 It's my pleasure to introduce our speaker, Dr. Rando Alecmitz. Just a little bit about his background. He earned his BA and his MS at Moscow State University. He earned his PhD at the Institute of Bioorganic Chemistry in Moscow. He completed fellowships at the Karolinska Institute in Stockholm, Sweden, and at NIH in Maryland. He has published over 60 papers and projects and probably way more than I can ever count on photoreceptogenetics, macular retinal, and cancer genetics. He is currently the research director and Akavela professor at Columbia University, as well as a professor of ophthalmology and pathology and cell biology at Columbia University. So let's welcome Dr. Alecmitz. Thank you very much. As you heard, I've been around the block a few times, spending time in Russia for 10 years where I got my practice in vodka drinking. But today, I'm doing a few firsts. I've never given a clinical talk because I'm a geneticist. I'm not an MD. I'm a PhD. So be patient and nice with me because this is an area that I really am learning. But I don't know that much about. At noon, I'll be talking more about genetics that I know a little bit better. So the Stargard disease and the ABCA4 gene was the reason I am in high genetics and I am at Columbia Heartless Eye Institute. Before that, I was doing cancer genetics mostly and worked at the National Cancer Institute here in the States. And I was working on ATP binding cassette transporters that are involved in many different diseases, the most prominent and well-known cystic fibrosis, for example. And these proteins are also multidrug resistance pumps that kind of oppose chemotherapy. So they pump drugs out of the cells. However, one of the members of the super family named ABCA4 was the one that is causing Stargard disease and many other phenotypes that I will show today. And that brought me into high genetics where we have had quite a bit of fun in both early onset Mendelian retinal diseases and with Greg Hageman here in age-related macular degeneration that I will not talk about today at all. So every good work takes a good team. And my team is usually quite small here. What I'm talking about today is mainly done by Winston, who is my study coordinator. But I bet it's the best study coordinator ever because he does everything. He not only coordinates the study, he does imaging. He analyzes imaging. He writes papers. So he does way more than he's supposed to. Jana has been in my lab for many years. She does most of the genetic studies on ABCA4 and the whole exome sequencing that we've done the last five years or so. We're done by Angela, who just got her PhD and left the lab. So here are some other people. Again, here is Winston. But there is also Caleb and Tobias, who did a lot of this work I'll be talking about today, and mostly on imaging of different retinal disease patients. So Starter disease, as any retinal specialist know, it is the most frequent Mendelian retinal disease. It has mostly juvenile onset, but you'll see during this talk and later today that it really spans the entire spectrum of age. It is characterized by loss of central vision. Usually you have macular atrophy as one of the first signs. It has some characteristic flex in macular region on the periphery that are made of lipofusion, and we'll talk about that also. And for me and for geneticists, especially what is the main cause of the disease, what is the cause of the disease. You have to have two mutations in ABC4 gene. What is the transporter of vitamin A derivatives in the visual cycle? So it has very specialized function and is expressed only in rod and cone photoreceptors. There is some discussion that it may be expressed also in RPE, but we haven't really verified that. So this slide is borrowed from Chris Balciowski. Actually, Utah in 2000 was the place where I met Chris first. He's at Case Western. He has done a lot of biochemical studies as you know on visual cycle, and he also was trying to do the structure of the ABC4 gene. It's a large trans membrane protein. The structure didn't come out that great, and we're still waiting for a better one. But this slide shows it's located in the raw outer segment disk rims, because it's a big protein. It can't be in the disk itself. It's on the rims. And this is kind of a schematic structure of it. It has two nucleotide binding domains that use ATP as the energy for transport. And it has also several trans membrane segments. The function of it, as I mentioned, it's a transporter of 11 cysts and all trans retinal in the visual cycle. This slide is borrowed from Bob Molday, who has done probably the most in terms of figuring out ABC4 function. And it shows this is how the visual cycle works. And ABC4 is in the rim, making sure that at first it was thought that it works only with the light, and it transports all trans retinal out of the membrane. But it also has been shown that it transports 11 cysts. So it keeps the balance of 11 cysts and all trans in the correct balance in the disks of photoreceptors. And if the balance is disturbed, you'll see you get into the disease, and you start accumulating those bisretinoids. The most famous is A2E, which are phototoxic and are either a cause of the disease. Now they say that maybe they are not the single cause of the disease, but these are the greatest marker of the disease and probably a cause in many cases. So this slide was made when we were talking about therapeutic applications. I will not discuss those today, but there are many ways and many stages in the visual cycle you can really intervene if you have mutations