 We'll go ahead and get started. My name is Marshall Huang and I'm a first year resident here at the Marin Eye Center. It's my pleasure to introduce Dr. Y. T. Wong here as a grand round speaker this morning. Dr. Wong is an investigator at the National Eye Institute and has been involved in numerous studies related to diseases such as macular degeneration, diabetic retinopathy, and macular telangentasia. He has also been the primary investigator, the principal investigator in several NEI-led trials including the landmark AAREDS II trial. He has a particular interest in the role of microglia and how it impacts age-related neuro-information. Today, he'll be presenting on clinical findings in macular degeneration and the current clinical interventions. Please join me in welcoming Dr. Wong. Good morning. Thanks very much for having me. It's my first visit here in Salt Lake City so I'm really looking forward to learning more about the department. I've known colleagues from a long time that have done great work here so this will be a treat for me to see how it's done in your own backyard. This morning I'll share some findings from not our lab which I'll talk about in the afternoon but perhaps from our clinics in which we at NEI are trying to find out basic mechanisms in the lab and we have the opportunity there to translate some of these findings into perspectives in the clinic and to design clinical trials for interventions in diseases for which treatment is still wanting. So today I will focus on a disease that's probably familiar to a lot of you, geographic atrophy in AMD. So I'll start by talking perhaps for the sake of the residents, what is geographic atrophy and where in the spectrum of AMD does GA feature. I'll get into some of the details about the clinical features of GA, how does GA arise in the first place, how is it born and also how does it progress, how does it grow with time and can these clinical observations looked at carefully give rise to some ideas about how the process may be driven and how it should be studied in the lab. And so this will lead to some of the mechanisms that we can think about about how GA comes about and how it progresses and this might then open the door towards a search for new treatments for GA and I'll review it in some of the recent strategies and approaches that have been taken. Of course the treatment, the disease does not have a treatment right now so this will be things that have been tried and have not worked. So to begin, just to kind of say what is obvious to everyone in the room, AMD is a common eye disease with the large negative impact on healthy vision and this is global. And although the prevalence of disease in different parts of the world and different ethnic groups are quite diverse, this graph shows that regardless of location, the rate of early AMD, the prevalence of AMD increases with the aging so that across the entire world when you hit your 70s and 80s the prevalence is actually quite significant and occupies a significant fraction of the population and the similar trend is also seen with late AMD so as a result these authors have estimated about 200 million people around the world come 2020 will be affected by AMD and with the aging population that is going to edge towards 300 million around the world so AMD is indeed a significant threat to healthy vision and to put GA in the overall context we can think about AMD as a disease of stages and this is what I tell my fellows when I speak to them about AMD in the first place, remember that all the stages of AMD and when you're 17 of course our funders looks great it is devoid of any kinds of deposits anatomically perfect but with time and with exposure some of us not all of us develop a science of intermediate and early AMD and this takes a form of deposits that form underneath the RP layer in the form of large soft drusen and also above the RP layer in the subrennial space in the form of pseudo reticular drusen these patients do not demonstrate much symptomatology they may have slightly decreased visual acuity or some decreased archivitation but in the whole they are quite functional and may not even know that they have intermediate AMD so the significance of this disease this stage is that it puts upon an increased risk of progressing to the late forms that do cause the loss of central vision and you're familiar of course with the progression to the late exudative of wet form in which blood vessels grow from the retina from the coroll into the retina and these blood vessels demonstrate exudative behavior that causes structural disruption and functional loss and the other way that intermediate AMD can progress is towards the late trophic form or GA which is what we can be speaking about today there's no exudative or neovascular process but instead there is a loss of retinal substance as characterized by the GA lesion so in laying out all these stages of AMD one can perhaps think about in a kind of in a compartmentalized way what are the targets for AMD treatment and where are we now it would be wonderful if we can get the process in the bud and stop the formation of drusen in the first place and as you know there was no current way that we can do this successfully we cannot prevent the formation of drusen in the average trial the average supplementation in the form of antioxidants and zinc was actually given to patients without any drusen and of course with that study it was shown that compared to the placebo group the rate of formation of drusen was not changed by arid supplementation so when you guys are providing supplements to patients it's perhaps clear to is good to communicate