true for PFAs. It's the same like, you know, they're, they're shortening the chain of pfas. All these tricks that are being used by the new psychoactive substance space is the same trick that industry is using. Why not? Like, you know, we know that the property of a chemical is associated with, with its structure. So for example, you know, a very simple structure is example is BPA free products. Right. When you think that, you know, when you're actually buying BPA free, it's well, it's BPA free, but it is not bisphenol S free, it is not bisphenol F free, it's not BPAF free. And some of these replacements are actually more endocrine disrupting than bpa.

So they're playing with the word.

Yeah, semantics. Well, we know as chemists that it's the structure that dictates the property. Right? And so if you have optimized the production of that specific chemical, it's cheap to produce that chemical. If you're suddenly being forced to produce a different chemical, you would only mine you would want minor modification in the production line, right? Because then you know, your bottom line stays high. So that's what's happening. Yeah, it's sad. But there are some green chemists though. So there are chemists that are trying to develop chemicals that would have higher degree of biodegradability sa.