Update on the Neurobiology of ADHD and implications for clinical practice by Prof Mark Bellgrove

good morning everybody and welcome to the ADHD special interest group this morning I'd like to welcome Mark belgrove who is a professor of cognitive neuroscience at the Turner Institute for brain and mental health in the school of psychological Sciences in Monash University in Australia now Mark did his postdoctoral training with us in Trinity College Dublin with Ian Robertson and Michael Gill and he maintained strong collaborations with Irish County so many of you may have met buckbell growth previously the work of his laboratory focuses on understanding the neurobiology of attention and attention deficit using a variety

of methods from cognitive Neuroscience with EEG and MRI genetics pharmacology animal and cellular models he is founder and president of the Australian ADHD profession Association which is an interdisciplinary professional group with over 500 members and currently Mark is Secretary General of the world ADHD Federation so I'd like now to hand you over to Professor mark belgrove okay look at some my pleasure to be speaking to you uh this morning uh your time uh on ADHD which uh as Ashland uh has mentioned I really cut my teeth on this topic uh in Ireland when I was

a poster back in in 2002 to the uh into 2004 and have then sensed um really managed to maintain a really productive collaborations and in a sense this is a bit of a an homage to that period uh in Dublin because when I was in Dublin I traveled around the countryside actually with a neuropsychological testing kit and at the time tried to link uh variation in candidate genes to neuropsychological profiles and so really what I'm going to tell you about is where that sort of work has progressed uh since the the Advent of genome white

Association and polygenic risk scores and the like and try and Link that into an overall picture of the neurobiology and and why you should care as clinicians these are my disclosures the um recently I have been leading the development of a national clinical guideline for ADHD which follows on from the nice 2018 guideline which I'll give a Shameless plug for at the end so my lab as Ashling says that Monash is buried interdisciplinary in center of screen there is a colleague who I met in Dublin in fact I often say he taught me genetics that's

Dr ziara Howie who work worked in Michael Gill's laboratory for a long time and did some of the first Gene mapping for ADHD for the rd4 drd5 and the dopamine transporter and ever since my time in Dublin he's been working with me first in Brisbane and now at Monash and he lives around the corner so we're literally a family at this point uh we also have parallel Labs like literally next door where we do traditional cognitive Neuroscience uh work and then usually we try and connect these things so that we're trying to understand how genetic

variation plays out to affect cognition and how that might create a susceptibility pathway for example for disorders such as ADHD so today I'm going to tell you about uh ADHD I'll give you a little bit of a Cook's tour a bit of a whirlwind tour of the basics from epidemiology environmental factors where does the current state of the art sit on these factors particularly in reference to the new International consensus statement which Steve Ferrone has LED and I'll provide a reference to that for you if you haven't read that I thoroughly recommend it to you

and then I'll move to through some neurobiological theories uh into an example of how we can in the modern era link genes to cognition to susceptibility and try and bring it back with some clinical implications for you more than happy to take questions at the end also can you hear me okay uh Helen just before I keep barreling along um okay so these are the Articles I'd recommend you dig in if you want a really overall or bird's eye view of ADHD the first one is a nature primer article a little bit old now the

second one is the consensus statement and you can find just a real wealth of evidence-based knowledge in those articles uh very accessible and I think really good for clinicians to be able to provide evidence to info information to patients particularly about you know what is uh what is real and what is uh perhaps not so really good for myth busting we have lots of web resources available at our ad per website as well so ADHD I'd like to start with this a description of ADHD because it really for me gives a really accurate picture of

some of the problems people with ADHD speak of and it's also good for myth myth busting because Alexander Crichton wrote this condition way back in 1798 so it can be really useful for counteracting the perception that ADHD is just a modern disorder I'm going to read through it that he spoke of a nervous problem which may be born with the person will be the effect of accidental disease we know that is true often ADHD is born with a person but if you have a traumatic brain injury for example you can have a syndrome or end

up with a syndrome that's very similar to ADHD when born with the person becomes evident at a very early period of life and has a very bad effect in as much as it renders him incapable of attending with constancy to any one object of attention and that's a really brilliant description of a problem sustained attention how it is we managed to maintain our focus our attend with constancy to an object of attention but it's seldom so greater degree as to totally impede all instruction and what is very fortunate It generally diminishes with age so we

now know uh that ADHD does remit in some cases but more often than not it will persist and the most recent data I think suggests Persistence of around 60 percent uh so as those of you who are treating kids many of those will go on to have ADHD throughout their life as well every impression seems to agitate the person and gives him an unnatural degree of mental restlessness a slight noise too much light too little light all destroy constant attention in so much as it is easily excited by every impression and that's a really good