that this does not prevent the formation of drusen so as a result we do not prescribe arid supplementation to people without drusen and what about people who do have intermediate AMD can we today prevent the progression of early intermediate AMD to the later forms and the answer is to some extent the average to supplementation as found in the average to study also built on the average one study can decrease the risk of progression to the late forms but with two caveats one is that the effect is only partial that it takes about the 25 to 30% discount of the overall rate of progression and second the protection that this partial protection it's only for the auxiliary form and if you would on its average supplementation it does not decrease the rate of progression from the intermediate form to the GA form so average supplementation does not decrease the rate of progression to GA only to AMD and it's also perhaps important to say that in the case of wet macular degeneration can we treat wet macular degeneration the answer is yes that the onset of anti-vegetable treatments in 2006 have really made a sea change in the overall approach towards this disease but however can be today in 2019 treat GA and the answer is no we don't anything that can slow down the rate of progression of GA once it has arisen it's perhaps also timely to say that average supplementation does not decrease the rate of GA once is arisen the rate of growth of GA with the unsublimitation or not unsublimitation is not different from each other as shown by the errors analysis so given that that's the overall kind of scope of the way we are today perhaps we can dive in a little bit deeper into specifically GA the definition for the residents is indeed that this is a late-stage manifestation of the atrophic form of advanced AMD and clinically speaking we can see that this exists as a sharply demarcated area of hypo pigmentation indicating a loss of retina and RP atrophy of a minimum size of 175 microns in the biggest dimension on the reading center level this is varied from 175 to 400 and through this error of atrophy we can see an increase of visibility of a quarter vessels underlying it this clinical definition is helpful for us to make the correct diagnosis there's some masquerading syndromes but also important on the reading center level to monitor define and measure the limit of the lesion of time on fun is our fluorescence our life is made easier by a higher contrast view the RP gives out a overall degree of auto fluorescence and in absence RP a sharply demarcated area of decreased auto fluorescence observed in the GA lesion eye both on blue and green like excitation and given that these perspectives have been well established of the field has been turning more recently to defining GA on OCT imaging and a consensus group has kind of defined OCT as a zone of RP loss of created at 250 microns in diameter I can see this is the region they're talking about the absence of RP cells and with that accompanying overlying decrease in overall retina thickness and photoreceptive degeneration in this area and because the corollate can attenuate the signal that passes into the RP can attenuate signal passes into the corollate the region of RP breakdown is also highlighted by this region of hypotransmission into the corollate so these features help one recognize GA on OCT and also allows the reading centers to define and measure progression of GA and lastly events in OCT and geography has allowed us to observe flow in the corollate circulation certainly in the GA lesion we do also observe there is a loss of this fine features that represent the corollate capillaris in the GA lesion itself although the large GA large vessels are well observed in the GA lesion the dark empty spaces in between these large vessels indicating the loss of corollate capillaris and emergence of these flow voids so clearly from OCT you can see GA is a multi-level disease extending from the retina into the RP into the corollate and these perspectives have been supported on histopathology if you look at the retinal section you can see a clear zone of atrophy in the GA lesion and zooming in on the edge of the lesion we can observe paralleling with OCT a loss of RP cells and overlying photoreceptor atrophy and Jerry Levy and his colleagues have done great work in demonstrating that this loss of corollate capillaris seen on OCT is indeed anatomically present they mounted in the form of flat mounds RPA corridor samples and examine zones in which RP cells have been intact moving into the transition zone and moving to GA lesion centrally you can see indicated in the loss of brown the loss of RP cells and corresponding to that also a loss of corollate capillaris that comes along with it so given that this is you know how GA appears the field is approach the question of asking how does GA arise in the first place and in the arid study they have been folks entering into this study with large reason they have developed GA in the course of the study so some years ago Rick Ferris and Michael Klein had collected all the patients that developed GA during the arid study and then walked them back in time to say what was going on prior to the onset of GA so this is an example of one of the patients that they had examined you can see that in 1993 to some years ago some of you were not born and this patient had a large drusen in the center of Immaculate and with time the number of drusen and the extent of drusen increase with the emergence of hyperparameter changes and during the course of this progression there is a regression of drusen in this particular area and emergence of