in ABC4 to try to alleviate the issues. And several of them are in clinical trials or close. The gene therapy we did at Columbia is in clinical trials right now. The chemical modification of A2E is in clinical trials. And some others are being proposed, but not yet made it to clinical trials. So the main way we image patients is the short wave autofluorescence. And this is slide again from Janet Sparrow's work. She has done the most in terms of figuring out the bisretinoid A2E I already mentioned. And it can be really shown that the A2E accumulates at very high levels in Stargard patients, in mouse models, a knockout mouse, and so on and so forth. So it is a marker that is greatly visualized. And now what we do at Columbia, we also quantify it. We modify the autofluorescence imaging by adding a reference bar so we can precisely quantify the amount of autofluorescence and give it a number of what it is at different stages of Stargard disease. And again, here is the graph that shows that we all accumulate A2E during our lifetime. And it grows. This is the normal curve. And you can see that most of ABCA4 mutated patients have way elevated lipophysicin. However, there are a few that do not show that. And this is also kind of interesting side of the story. And I'll touch upon that a bit. Another imaging technology we are now using is near-infrared autofluorescence as opposed to the short wavelength that measures the lipophysicin and A2E in both RPE and photoreceptor outer segments. The near-infrared signal detects the melanin in the RPE cell. They also discuss that probably melanolipophysicin, however, it hasn't been very well determined. But pretty much with these two methods, you can separate the RPE part from the photoreceptor part. And you can figure out quite a bit about the disease mechanisms. So here is the same person imaged by short wavelength and near-infrared image. This is a what's called normal. It is an un-diseased patient, an un-diseased individual. So now when you do the image the same patient here, we have a Stargardt patient. You can see a few things here on near-infrared where there are dark dots. That means that RPE is gone. So RPE is dead at those places. And you can see that they overlay perfectly with the flex. So when it was postulated before that what autofluorescence images, it images A2E lipophysicin in the RPE cell, actually I think, and this is what Janet Sparrow says, that most of the autofluorescence really comes from the photoreceptor outer segments because RPE is dead in those places where you have this very intense lipophysicin signal and here, from one of our papers again, you can see that the flex appear later. So this is the same patient first imaged with short wavelength near-infrared. You can see that there is no flex in these spots where the RPE has died, but it appears there later exactly at those locations, meaning that it is not in the RPE cells, it is really in the photoreceptors. OCT, of course, is another method that is very widely used and we use that a lot in all retinal diseases and also in patients with ABC4 disease. And these days you can really have a very detailed images of the retina and there are several discernible bands in the retina, for example the ELM, external limiting membrane, it has several names, we call it ELM, which is very faint and narrow band in healthy subjects and this is an RPE band. And you can see that in ABC4 disease, in Stargard disease, one of the first things you see is the thickening of the ELM band. So what is ELM band? It is suggested of being the connections between Mueller cells and between Mueller cells and photoreceptors. So the kind of suggestion that is not confirmed at this point is that this represents Mueller cells fighting the disease since Mueller cells are involved, as you know, in cone visual cycle that is different from RP visual cycle and just trying to keep the photoreceptors alive and going. This is just the plots that this is one of the first signs of the disease and we also have looked at the thickness of ISC band and then the relationship of these two and this is a very well quantifiable feature in Stargard patients, in all patients with ABC4 disease. Another interesting phenotype, sub-phenotype, that we see is called optical gap. This is something that we see in many patients but mostly in those that present with bullseye maculopathy as opposed to the classical Stargard picture. Interestingly, this phenotype is associated mostly with this most frequent and quite famous mutation in ABC4 1961 and it's a gradual loss of photoreceptors and cells and RPE in the macula where a large gap is forming that will later collapse and you'll have the macular degeneration, so these are different stages through which a large fraction of Stargard patients actually move. Interestingly, again, I may have a slide about that. These patients do not accumulate much like a fusing at all. So again, it shows that the same disease is happening by several different methods and we're trying to find out specifically what is the biology behind that, why some mutations like 1961 cause a very specific phenotype that is not associated with this well known classical high level of accumulation of lipid fusing. Some other more recent methods and quite fancy are in FAS, OCT, where you do the segmentation of the OCT image and then you can look at the same place in greater detail and also the OCT angiography where you can really look at the coriocapularis here in healthy. You see it in the diseased individual. It's completely gone. And you don't see it. So this is