description of what we in cognitive Neuroscience think of selective attention or distractibility how it is one's attention can be caught by extraneous stimuli in the environment and that you need to be able to exert top down control to focus your attention back on what is needed so just a really nice description by Alexander Crichton you'll see it's called mental restlessness we know that man culture has changed a lot over time for ADHD but early descriptions even way back into towards the end of the 1700s so in terms of the epidemiology of ADHD in here I'm

really referring to data that always comes from large large either population based or clinic-based studies but where we can really get a good perspective of prevalence at an international level so we know the global prevalence in Children and Youth about 5.3 percent in Australia actually it's a little bit higher some Studies have reported as high as 10 to 20 percent but really if you look at a Global Perspective it's around five and a five and a half percent and there's really little uh evidence for a real increase in the prevalence of ADHD over the last

30 years what you may see is that the diagnosis has gone up in some areas due to Greater awareness and greater reporting but if you look at the structure of attentional hyperactivity impulse impulsivity symptoms over time what you can see is those distributions largely have not are changed so although we might have a few more people coming to attention really the distributions of those symptoms are fairly consistent over the last 30 years and in many ways I think you can view ADHD symptoms as a a quantitative trait uh that varies or a continuous trade that

varies in the normal population uh really uh par excellence and I think some of the data I'll deal with today will speak to that we know the Persistence of functional impairment is high and it's high into adulthood there's some data from Steve Ferrone the female to male ratio continues to be much discussed we know there's a bias favoring more males are diagnosed in childhood in the clinic or populace settings population settings in young people but then by adults that largely diminishes and so there's lots of talk at the moment particularly in Australia about this phenomenon

and what is occurring is it boys are remitting is it that girls are going undiagnosed in childhood and coming to attention later in life and the latter would be probably my perspective on what's going on there but clearly definitely a big Hot Topic in research going forward to be able to get a better handle on these gender differences we know sort of classic environmental factors do influence the development of ADHD and that's in the context as I'll talk to you about a very high heritability rates for ADHD on the left is some uh you know

old data I think this is actually another plot of Steve frones but you see you can see cigarette exposure alcohol exposure drug exposure in utero a low birth weight interestingly low birth weight there's a recent study just come out suggesting that perhaps low birth weight may not be such a a negative predictor of Life lifetime ADHD in a recent study uh psychosocial adversity is a big one we know trauma all these factors are really important risk factors for ADHD and over on the right there really I've just put an update to this from the new

consensus statement you can see that maternal smoking is coming through a little less so if you control from maternal ADHD suggesting that it's perhaps not a true environmental effect but rather women who have ADHD symptoms are likely smoking to control some of their own symptom problems and then you can see a whole raft of things even down through to maternal use of paracetamol as sort of one of these laundry lists that makes you you know really worry about the range of things that can be on there as risk factors but you know these are are

all big population studies often the effect sizes are relatively small but the consensus statement does certainly uh lay all these out what we've had for a long time is a neurobiological uh theory of ADHD that has focused predominantly on the catecholamine hypothesis of ADHD I'm sure you're all familiar with this the theory goes that since the uh stimulants like a stimulants such as methylphenidate or dexamphetamine are effective in reducing the symptoms of ADHD then we assume that that is happening by at increasing the availability of dopamine and noradrenaline in the synaptic clef and this is

really powered forward a whole neurobiological theory of ADHD regarding the catecholamine hypothesis we had for a long time hopes that we might be able to find objective markers of a catecholamine deficit in ADHD using techniques such as dopamine transporter Imaging or spect and this is a classic study by the General German group Drexel for example and Alex were on the left here they show a patient who has a radio ligand for the dopamine transporter sorry guys uh dopamine transporter uh injected and you can see the fluorescence in panel a is very bright suggesting that that

ligand has been uptaken very avidly in panel B is the patient when methylphenidate is on board suggesting that has reuptake has been inhibited by blocking of the dopamine transporter and then in panel C is your stereotypical control subject who looks as you can see more akin to the medicated ADH down or and for a long time this was really uh I guess held to be the promise the Holy Grail to identify a biomarker but in time this study this finding really hasn't held up we've found that a whole range of factors can influence the transporter

binding whether the patient smokes whether they've had a past history of exposure so really at this point we're left with a dopamine catecholamine theory of ADHD that is really predicated on the clinical effectiveness of the drug but for which we lack objective neurobiological evidence there is no evidence for example that dopamine neurons are aberrant in individuals with ADHD and this is largely Akin I would say to the situation with serotonin transport as it relates to depression for example although the transporter itself the jury's very very still very much still out on its role in depression