hypopigmentation in this region and with the collapse of drusen there's also an emergence of a well circumscribed atrophic lesion that is known as GA so this is how GA is born and looking across multiple patients Rick and Mike were able to confirm that the sequence events leading to GA is one of drusen regression of large confluent drusen and also the often the presence of hyperpigmentation hypopigmentation indicating RPA changes as well as the emergence of calcified drusen that are described as refactile deposits and this process of drusen regression leading to the birth of GA is also actually in OCT findings in which more recently in which you can see in this particular case a long standing drus which can be present for many years undergoes regression we don't know what the trigger is and with that regression there's also a subsequent loss of RPE cells and flores of the cells and an increased transmission into the core of it. In a smaller study we've done at NIH we also looked at patients with intermediate AMD that demonstrate drusen regression in the early aftermath of drusen regression as seen in this funnest picture over here the drusen following regression this area looks quite unremarkable on the clinical examination at first but even at this early time you can see stick matter of RPE change in this region indicating that drusen regression is accompanied locally in the same area by degenerative changes on the RPE layer indicating that drusen regression may potentially be causally driving these degenerative changes and given the choice of having drusen and not having drusen we usually choose not to have drusen but it's also paradoxical that a regression of drusen is the gateway towards atrial pathology so investigators some years ago have asked the question can we get rid of drusen but not have the bad consequences that come along with it can drusen be used to regress without negative consequences so some years ago Maureen McGuire and Stuart Fine had organized the CAP study this randomized kind of good trial which employs a low intensity argon laser causes minimal damage to the retina in the treatment of eyes with large drusen it has been observed that a laser treatment of this nature can induce some degree of drusen regression and if we use laser to induce drusen regression would this be good drusen regression or would this be bad drusen regression so the overall result of the study some years ago did not however demonstrate any changes in the rate of regression to advance from of AMD with without in treated or observed eyes and specifically for GA both the treated both the treated eye in the dark line as was observed I progressed the geographic atrophy at the same rate neither increased nor decreased this strategy was a more recently revisited by Robin Geimer and her colleagues and Australia in this case they didn't use argon but he's a different sub-threshold nanostatic laser that does not induce retina degeneration but is capable of inducing drusen regression and they also ask the question if this caused a decrease of progression to the late forms of macular degeneration they have a small number of eyes in this random life study the ultimate kind of bottom line was that it did not significantly change the rate of progression although there's a lean towards decreased GA progression but interestingly in post hoc analysis and we're dealing with very small numbers here they did find that in eyes that have large drusen but no reticular pseudo drusen that these eyes actually were helped apparently by laser whereas eyes with drusen and pseudo reticular drusen were hurt by laser in a sense that they progressed to advanced disease at a faster rate so does true does laser treatment work only in eyes lacking of particular pseudo drusen that's something to be considered for the future but it's likely that we've not heard the last of the drusen regression using laser as a potential therapy I'll mention very briefly other ways of trying to get drusen to regress Demetrius Wallace and Joe Miller at Mass Eye and Ear had published a small study in which they treated our patients of high dose of silver stat and 80 milligrams per day there is the theory that drusen formation occurs because of an imbalance of lipid transport in and out of the corollate from the retina and so if you were to change this balance using a torbostatin and perhaps again change the accumulation of drusen so in the small study in the half of the patients less than half they found that these patients over a course of 12 months demonstrated drusen regression that perhaps was they thought unexpected from the natural history of the disease so whether this is something that it will bear for the fruit or remains to be found so once drusen is born how does sorry once J is born how does J then subsequently progress I think everyone is familiar with the rate of expansion of the J legion once it occurs this movie shows that and patients that are half a skatoma form during the birth of J will see the skatoma increase and deepen with time and visual acuity also progressively decreases the functional time to and this overall progression of GA also seems to be a little bit different and in this phenotype from the initial birth in the sense that the sudden regression of drusen is not a precondition instead it spreads progressively as a burning brush fires or extending from the center out to periphery so to recap where are we in 2019 we do not have a way to prevent the formation of GA we don't have a way to treat or slow down or rest of progression of GA once it's arisen and suddenly we don't have any treatment to restore whatever vision has