in general where you talk about Stargard disease. It's really not that simple and not that very uniform disease. All of these autofluorescence images are from people with two mutations in ABCA4. And as you can see clinically, it's often very, very difficult to put the precise diagnosis for every patient. And I always say that, and I will specifically talk about that today at noon, that these days you have to do genetic analysis. And only then you can confirm the disease because it can be from a very small lesion in the macular, like a bullseye, to really pan-retinal degeneration. And actually, we have shown that patients progress very differently depending on the ABCA4 mutations and some other modifiers that we are trying to find out, meaning variants in other genes. Now this is more of a classical way how Stargard disease progresses. So you have a juvenile onset maculopathy. Again, first it starts with a kind of a small lesion, but flecks start developing, then they spread all over, then they consolidate. And really, once you get to this stage, you have a very serious vision loss. Usually this is happening over quite a few decades. Now, as I already mentioned, this mutation that is the most frequent mutation, about 20% of patient carry at least one 1961 mutation, is quite different because when you see these patients, you can tell they will never develop into severe disease. They are always, they will stay pretty much, I think. Now this is a, OK, let me, I'll come to that slide later. Again, this one I already showed, 1961, does not accumulate lipofusing much at all. And it stays within this kind of a clinical picture. You have a loss of central vision, but you never, you don't get flecks, you do not accumulate lipofusing, you will never get panretinal degeneration. And age at onset and fleck distribution is also kind of evenly distributed between the groups. But if there is something in terms you can tell to the patient when you know the genetic basis of the disease, is that if you have at least one of the two is alleles, is G1961E, you can tell them quite surely that this is the disease progress they have. And they will not have a major degeneration of the entire retina. Many patients with mutations in ABC4 get late onset disease as opposed to juvenile onset. Age of onset can be as late as in the 60s. Sometimes they are then confused with age-related macular degeneration, it has happened quite frequently. We have many patients who were diagnosed past 60 years of age and at first with AMD and by only genetic screening we figured out that this is late onset Stargardt and this is kind of a classical picture of late onset disease. Another thing what we see, and again it makes sense, late onset, usually the disease onset is determined by the visual loss. These patients have foveal sparing, meaning that they have a small patch of cells right in the fovea, as clearly seen here. And although they have quite substantial degeneration, their vision is sometimes 20, 20, sometimes 20, 30, so they do not experience significant vision loss. And I would say 80% of patients with late onset Stargardt have the foveal sparing and very benign or minimal vision loss. So the phenotypically, they look similar, but actually 1961 is involved in early onset disease and not late onset usually. So now there are some more severe forms, cone rod dystrophy. Usually these patients are human knockouts, meaning they have two deleterious mutations in ABC4. They have no functional ABC4 protein. They have very early onset and they rapidly develop into pan-retino degeneration. Again they have major lipofusin accumulation and aggressive lipofusin accumulation and cone responses are diminished at quite early age. This is also pictures from patients with ABC4 disease. If you look at this clinically, you could say that this is probably looks very much like retinitis pigmentosa. Although there are several differences between the real retinitis pigmentosa and the one that looks like it caused by mutations in ABC4. Again, these patients are knockouts. They have no functional ABCA protein. But as you can see, despite the fact that they are genetically knockouts, these patients and then the cone rod dystrophy patients, their disease phenotypically is quite different, which we still have to figure out why. Because you would say that knockout is a knockout and you should get pretty much the same phenotype. But again, you have here cones dying first. So this is not like classical retinitis pigmentosa. So it's not true. And in autofluorescence image, in the fundus photo, it looks very much like bone spigil pigment. In autofluorescence, you can see that you have flecks and you have peripapillary sparing. So features of ABCA4 disease. So this is kind of the general picture of how what we call Stargard disease, now we call ABCA4 disease, is happening in different groups, depending on the mutations that are in ABCA4 gene. And some people never develop into the end stages. This is kind of what we call the critical stage or point of no return if you have mutations. And we call that also a classic Stargard group that develop into this stage, then you develop the disease further. However, as I said, if you have 1961 or you have very hypomorphic variants in ABCA4, you will not get to this transitional stage at all. And you pretty much stay in this stage, first three stages as we classified until the end of your life. So one thing what we are trying to figure out, and we know that right now a lot of discussions are about precision medicine or precision of tomology, is genetically