we know that antidepressants have clinical Effectiveness in treating other condition and so also we've had a whole wave of neuropsychological theories uh and these are really driven by the Striking similarity of the neuropsychological profile of patients who have circumscribed lesions for example after stroke and the symptoms presented by folks who have ADHD and so the predominant theories here have really focused in on the role of the prefrontal cortex and the cognitive functions that that subserves so as I'm sure you're probably all familiar Russ Barkley spoke of executive function and for here mediation was primarily a

disorder of Behavioral inhibition coming from the front prefrontal cortex and other behaviors really fell with below that Joe Sergeant emphasized really the interaction between bottle-up arousal activation and effort systems so for example you can think of the locus cyrillus that uh where an adrenaline is produced midbrain dopamine for example he saw the problems of ADHD as very much a a problem of the coordinated um uh communication between these top-down executive areas and these bottom-up arousal systems and I have to say I think that's a really good uh theory of attention and one I think a

lot of our own work has supported and then you've got other folks like sunooga bark who really emphasize what we might think of as the hot and the cold aspects of executive function so the the you know the cool elements being Barclays classic executive function the hot elements being those related to reward and so we do know that folks with ADHD have a uh alterations in there a reward system whether that be on delay discounting tasks where they choose or have a preference for smaller rewards uh sooner rather than larger rewards later uh and we

know this links to impulsivity uh for example uh in other disorders so that's sort of an emphasis between what might be uh seen as a more uh dorsal uh uh basal ganglia prefrontal system and a more ventral uh system being related to reward and then still uh further we've had other theories from uh castiana satanic in our own work definitely in this domain saying that one of the most ubiquitous findings in ADHD is really that folks with ADHD just have a lot of trouble maintaining a consistent task set so if you give them any sort

of task uh what their their performance is really characterized in the main part by just a fluctuating performance over time and that performance can degrade very quickly and my own feeling is that reflects a failure to be able to exert these top-down control influences but also most likely also impacted by waning arousal systems uh subcortically and so here's a schematic for you which I find tremendously useful and um Amy arnston uh at Yale is sort of I've got a bit of a scientific crush on her oh this is going on YouTube oops anyway apologies Amy

um this is um this is a schematic that really shows the the interaction between these top down and bottom-up arousal systems the red arrows for arousal really should be uh um you know bi-directional because the arousal systems can actually flood the cortex with noradrenaline and really help to bring attention uh back to uh back to control so if you look at the very top we've got the top down guidance of attention which is very much the sustained attention uh processes and that's really subserved by these long-range front of varietal connections working memory as well as

well as we move more eventually down that prefrontal cortex you start to bring in control and in the inhibition of actions so for example the inferior frontal gyrus and its connections to the basal ganglia all these areas have been shown in large-scale studies of ADHD to work less effectively or to have subtle structural differences and then as we move right down more towards I guess the the bottom or the ventral almost the orbital uh frontal areas we have areas that regulate emotion with their connections to both the limbic system but also arousal system so my

sort of take on the cognition of ADHD is very much this push and pull between these top-down set control systems that try to keep us on task and consistent over time and these arousal systems that can work in our favor to bring us our attention back by flooding the cortex with dopamine or noradrenaline but unfortunately do wine with time and so the poor old child with ADHD in class needs to be able to maintain attention uh over long periods of time often but really their attentional system is just not geared for that you know the

the attention decays over over much uh more rapid time courses this is also another one that I find really useful uh and hopefully you do too uh because I think you know the portal clinician when a patient rocks into the lab needs to resolve this conundrum here and that is where is the patient going to be on this inverted U-shaped curve or on the left you've got a situation where you've got too little dopamine too little noradrenaline stimulation and a situation where the patient may be showing distraction or impulse control unguided poorly focused behaviors and

then you know with a little bit of stimulation of alpha 2A and D1 as would be occasion for example by the use of methylphenidate or dexamphetamine you're hoping to get the patient to this sweet spot where they've got well-guided well-controlled attention but of course if you push them too far you can go to the other end of the uh of the curve here and you can get a quite a rigid inflate flexible cognitive State and the patient can even become quite a stimulus bound or inflexible sometimes I've heard that described as almost like a zombie

state where the patient really doesn't have any cognitive flexibility so balancing these is incredibly difficult for you guys and you know which would really is a motivation for us to be able to identify biomarkers that can help to allow you guys to position or work out where people fall on these curves and optimize treatment so you have my respect for doing this day in day out and trying to optimize treatment for children and adults with ADHD so we moved through to some genes now and the gene work which is uh as I indicated earlier where