been lost in GA so our challenges up in front of us are large and in thinking about this scientists been going back to the lab and asking do these clinical observations give us some ideas about what the underlying mechanisms might be what are the specific processes that we can think about in terms of trying to come up with something that can be a successful intervention so in the lab we've been having this conversation with us ourselves this is a list of questions for which I don't have the answers but I'll pose them anyway the question may be phrased how and why did you inform what are the cellular sources of drusen where did they come from who makes them RP cells seem capable of making many components of drusen great Hagerman here I have also suggested immune cells may also contribute to some extent to the composition of drusen and also proteomic analyses have also speculated on whether the circulation brings in proteins they also contribute to the formation of drusen and this where drusen comes from may be important part of preventing drusen from forming the first place there's also the other side of the question perhaps formation is only half of the balance there's also a clearance of a drusen maybe 17 year olds make drusen all the time but they're successfully cleared is there a decrease in clearance and what mechanisms might apply it's been proposed a little bit barrier to diffusion of drusen material from the retina may be a factor as are of the cordicapularis atrophy maybe the pipes that take away drusen into the circulation are decreasing with age and allowing formation of drusen so some of these perspectives we feel may be important in thinking about what how to prevent drusen formation and leading back to GA how do we understand drusen regression that gives rise to GA the drus is actually a lot structure and when it regresses over a course of a few months what is actually contributing to that regression who ticks the drusen away it's been proposed that immune cells were capable of phagocytosine material might contribute it's also been proposed that RP cells which phagocytosis other segments every day might also turn as attention to drusen to participate in phagocytosis and Christine Gertjo has also proposed that a drusen regression may be a passive process the RP cells may making drusen all this time but when they're sick they stop making drusen and the drusen naturally regress through a lack of reduction so I think that this idea about drusen regression and what drives it maybe something that's important and also the consequences of drusen regression so drusen regression doesn't happen often without atrophy but how do we link to these two aspects how this drusen regression result in death of RP cells for receptors and cordicapularis what are mechanisms the cost RP cells to die might there be some processes happen within cells such as oxidative damage and inflammatory activation always are these cells killed from the outside perhaps from immune cells that are found in a cycle of the lesion so these and also with all these hypotheses that can be generated there's also a set of molecules that have been indicated in genetic studies to indicate about 30s for loci that have been linked to aim the pathogenesis how do these processes if they're relevant relate to these molecules is something that the field is working on also to extend the argument a little bit not only to the birth of the jib the progression of GA we're also looking at a phenomenon which is happening across multiple layers of retina and it's also the question about which affects a primary and which affects a secondary does the death of RP cells drive the death of photoreceptors of vice versa there's something that the field is also dealing with and also the there's also the expansion of GA if your RP cell this next to a dead RP cell chances all that you're undergoing degeneration much increased how is that driven and what's responsible and that's something that's also being thought of lastly GA is not perhaps only a local problem there's also indication from adaptive optics imaging also perhaps flio imaging the overall macula is also compromised as a whole in the disease that the overall ground for GA expansion is actually undermined and the retina sensitivity in these regions also decreasing how do we understand a more global change that's going on how do we strengthen the Britain as a whole to these degenerative processes so I wanted to kind of a list of all these overall thoughts and provide overall framework in thinking about GA and perhaps in the last part of talk I'll talk about some of the approaches we've taken and following some of these leads so one aspect that we followed is oxidative stress suddenly oxidative stress there is increased it's found to be increased with aging the macula is a very oxidative actively active environment and with age of these changes due to oxidation accumulate with time and there's no better example than smoking smoking does amokally increase the rate of AMD I tell my patients if there's one thing you should do it's smoking to stop smoking that's the perhaps the largest environment that contributed to the risk of AMD it can see from these graphs over here that regardless of your genotype that can progress the increase of risk of progression for all the genotypes they're shown here smoking in these back row compared to non-smokers increases the risk of all the genotypes across multiple genes and Jim Hannon his group also shown that Hopkins that oxidative stress can decrease subtle protective mechanisms within RP cells causing RP cells to come perhaps become a more vulnerable to to