it means we have to figure out all variation in patients that cause a specific phenotype. Because then you can advise the patient in terms of progression of the disease and in terms of possibly available treatment options. So I will talk about that at lunch quite a bit more. But here I'll just show one interesting example that we very recently published. Again, we had a pro band that had a pretty 19-year-old female and had an early onset vision loss at 19. She was 2050 and had younger sister with quite similar symptoms as it always happens. If they're affected, so this is the pro band, so this is the short wavelength auto fluorescence. It has pretty elevated lipofusin accumulation again as the bullseye phenotype. And as you can see, has also the optical gap on the OCT images. So again, most of these kind of cases are clinically diagnosed as Stargard disease and we screen them for mutations. However, there are some other diseases that can give very similar clinical presentations, occult macular dystrophy caused by mutations in RP101, which is usually dominant disease with very variable penetrance, depending on age of onset is different. Atromatopsia, again, that is caused by mutations in a slew of different genes, but is mostly characterized by color blindness, what Stargard disease is not. So we, as I said, automatically always screen similar patients for mutations in ABC4 and again, in this case, as expected now, we find that he is carrying one 1961 mutation, therefore the optical gap and the bullseye phenotype. The other variant is also very well-known mutation in ABC4G and when we were looking at the family, the sister showed quite similar phenotype and the mutations segregated with the disease in the family, both sisters were compound heterozygous for these, one mutation came from mom, the other from the father, and the brother was lucky to have no mutations in ABC4. Now, in the follow-up exam, the disease was quite stable, as expected, I mean, there was progression, the optical gap was getting larger, but one other thing what we noticed with this patient, specifically with the white field, auto fluorescence was this interesting kind of mud splatter phenotype on the periphery. So this is the proband imaged from almost all angles and again, this is a quite specific phenotype that is not associated with ABC4. So, but there are patients that present with this phenotype and this disease is called X-linked ocular albinism and you have this mud splatter type, original appearance. So this is a phenotype with carriers, females, and they have the diseases caused by mutations in GPR 143G. So this has been described, the phenotype has been described and there are many papers that talk about ocular albinism. Now in this family, the healthy brother was absolutely fine, but affected sisters had this additional phenotype feature. So this is again to put these disease phenotypes in perspective, this is a carrier of X-linked ocular albinism and these are the affected from this family. And we of course, since we had that good lead, did sequence the gene and we found that there was a disease associated variant in both affected sisters, not in the brother, not in the mother, but also in the father. Now, as many diseases go, there is the non-penetrance issue because the father should have been really affected. This is a pretty severe mutation, tyrosine to cysteine, but father showed extremely minimal signs. So he had no vision loss symptoms, but there were, if you look very carefully, some signs. So you can call this case a reduced penetrance that you see a lot in retinal disease, in very well-known diseases such as best disease, where you have sometimes very severe phenotype in kids but their parents are absolutely fine visually, although best disease is mostly also dominant disease like in this case. So this is the mutation, it is a new variant in that gene, it is highly conserved area, and it is predicted to be deleterious by all programs. So it is definitely this variant that causes this mud splatter phenotype in this patient. However, interestingly, you don't see that pattern in the macular area, and the reason is simple. You have accumulation of some lipophiesin, so it kind of masks that phenotype in the macular area. And so this is kind of the summary of the work where two sisters were found to be affected with really Stargard disease, but were also carriers for the mutation in GPR-143. The mother was just a carrier for a mutation in ABC-4, and the father was very mildly expressing the disease that should be the ocular albinism phenotype. So to summarize some of this what I said, so the optical gap phenotype that I mentioned is seen in quite a number of Stargard patients, and it is associated with the 1961 mutation. The mutation really doesn't matter what you have on the other allele. It can be a very severe mutation, can be very mild mutation. The phenotype you get is pretty much always the same. The disease is kind of arrested in this first few disease groups and has bullseye phenotype, no lipophiesin and so on. It is very difficult to make any correlations, except maybe with 1961, and we have now a few more examples with the disease severity. Yes, if you are a knockout, you are likely to develop early onset, very severe disease and have quick progression either to cone rod or the RP-like phenotypes. And a big problem with Stargard disease and clinical trials for Stargard disease is that disease is not very rapid, so doing clinical trials is pretty tough, so you really have to select a very specific group of patients where you can