I started in Dublin and this we really started through the candidate Gene approach whether we were trying to I guess find evidence that genes of the catecholamine system were altered in frequency or gin variations while altered in frequency in cases versus controls and a lot of this work was done in Dublin by a whole range of people including Ashling and Michael Gill Ivan Keely a whole range of people did really groundbreaking work at the time relating to identifying susceptibility genes of the catecholamine system this is not I'm putting this up not to dwell on each

of these individual ones but more just to show you the way the field has developed and then of course you know we had concerns that these genes really weren't replicated and that was the entry point really for end of phenotypes we thought if these genes of small effect uh if we're trying to link them to phenotypes that are subjective uh that are often perhaps a little bit imprecise maybe a better way would be to try and Link them to more refined phenotypes so moving down through this hierarchy to objectively measured um aspects of cognition all

the way through brain networks circuits and cells and the expectation was as we moved down this hierarchy we should be able to pick up stronger more meaningful and hopefully more reliable signals and this was the whole concept of the end of phenotype and one I've spent a lot of time in my life working on and I'll come back to it uh in a in a little minute but the um and these just some examples of the work we did uh partly in Dublin and then partly since uh really Prosecuting this argument or trying to test

the argument whether copomb engine variations were related to objective measures of attention and inhibition and also a lot of work are done by groups in Toronto at sick kids for example but then as we move through I guess we had the Advent of genome-wide Association and for a long time these weren't particularly successful because uh in hindsight again we realized that the sample sizes uh for the discovery were just too small particularly when one had to take into account uh the very very stringent Corrections for multiple comparisons so bear in mind you're testing about a

million points on the genome so you have to divide your Alpha by a million to correct for all those multiple comparisons and the chance of picking up a type 1 error just by accident so that leaves you with what seemed for a long time as a impossible p-value to meet um enter the eyesight study in Scandinavia where they were able to go to dried blood spots and amplify DNA do a g wasp and then link that in the uh the population based records to see the who had ADHD and you know it really changed the

game a lot of the cohort went from being quite small to over 20 000 patients and they identified uh 12 genomic loci and you can see them on this so-called Manhattan plot Manhattan because it obviously is meant to resemble the skyline of uh of Manhattan and these uh genes turned out to be probably predominantly those that coded or encoded are neurodevelopmental processes in the brain there weren't any obvious dopamine or non-adrenaline related candidate genes there were some genes such as dust six and fox P2 that have interactions with dopamine but they were certainly none of

the original candidate genes that we'd been hoping for what they found very interestingly was that these um from this G was they were able to run genetic correlations with other traits and what you can see here is the genetics of ADHD relate to a whole range of other genetic correlations and many of these as clinicians I'm sure you probably wouldn't be uh surprised with so for example there's positive genetic correlations between ADHD risk and obesity risk for example there's negative correlations between ADHD risk and childhood IQ and years of schooling so these are all fairly

uh interest very interesting and certainly very robust uh relationships from these large data sets then because they weren't satisfied with the sample size that they had they found some more uh patients and this is the most uh recent GEOS again driven by Diddy domantas where the sample size has now increased over 30 000 and you can see here that the Manhattan skyline is now getting significantly uh denser you've got more high-rise spiers uh going up and uh now there is 27 independent genomic low slides so this paper is now accepted uh nature Jacks but if

you want to have a look at it you can grab the the pre-print on um on um mad archive uh interestingly in this one there is a bit of a signal for ADHD genes and genes expressed in uh dopamine and but dopamine dopaminergic sorry uh midbrain neurons so the you know the the dopamine hypothesis and catecholamine hypothesis uh has a little bit of a little bit of in it uh still and that is one thing that we're pursuing in our cellular modeling work for ADHD so where I think the end of phenotype approach does still

uh hold utility is as a mediator what we came to realize in time was that the genetic architectures of the end of phenotypes that we hoped would help us reveal the underlying genetic structure of these disorders turned out to be just as complex as the ADHD trade or the depression trade or the schizophrenia trade that we were hoping to get get a handle on and so we know this from large-scale studies of either structural or functional brain measures for example you still require you know large 20 000 plus GE wash samples to pick up reliable

signals uh equally but where I think the end of phenotype concept is useful in a mechanistic way is following uh the logic of Kindler and Neil who are very uh you know famous Behavior geneticist is as a mediator of genetic liability and broadly what this says is the genetic variation causes variation in the end of phenotype and that in turn causes variation in the psychiatric Disorder so following this example it allows you to set up some quite mechanistic studies to try and ask okay in my case how does near how do neuropsychological endophenotypes for example