degeneration so with that in mind we some years ago in 2006 we tried to pursue antioxidants as a as avenue for treatment in this study we use a class of compounds known as preparatory nitroxides these are free radical scammers and we have used this formulation of medication which is a lipophilic compound that can now when given as a eyedrop can pass through the cone is glora and within the eye be converted by intraocular asterises to this compound temple age which is a free radical scavenger and this compound in in vitro studies seems to protect RP cells in vitro and also in animal studies seem to protect the retina from a light induced injury so with that in mind we recruited patients with bilateral GA I can see from these pictures in the vitro patients there's a high degree of symmetry between left and right eyes and in these cases and there's also a good correlation between the rate of growth in the left compared to the right eye so what we did is to randomize our patients to receive the eyedrop treatment and only one eye and use the fellow eyes of control I know they try to detect a signal unfortunately although we found that the eyedrop was well tolerated and associated with few side effects in this report we did not find that the eyedrop resulted over a course of three years any significant effect on the rate of expansion of GA and also did not have an effect on the overall retina sensitivity around the lesion didn't seem to rehabilitate the retina that has not yet undergone degeneration so I think that this approach towards anti-oxidation is still something that is worthy of consideration perhaps there are better agents out there certainly we there are still other folks that are interested in pursuing this strategy another strategy that we also thought about it relates to the immune response and and suddenly the reasons for thinking about that are multiple for example in the GWAS studies and multiple genes have been indicated as being important in AMD risk I mean of these genes including the list over here have been found to be expressed by immune cells and and you can see that some of them are complement genes that have been featuring a lot in the amd pathology but also some genes such as tjaveta R1 apoe and also recently pill RB and PRA which are expressing immune cells and influence immune cell activation and certainly immune cells are found to seem the crime at the edge of the GA lesion you can see activated immune cells close to where the degenerating RP cells are and also been a number of animal models in which manipulating immune cells as well as providing means in changes can induce changes in the retina that resemble AMD in some extent so with all this in mind we thought that perhaps we should think about GA treatment in the form of an immunomodulation so we started we did perform these trials involving serolimus of rapamycin which is a inhibitor of a protein kinase called mammalian target of rapamycin mTOR and this is of course a immunosuppressive medication using uveitis but also FDA approved for as a immunosuppression post-venom transplant there was a place in and capacity stands to provide to prevent the recent analysis so with this immunomodulatory agent we worked with santa to create a formulation that is suitable for intraocular injection we perform two studies one involving subcontractive injection every three months and also intravisual injection every two months and an attempt to try to decrease immune activation within the eye to see if that is helpful for decreasing a rate of expansion of GA if we if this has worked we you probably heard about it so we did not find any positive results we did not see any large negative consequences but neither the subcontractive injection nor the intravisual injection are resulted in the rate of decrease of greater growth of GA we saw some in one patient we saw a strange expansion of the GA lesion and the thinning of the retina but suggesting in some situations this may be deleterious but a larger study published last year as part of the heuristics that are using serolimus intravitually did not show any positive or negative effects so we right now are perhaps narrowing our focus a little bit and focusing on one group of cells microglia cells which are innate immune cells that live within the retina I'll be speaking a little bit more about these cells in the afternoon and certainly these microglia cells which live normally in your retinas at all times are changed in their location coming close to drusen and coming to the areas of GA and suggesting that they may be abnormally activated and driving the rate of degeneration so we want to see if suppressing microglia activation in eye might be helpful so currently we're doing this study in which we're using a FDA-approved agent minocyclin which is an anti-biotic as well as anti-inflammatory I have been shown in animal studies to decrease microglia activation in the eye we're using also a novel design in which we're observing patients without treatment for a course over a course of nine months and then instituting or minocyclin afterwards from a preliminary natural history study we find that the square root transform of the growth rate across a number of Asian sizes actually linear over time but but however we introduce a intervention we hope to see the bending of the curve in a way that can reveal a signal so each patient is its own control each eye and history of the eye in terms of overall expansion will serve as a comparison group to the effects of treatment subsequently so we're doing this study using this novel approach and also a novel design and hopefully we'll be able to find