measure some quantifiable features because one of the features, I don't know if I have it here, probably not. So one of the features is of course the expansion of the geographic atrophy. In many patients such as 1961, it expands very, very slowly, so you don't see much difference in a year or two. And of course nobody wants to do or even will fund a clinical trials which go five years or longer. So you have to really classify the patients, you have to know the genotype and then you can select a group of patients that can reasonably be measured in terms of quantifiable differences within two years. So again mutations in ABC4 cause very many different phenotypes, variation in the gene and the genomic locus is really vast. I'll talk about that at lunch. We know more than 1,000 mutations, individual mutations that are causing the disease. The most frequent is in 20% but most of them are very, very rare. In individual cases. And another interesting thing is that we are working on now is that even with a comprehensive screening of the entire locus and this means the genomic locus is screened, the coding regions, non-coding and everything, you still have many cases that show only one mutation but they definitely look like they have ABC4 disease and we're trying to figure that out and I'll discuss that also a bit more during my genetics talk. So this is another issue here is that one in 20 people carry one pathogenic ABC4 mutation. The carrier frequency is really high. Not all combinations though cause the disease so because if you do the straight calculations you would see that the Saga disease would be really highly prevalent. It is much more prevalent that it has been stated, it has been stated like one in 10,000, my guess is at least two, three times more frequent because a lot of people as you saw are not clinically diagnosed as Stargard patients because they look very different. And the other problem is that right now the clinical trials that are ongoing say that the inclusion criterion is at least one mutation because of this high carrier frequency and wide heterogeneity in genetic and clinical presentation. This is really not a good criterion. You have to have two mutations in order to have a definitive diagnosis of the disease. Okay, so this is the quiz that usually comes with clinical presentations. So one row is ABC4 disease, the other row is not. So who are the retinal specialists here and tell me which is which? And this again shows how difficult it is clinically to diagnose the disease. Any guesses? So the bottom row is confirmed ABC4 disease with different mutations, severe disease, milder. And again, this is a homozygous 1961 patients. And these patients have mutations in RDS periphery. That usually the disease mutations cause in PRPH2 is called pattern dystrophy, but they are very often very much alike. So you really have to screen all genes. Actually the case in the middle, actually this case, is a person who works at Columbia and she was always asking me to figure out so what disease she has. We had screened ABC4, we didn't see anything. And only when we started doing the whole exome sequences so we quickly figured out the disease so that this is really a dominant pattern dystrophy what is clinically. This is mutations in CRX. Again, this looks very unusual for CRX disease. So it looks very much like the bullseye we have with the 1961 mutation in ABC4, but again this is another way the precision medicine is currently developing is that again only after you know the specific variants in specific genes you can call the disease. And what is also called phenotypic expansions we see because CRX mutations cause dominant early onset retinitis pigmentosa. So this is very unusual phenotype for CRX. It's a bullseye and as you see the mutation is actually a stop codon. So it's a severe mutation. So this is the CRX family we had. This is a pretty large family and so all affected except the grandmother and I always say that sometimes old ladies are really tough. She didn't have the disease. The all the others, kids, even grandkids who we were able to screen had the disease. We even got a cell paper out of this. Actually now we were on the cell paper and that was actually a very interesting paper out of Baylor where they were doing the mutagenesis in Drosophila and looking at machines in the phenotype in the flies to the human phenotypes and this was one of the genes that came up and so we had this specific family. With this work it was fantastic to learn that the method, I mean after they do the chemical mutagenesis, the method they look for the mutated flies is ERG. So actually they do ERGs on thousands of flies and when our ERG group is complaining that they can do four people per day I say well these guys did it on the flies and they did hundreds quickly. So but this is the example of the CRX. This again is just the same slides and I pointed out to this patient and this was the family that I mentioned at Columbia, so again the sisters that we saw and their mother who was at very advanced age where all had different clinical diagnoses. Yeah mother of course had AMD because she was well into her 90s and the sisters, yeah one of them was diagnosed with Stargard, the other with macular degeneration but it turned out that it is really not ABC4 not Stargard but it is pattern dystrophy and there is a deletion that causes this other variant is just a benign variant that is on the same allele on the same haplotype with the deletion. And I think this is pretty much all I have today so again a take home message. Not all ABC4 