work in this schema can they act as a end of phenotype where for example polygenic risk might operate through that endophenotype to influence ADHD traits and so that's what I'm going to tell you a little bit about now a study that we've done where we use the ABC data which is a large population-based child study in the U.S publicly available data and what we're asking is how does genetic risk for ADHD operate through this cognitive endophenotype to influence ADHD trades I'm not sure you can see the mouse I'm waving around with my my finger here

in front of me or not no one you can okay um so as I said we use the data this was actually a lockdown project in Melbourne uh done by a very talented honors student I had Mia Moses who I don't think Arena and I her other supervisor met the whole year in person but nevertheless she managed to pull off this uh really great project and get a lovely paper which is uh in um uh will be online soon at molecular Psychiatry so the logic of this one was we took the large data and then

we looked for cognitive measures which were theoretically related to ADH stand for which there was evidence that they might act as end of phenotype so we used working memory uh we used a response inhibition measure from a stop signal task and we derived a measure of reaction time variability also from that stop signal task working memory is I'm sure you are probably all aware refers to the ability to behold information online in order to be able to operate it on in a goal directed way it's incredibly important cognitive function for a whole range of things

Maths for example even even uh even spelling I would argue uh and very linked to educational outcomes so we wanted to understand how these performed within that framework we derived polygenetic risk scores which um I think a lot of people are scared of this sort of concept but it's actually I think uh a really useful and not that hard to understand what you do with apologetic risk score is you take a base data set which in our case is the large international g-was data set and what you're creating is a weighted sum of the risk

variants that were identified in that study and then you're asking how they map to your cohort under investigation and what poliogenic risk scores or the way in which they behave is that they form a normative distribution where increasing polygenic risk is associated with increased levels of the trait and so what we're able to do is take the base data set from Diddy de Montes which you generously allowed us access to pre-publication and then we're able to calculate that polygenic risk in the ABCD data for uh for for the ADHD trade and then what we uh

then did was look at the uh traits ADHD trait data that they had and here we were using cbcl data and also items from effortful control and creating a factor uh score estimates which uh we did that to make sure we had a a really reliable measures of the of the trade in that population base color so when I'm talking here I'm not talking about diagnosed ADHD I'm talking about ADHD trades if you look at these measures of course individuals with ADHD would score higher on these Factor scores that we've derived and we did check

that that was indeed the case for the relatively small number of folks who had a clinical disorder of ADHD so just to reiterate these are what we were using working memory response inhibition and variability I'll talk you through in each case uh with the direction of these effects and so here what we're asking in this mediation analysis is first does the ADHD risk polygenic risk relate to the ADHD traits second there's a variation in the cognitive mechanism relate to the trade uh thirdly does ADHD polygenic risk relate to the cognitive mechanism and fourthly can we

see mediation in the sense that when we consider the cognitive mediation does it account for the relationship in part between ADHD polygenetic risk and ADHD like traits we did this in a very rigorous way you have to control for a lot of comparisons here and it's actually quite hard in this type of analysis to pick up all of these associations and in fact the first study to show what will show in a population-based data so first what we did was derive the polygenic risk score uh you can see on the on the left hand side

here this is the process by which we identify a p threshold that we'll use or take forward for subsequent analyzes and what you can see on the right is the association between the ADHD trait measures and the polygenic risk scores so what you can see is that higher ADHD polygenic risk is associated with more pronounced ADHD like traits which is exactly what we were expecting and hoping for if we hadn't got that we would have been in a difficult position to go forward with these analysis Nick next we check the Second Step does the cognitive

mechanism relate to the trait and what you can see here is it does statistically these are very robust effects uh effect size wise they're still pretty modest because you've got um you've got fairly subtle effects that we're picking up in large samples here but you see that hierarchy memory related to less a pronounced ADHD traits higher stop signal reaction time which is indicative of poorer inhibition was associated with more traits and a higher variability that is more fluctuating performance was associated with more traits as well so these were all really encouraging as well and next

we tested polygenetic risk to the cognitive mechanism and what you can see here is that we found effects on the very left for working memory and polygenic risk and on the very right between variability and polygenic risk all in the expected directions we didn't find an effect significantly that is for the Stop signal reaction time and that's also been followed up by subsequent study I've done in a collaborative meta-analysis and I think that this task is just to uh variable in kids with ADHD to pick up a reliable effects then we tested the mediation and

of course uh we found something otherwise I wouldn't have built a whole talk around this that would have been really disappointing for you all um is that we found that the relationship between ADHD polygenic risk and ADHD like traits uh was indeed partially mediated by the by sorry working memory partially mediated this relationship you wouldn't expect full mediation here we only ever expected a partial mediation and we were you know really delighted to see a really robust effect for this so if you can think of this as putting working memory in a causal relationship between