some signal going forward. Others in the field have also been looking at immunomodulation perhaps in a more molecularly specific level with respect to complement and you probably heard a lot about this. I'll review some of this literature complement of course has been you know featuring heavily in the pathogenesis of AMD and the overall broad theme although this is not really substantiated by detailed clinical studies is that over activation of complement is something that is happening in the AMD eye and driving degeneration many aspects. The central molecule of complement is C3 and cleavage of C3 gives rise to the production of various mediators that can facilitate immunomodulation across cellulises so and this overall cascade is regulated by a series of positive as well as negative regulators so Roche in their studies have targeted one positive regulated complement factor D. The common factor D is a positive regulator that can drive the activation of complement so the investigators at Roche reasoned that if they suppressed CFD using a drug they call Dampen-Liesemapp they can change the outcome of disease in the eye. So Dampen-Liesemapp is a FAB fragment so it's an antibody directed against CFD and although the phase two results published a few years ago were positive and showed that perhaps there was some efficacy the recently concluded phase three studies involving a large number of patients did not show any effect so this graph shows the rate of progression the rate of growth of GA in three groups the sham group eyes are treated every four weeks eyes are treated every six weeks you can see that they overlap with each other exactly indicating a lack of effect of treatment the phase two study also indicated that the patients with polymorphism in a gene called CFI might have extra benefit again the study did not show any benefit based on that so the fate of CFD and the field is a little bit uncertain right now but that has not stopped folks from looking at other targets and ongoing study right now involves C3 the central molecule involved in Compton Gascade this is a recently concluded phase two trial and ongoing phase three trial where this drug Pocetagopin which is a peptide inhibitor peptide to C3 is provided to patients for a course of 12 months given individually every month every other month followed by observation from month 12 month 18 the results of phase two study are interesting and have been published last month they show that the growth rate of the patients are treated every month was decreased by 29 percent folks are treated every other month decreased by 20 percent but and you can see these graphs showing that treatment was initially quite subtle the treatment effects of subtle at the early phase of a treatment these are the treated arms versus the sham arm but with the length of treatment towards months 6 and 12 there's increasing difference between the groups indicating a greater treatment effect the longer the treatment has been around and when they stopped the treatment between 12 month 12 month 18 things became back to normal again indicating that there's a good relationship between dose and between the presence of fact arguing that this effect is real most interestingly perhaps that this is what's not the only effect of the drug was related with there's also the emergence of new onset wet macular degeneration in the eyes to be treated 20 percent of the eyes over a course one year develop wet macular degeneration with treatment and in the in the eyes of treated monthly in the eyes of treated every monthly this is a little bit this half whereas the the sham shooter eyes had a very low rate of progression so this is actually a also a real effect of the treatment that induces wet macular degeneration in addition to perhaps slowing down rate of GA growth so the balance the relationship between a wet macular degeneration and decrease of GA growth is something that the field is very interested in looking at so the phase 3 trial for this study is ongoing and recruiting new folks maybe a site and we're very interested in looking at what this trial might show us so with that I'll bring my comments to a close GA remains a big challenge and 2019 is a disease that's without any prevention or no any treatment or restoration at the current moment folks that work in the labs are very much involved in looking at the clinical clues that the patients are offering to us and looking at the causes and perhaps the drivers of GA at each manifestation of the disease both in terms of its birth and also in terms of its growth and in the clinic we're also very much interested in doing more clinical research looking at clinical outcome measures to more precisely measure and detect any effect that we might see in clinical trials we continue to invest in phase 2 clinical trials under proof of concept levels any phase 2 trials that show some efficacy would change and reorient the field as a whole and of course getting a treatment to patients will require confirmatory phase 3 trials that can give rise to a treatment so with that I'll thank my colleagues and at NEI and my clinical collaborators especially Emily Chu Rick Ferris Captain Kruper and also Karen Aquina who recently left Utah to come join us at NIH and also all sources of funding thanks very much for your attention happy to take any questions yes please so obviously the quest for doing something about GA continues where's Paul but I see Paul so there was a big headline of something about really exciting phase 2 2 a study about some modulation I mean this I'm talking this was like a you know ophthalmology times type thing that I just saw it