ADHD risk on the one hand and ADHD like traits on the other we also found an effect for Reaction Time variability in exactly the same way so Reaction Time variability partially mediated the relationship between ADHD risk and adhc trades and for me personally this was ratifying because you know variability was one of the traits that I worked on way back when uh in Dublin with Ian and Michael uh in the Canada Gene era but here we were able to take those learnings and use it within the era where we've got G was and polygenic risk

scores and fitting them with within causal model models what we did was also check the specificity of these analyzes uh because it could be for example that any old genetic risk would show this relationship so one of the beauties of the polygenic risk method is that you can calculate orthogonal measures of risk or liability for any trait for which says reliable gwos evidence so we focused in on the major psychiatric disorders bipolar schizophrenia major depression and autism and we asked whether these same relationships would hold true we first showed that major depression polygenic risk did

relate to ADHD traits for you clinicians out there will know that ADHD and major depression certainly have some overlaps and there is some past literature are relating to that and we also showed that ASD risk apologetic risk related to ADHD traits again I would argue probably not a surprise for you guys we didn't see any effects for bipolar or for schizophrenia then when we put it into the next step what we found was that none of these uh polygenic risk scores that we took forward for major depression or Autism Spectrum Disorder related to our cognitive

phenotypes so although the polygenic risk did relate to the traits the polygenic risk did not relate to the candidate cognitive endophenous parts so given there's no association really statistically we can't move forward to mediation and it really shows us that the prior effects that we had for mediation really do appear to be ADHD ADHD specific and specific to the the cognitive mechanisms by which genetics plays out to influence symptoms so just to summarize that what we've shown is that we can indeed uh see a partial mediation uh for variability and response time uh sorry working

memory and response time variability uh uh intervening or mediating between genetic risk and ADHD trades it does appear to be specific for genetic risk for ADHD it isn't present for a major depression or autism spectrum disorder and what was really interesting to us was that this confirmed uh findings that were found earlier by Joel niggs group in a in a clinic and clinic and sorry case and controls and apple Clinic controls but here we confirmed it in a population-based data set of much larger sample size as well so this really you know gives us confidence

that these are meaningful effects so what does this mean why is this interesting I think it's interesting because putting it back in those models that I gave you before we know that these top-down uh guidance of an attention and thought uh for example they're very much the fronto parietal mechanisms we know those networks of the brain are really robustly modulated by catecholamines we know this from animal studies both rodent and macaque for example the work of Amy Ernest and earlier earlier even Patricia Goldman or a peach we know those areas the prefrontal cortex are exquisitely

sensitive to their catecholamine environment and here you've got those very networks uh being played out here in the data that we're showing you both for working memory and for variability other folks Chan at our show that these long-range front of posterior white matter tracks uh you know the the Integrity of those white matter tracks uh was uh different in folks with adiation versus control and if you put it in a task based setting there's an old paper of ours where we go kids with ADHD and mental rotation task which is almost a paradigmatic measure of

working memory and you see that these very same frontal parietal areas particularly in the right hemisphere of folks with ADHD kids with ADHD were under activated compared to controls so this sort of moves us to an area where the genetic risk as it relates to working memory and ADHD trades I would argue is probably going to be underpinned by the Integrity of these from the parietal networks and that gives us a next step in the puzzle that we can pursue using large-scale data sets that have both genetics and brain Imaging for example uh I want

the point here is also that the clinical heterogeneity in ADHD is massive right we are picking up with relatively small effects here in terms of the uh the effect size we know that kids that come into your clinic are almost as different one is almost as different from the other one and so now what we've got uh is this scenario where we've got a neuropsychological profile that is although can be picked up as different in a group of ADHD cases or controls it's not reliable enough at an individual level to be able to be used

for Diagnostic utility for example so the likely scenario here is that we have multiple cognitive mechanisms operating in parallel each of these will have likely its own genetic idiopath pathophysiological mechanism and to put it here in another framework you've got both you know the cognitive in the middle there which is really the ones I've been referring to uh and and prefrontal areas looping down through the basal ganglia and out to the thalamus but then you've got these whole other limbic areas with recurrent circuits as well and then yet again motor areas of potentially related uh

to the hyperactivity and the motor symptoms as well and in any one child you may have any some multiple of these operating so it becomes very difficult to map out the clinical picture in these so clinical and neuropsychological heterogeneity in my view is almost certainly going to be underpinned by these uh ideological heterogeneity and the multitude of genetic risks that we've now identified for ADHD and this is certainly going to give rise to multiple phenotypes and probably different patterns of of car mobility and so mapping that going forward is clearly a very big a very

big challenge so in terms of these neuropsychological domains a working memory inhibition arousal you know they do relate reliably to symptoms of ADHD across case control and now here population-based studies and I think working memory is a really interesting uh cognitive phenotype it's comes through from a whole range of work as a valid phenotype for ADHD and I think here we're giving insights into some of the brain Behavior Pathways by which that genetic risk for ADHD might play out and might relate to the symptoms of ADHD and as I said we know that it's underpinned

by these front of parietal networks that have potently modulated by dopamine nor adrenaline many of you will know that there have been a lot of attempts in the literature to strengthen working memory networks via for example working memory training and I think there's the jury on whether that is effective uh for ADHD symptom reduction is certainly still out uh I think what the evidence suggesting that working memory training improves work in memory uh is not too bad but whether it reduces symptoms of ADHD uh in the way that I guess I'm suggesting it may within

a causal framework I think the jury's very much still out on that but there are other techniques so for example we know that targeted brain stimulation using transcranial magnetic stimulation for example is also able to up regulate the activity of these frontal parietal networks and potentially in combination with cognitive training as some people have done could be another way of strengthening these working memory Network as well these um are the folks that have been involved in this work I want to thank all these collaborators I'm not quite sure I'm pretty okay fine right um and

just lastly before I did I don't want to inject your attention to one of the work a piece of work I've been doing on the side as it were is as president of the ADHD Association Australia is developing this Australian evidence-based Clinic practice guideline for ADHD so this is the most up-to-date evidence-based guideline for ADHD since nice uh you can go to this web link at the top and download it it's got 111 evidence-based recommendations for uh well they're not all evidence base 111 recommendations uh some of which are evidence-based clinical consensus or clinical practice

points and also a whole range of fact sheets that have been developed there which you may find are useful the other project we've been doing is trying to shift the dial with respect to stigma around ADHD particularly in Australia we have a lot of ongoing issues from negative elements of the media uh which constantly are trying to you know push ADHD down so one of our Advocates Lou Brown who's an appointed director on our board developed this talking about ADHD guide which is really uh our effort at advert to try and promote uh the best

way uh and the best language to use to talk about ADHD so that we don't fall into the Trap of engendering uh stigma Etc ourselves and so there's a whole range of little points here on this this is downloadable you can download it put it in your clinic give it to your families should they wish and it tries to really I guess both dispel a few of the myths but also put forward the best way to talk about ADHD not stigmatizing there is strengths based rather than deficit focused Etc so here's a few of the

sort of um uh little take-home messages from that we for example um you know many many folks with ADHD don't always regard their ADHD as a superpower but yet it's something you always hear in the media and particularly these days Concepts like hyper Focus Etc um and you know a whole range of things uh relating to Concepts that are often put out in the media so for example do children grow out of ADHD so I think it's really useful it's actually been endorsed by a lot of international groups including cadra uh if you guys wish

to make use of it in your own practices or your own areas of work all you need to do is is jump on the website uh and grab it and I think uh Helen and Ashley I'll leave it there I'm happy to take any questions you might have foreign yes thank you very much Mark that was a very interesting and stimulating talk and I actually have one question if I could kick off um you know your polygenic wrist score and study where you looked at traits of ADHD in general population study in a general population

sample I was just kind of curiously thinking as you're doing that do you think you'll do a polygenic risk score and traits of ADHD or measures of ADHD in a in a case control study as well or do you think that would be a value looking at cases and ADHD controls because do you think there'd be a bigger difference those graphs are quite you know you'd find would you find a bigger difference that'd be more convincing you certainly could do that you could you could relate apologetic risk to case control status uh there's no uh

problem doing that the issue will be getting a large enough international data set that also has the appropriate uh well-measured cognitive endophenotypes um there is a group within the PGC that has focused on cognitive measures and there has been talk of doing a similar type of thing within that group um but yeah you might expect the effect size uh to be larger if you if you actually had a group that was enriched with ADHD cases yes okay okay I have a question coming in here from Dr Hazel Fernandez hi Mark did you do any study

comparing with anxiety disorders just as you did with moderate depressive disorder and ASD um so uh Hazel calculating polygenetic risk for anxiety disorders uh if that's what you mean and linking it to the cognitive no we didn't we we focused on the the five major psychiatric disorders that are generally um looked at in this context but that would be absolutely nothing uh stopping you doing that I think someone else has actually mentioned that before uh and it would probably be a good idea because anxiety ADHD and working memory is a bit of a Nexus as

well thanks no worries well I I think for some of the take-home messages for clinicians I think some of the really interesting things are that there's now the Australian evidence-based technical practice guideline that we should look at I think many of us have looked at recently Steve forones and World Federation of ADHD consensus statement I think that's something um I know we talked about is at the college meeting um in in the spring College meeting we talked about that somehow the World Federation consensus statement so that's a take-home topic for for clinicians the other take