and I didn't get it in any detail but did were you was did anything strike you mean that they claim like there was some early data who's heard that so often that I think these come and go the face I saw just enough that it did not reach statistical significance which is yeah many of these as they say if their industry sponsored are forward-looking and they're looking to keep going to the phase 3 but yeah I to the NEI's credit they're doing a lot of cutting edge phase 2 trials and really doing them right and unfortunately there's going to be a lot of a lot of things that don't work but that's the only way we're gonna get forward but what about the visual psychic there was also some some a visual cycling suppression for doing a lot of these trials and they're not failed as well yeah they are they're not we're still looking at them but yeah for a lot of GA it's failed and they're now switching back over to Stargard to try to do proof of principle and Stargard disease interest because if they can get something to work in Stargard then they'll go back but GA has been difficult but you can talk about how you know you're there's a lot of trials you guys have done and how you decide if you're the vast array put the NEI's resources on these how does that how's that decision made yeah the selection of targets and they're multiple as I try to and in addition to you know immune modulation and also stress there's only visual cycle approaches the additional molecules that have been implicated as of course of the GWAS study even if we don't know the function too well such as HTRA or arms to people are trying to design drugs inhibit them in the hope that we might find something I think that the limitation as a word suddenly is relates to lack of animal models and good animal models in which we can use to generate narrow down our field and so far the narrowing down of Canada's has been difficult to achieve so as a result of proof of concept trials that are perhaps easy to do had there's an agent for just FDA approved and can be repurposed or agent it looks like the risk-benefit ratio may be suitable so we can try to attempt those so much of the decision-making has been influenced by accessibility to some extent I think that as we dig deeper into the molecular genetics of these processes perhaps we can get closer to a more judicious selection of targets I think the field is changing all the time to with new basic science discoveries that link well to the clinical observations so if you find something that we see a recapitulation of phenotype as precisely noted perhaps that can be helpful and we're looking at this disease all the time with new eyes to perhaps using imaging technologies as flio to see something that we haven't seen before that can link to perhaps if flio indicates that visual cycling is really abnormal in much of the macula in early course of disease that might reorient ourselves towards using visual cycle modulation we also limited in terms of when and what stage we should intervene and I think that the phenotypes of birth progression drusen are really distinct and although the gene the genes implicated abroad in the sense that it applies to overall disease a selection of the stage that's appropriate for finding a signal is also crucial so we have a lot of questions and we don't have a lot of tools for narrowing the field but I think we're in focusing on the problem perhaps we can get a little bit closer so at any I we just been looking at things that are ready for for trial just just one of the thing that we so we've been looking at a large group of genotype phenotype pure groups pre going forward and so we've been able to get a large group of HR and when arms to pure homozygous risk and we're amazed at how many of those will go wet or to GA and never have drusen yeah I don't so I don't know how kind of small a fraction of the entire END population that might represent that's a very small yeah I was like it's a small group but but it's interesting they're it there obviously is a process that moves forward that does not have to have soft drusen and and and they get reticular pseudo drusen quite often but they're not they're not getting classical they're not getting classical yeah it's it's a fair it's a fair percentage of that group and you brought up a really important point which is say that do all the the way in which we understand in the we're just really a control of the area study mostly of large drusen etc actually did not take into account of pseudo particular drusen because we couldn't see them at that stage and if we included them with that change our classification and understanding of so and so on and so forth if you didn't take them into account most people get to wet EMD or advanced EMD through drusen very few do kind of take a different route but if you took them into account then perhaps they can be you know different so we're trying to we can't redo the average study because you know because they were not genotypes right but what we try to do now is try to use AI artificial intelligence to look at those funders pictures for which clinical readers could not score reticular drusen and to see whether we can score reticular drusen they would have scored many of these people as normals we reviewed the scoring system and it was highly ever-prone just because we couldn't see them well and I can see them well clinically to funders auto fluorescence can show them much better so we kind of right now we're trying to train an AI algorithm using a funders auto fluorescence and then look have them look at historical data to be able to re-score all the images from errors to see whether a new algorithm a new kind of framework can be driven and perhaps some of these genetic associations and how things work would become a little bit more clear that'd be great this place that's a really good question so the trial was just published and I encourage everyone to read it it came up in ophthalmology last month and the others had asked me to write a short commentary to go over the so that's coming out next month I think so but and then I tried to kind of talk a little bit about your point which is in short we don't really know in the study the patients when they develop at macular degeneration were were kind of stopped and they dropped over the study and they were not really followed so data relating to them was not really collected in the phase 3 study I think there's a little bit of a gut team of they would continue to keep and follow those patients give them anti- veg of treatment and continue to give them the investigation treatment at the same time without without discontinuation so that will kind of give us some data about the question you're asking but you know looking for if you want to predict what would happen and really is not so clear would you can suppress wet macular new vascularization and perhaps that will maybe that's an adaptive mechanism if you suppress that the benefits of GA slow down will go away can this wet macular degeneration which is emerging in a very different context be actually controlled well with anti-vegeta we don't know so the many is a very interesting situation the many unknowns I can't wait to find out what this phase 3 shows hopefully there be good mods to make sure that the patients are not going to be harmed in the course of the study this week for their upstream talking about the regression of Jerusalem just with my superficial understanding it would seem more likely it would be part of the natural history rather than something that's causative and are we getting closer to understanding that piece I mean you presented several potential macadames is there a horse to bet on yeah I think that the the cause and effect relationship between Jerusalem regression and the emergence of GA seems to be quite good although there I tell my patients that Jerusalem although it comes to the German word rock I'm more like shifting sand dunes than they are rock gardens so there's a lot of fluxes going on in many patients so but the fact that large amounts of Jerusalem regression can be locally and temporally associated with atrophy this suggests that the two are if not causally like at least kind of temporally and spatially like if one of you seem that they are somewhat causally like then what is that process I don't think the field has a full understanding but if you asked me to bet on the horses I would say that the process of immune cell clearance is interesting process and this is true in Alzheimer's disease where I'm like packs also deposit to which the immune cells to try to clear and they're also is accompanied by a number of new generation so the activation of immune system of clear stuff there's not there may be a good adaptive process but perhaps one that lacks sufficient discrimination so that the clearance also extends to cells that are required and needed so I think that that is some perspective I would like to learn more about and study further it appears in a trophy game D that it's real not only a molecular genetics tremendous variation but in the imaging patterns of the boundaries of the GA that define I think that those patterns of the autofluorescence changes can define the rates of progression in those areas in a number of observations we found that the photoreceptor ellipsoid junction and ELM disappearance actually preceded the RPE atrophy in the geographic atrophy and was related to the rate of progression we found that the geographic atrophy around the optic nerve proceeded and actually predicted the rates of GA progression and the patterns of autofluorescence often would predict the local areas of progression so so there's a tremendous variation here that we've got to somehow try as you said there's so many things that are affecting it how do we decide so we can define better methodologies to try to prevent this yeah I think that we're just mixing so many together confusing our ability to separate absolutely I think the other bird looking at kind of sections are also going to try to relate photoreceptor degeneration to overlying RPE degeneration and I've found quite a bit of variation across individual patients Frank Holtz looking at different patterns have defined trickling patterns that confer increased rates of expansion so certainly it's not just the burning border per se as the only factor but also perhaps of more ground factors perhaps determined genetically that can determine the overall vulnerability of of the macula how do we kind of put how do we categorize and understand that it's been a long time since the events of autofluorescence imaging in GA but I think that we still don't fully understand beyond the lesion itself what the patterns outside lesion might mean we are thinking about using artificial intelligence to try to tell us in which we follow the patterns people with patterns over time and of course artificial intelligence is a great way of pattern recognition that doesn't require individual observation or individual observers making hypotheses I do look at everything and just pattern a sort across a large number of eyes so hopefully we can use that power to to try to understand and categorize different patterns and assign significance to that so we don't know so that's one tactic we might take in the future thank you