on point for clinicians I think is the the jiwa studies and the findings of the jiwa studies the genome-wide association studies and that we now have this very large genome-wide Association studies published in nature genetics so 27 rest loci and I think for families that's actually very important um that the genetics of ADHD have been confirmed in these major studies it has been confirmed that there are you know parts of the genome that are affected by ADHD and I think that's a take-home message for parents and that there are genetic findings which have been confirmed

in very large data population sets so we can say to people yes genetics is very important um I I personally think the polygenic risk score is a very exciting movement and I know we talked about that years ago trying to I had an old data set where we were adding together the three or four um genes that we were interested in at the time or genetic loci and but that's obviously in a new a new it's all moved forward new phase of that of course sorry I was just going to say actually the next phase

of that of course is whether polygenic risk scores can be used to inform clinical practice directly either in a pharmacogenetic way uh or just because it relates to or links to you know prognosis for example or outcomes Etc so there's tremendous interest in that and a lot of work uh ongoing in that space definitely definitely yeah I think a few years ago people thought there would be one area of the genome that we would be probing to try and decide if we could support a diagnosis of ADHD or predict a treatment response methylphenidate or side

effects and now things have moved towards the polygenic risks score yeah yeah I think people are building predictive algorithms right so it probably won't be a polygenic risk or on its own but maybe if you do have an algorithm that takes in for a gut it takes into account for example um you know symptoms your your uh your background your family environment and your polygenic risk there'll be a whole range of factors that can be put in and if polygenic risk becomes another predictor within those models that enables you to um you know or enables

you to enhance your clinical practice then I think it'll be worthwhile you know there's a lot of work going on in that that's based definitely yeah I I think that's the space to watch it's going to be quite exciting to see what happens next okay just wondering have we any further questions I just had a question that uh Mark can you hear me um I'm wondering about the maternal or paternal have you been able to differentiate I know it's a way it must be it probably will do it someday but I'm just wondering any thoughts

about that where is the lowest genetic load yeah the mode of transmission is it coming from yeah look they I I haven't seen anything from that with respect to G was studies recently that I can that I can pinpoint yeah some years ago folks in Dublin certainly did look at that for the rd4 and and Dowdy five from memory whether there was any differential transmission of risk alleles from mum versus dad um but I suspect it's probably not been shown to be particularly you know reliable or a big factor in the in the big scale

studies it probably also is due to the fact that a lot of the large-scale studies are case control right back in the day in Dublin we were always recruiting trios so we'd recruit mum dad and the child which enabled us to look for a mode of transmission from either the mum or the dad to the child but the big studies now are usually cases and controlled and they're not getting DNA from uh from mum and dad to be able to do that sort of condition analysis um so you know that might be why I haven't

seen it uh because the the data sets using trios these days are nowhere near as large as the the ones that are looking at ADHD cases versus controls thank you okay this is interesting yeah thank you there's a question on the chat do ADHD like phenotype individuals have ADHD or may have other diagnosis like for example PTSD or anxiety um so yeah folks with ADHD with a clinical diagnosis uh can you know definitely have concurrent or comorbid anxiety with respect to the phenotype um if we're analyzing it dimensionally we're kind of not setting a cut

off to know whether they would have at some level other other other extreme traits I guess in in the in the data set that I was talking about we haven't but certainly in clinical cases uh you would get more anxiety PTSD I guess yes but um I'm trying to think off the top of my head what the data relates to PTSD Ashley might know that one better than me yeah well there's an awful lot of talk at the moment about um chronic trauma and exposure to Childhood adversity and adverse childhood experiences and Trauma and that

predisposing to ADHD so that's and is that a phenotype in ADHD like phenotype or is that true ADHD and some discussion about that so I think and then some people call that PTSD but that's not firmly what PTSD is so um I'm not aware of any study that links through ADHD with true PTSD [Music] to trial s yeah trauma exposure to trauma is a risk factor for ADHD and there's no doubt about that and you talked about the the large effect of childhood adversity and ADHD symptomatology I think this that's really a clinical decision like

does somebody have ADHD or do they have PTSD or anxiety and you're going to be looking at the duration of symptoms and what the young person was like in preschool and or prior to the experience of trauma to try and tease out whether there's ADHD plus the experience of a trauma plus further symptoms okay have we any further questions well I think in the interest of good time keeping I'd like to thank Professor Mark belgrove I think it's been a very interesting morning and somebody said yes thanks for a fantastic start to my day that's

Hazel I think communion is it and I I think that's that summarizes where we're at with this talk