diseases, not all Stargards are caused by mutations in ABC4. Oftentimes they are very much look like Stargard but there are other genes that are involved and oftentimes even if they are confirmed to carry one allele they still do not have ABC4 disease. And the reverse as I showed is also true. We have found many cases where the phenotype does not look like ABC4 disease but is definitely caused by ABC4 mutations. The disease is extremely heterogeneous, the gene is very heterogeneous and you really have to use many different methods. Nowadays we pretty much resort to whole exome sequencing although we still do sequence ABC4 first if there is a good diagnosis of Stargard disease because it is a quick and these days quite a cheap way of analyzing and then if we are not finding two allelic mutations in ABC4 we find one or none and I'll talk about that also at lunchtime we put them through the whole exome sequencing and we do find the cause of the disease at a very high percentage. I would say close to 80% we find when we screen a case with retinal disease by whole exome sequencing. Again, very important is to collect all family and family members because doing analysis of whole exome sequencing on sporadic cases is very difficult unless you really hit something that is very well known. You really have to collect at least one parent if possible and siblings whether they're affected or not because then your power of the genetic analysis is really quite substantial and when I talk to other geneticists about other diseases, immunology diseases, their success rate is about 50, 60%. I think in eye diseases we really have research in quite a bit. In terms of retinal disease we know I think it's approaching 250 Gs that are causal in different retinal diseases. Sometimes simple eye diseases, sometimes they are syndromic diseases. I'll talk about that also at lunch. So this is what I always stress, although even these days when I lecture to residents and see what the geneticist is probably necessary, but I think in Utah we think differently but at Columbia we say, okay, let's get through with it. I say, well, this is the only way to do clinical research because you have to collaborate with geneticists in order to really precisely diagnose your patients. So that's it. Thank you very much for the pleasure. Fascinating, really appreciate your being here. I'll apologize that some of our retinal people had to stay for an Arvo meeting after the calories that I had, I know Paul. I'll meet with Paul tomorrow morning, yeah. So as you know, I mean, a lot of our approach diseases came from the German morphologists. And so essentially they're saying we're calling disease because it looks alike. And that's when we came up with these different categories. And then as we started getting genetic basis, the hope was, okay, we get the exact genetic and the phenotype will be very specific. So now it's getting confusing for clinicians because what we're finding out, now you can have the same genetic defect but you can have quite variable phenotypic presentation. So as you can imagine, and so we're kind of thinking in the middle of an area determining which way we are. I think we're still large use clinicians, morphologists, getting the genetics, and eventually it's gonna get way more personalized medicine where we're saying genetics is where the action is. And we have to realize that the phenotype may not be telling us much about the actual cause. Your thoughts about that? No, this is absolutely correct. And this is what I try to stress and I will talk about that more that on sometimes I title my talks like simple and complex ABC for disease. Yeah, because it is a Mendelian disease. So therefore it is simple because you have to have two mutations, you get the phenotype. However, the more we learn both with the new clinical methods, this advanced imaging when you can really image retina in a fantastic way and with genetics, with all the sequencing, so it gets again more complicated because it's not always, and ABC for is a good example because it's not simple. Yes, what I say is that for diagnosis of a Mendelian disease, sequencing is a must because if you have two mutations, you have the disease. At least you can tell the patient, let's say there is a clinical trial coming up gene therapy. So you have to know the gene in order to do that. And even here when I was talking about, let's say the therapeutic applications at Columbia right now, we have one guy who is feeding patients with deuteriated vitamin A. So the idea is that deuteriated vitamin A will make A2E at a much slower rate. So he's attacking that A2E part, that to eliminate the formation. Now as I showed, let's say 25%, at least maybe one-third of patients with ABCA4 disease do not accumulate lipofusin. So that doesn't make sense really to go there. Even if it's safe, although eating deuteriated vitamin A all your life is maybe another kind of thing you are not, although it has been shown to be safe and at least short term, but you know. So this is exactly the case. So we have to put much more work in the dogma or the premise of personalized medicine is great, but we are far away from it because we all learn about this great variety because we all have, even if you're closely related to somebody, hundreds of variants that are different from that person and what they do and how they interact is quite difficult to say. What I suggest always is to keep it simple because when we do whole exome sequencing, of course, you see a lot of stuff. So I say keep it to the eye. So if the patient comes in with Stargard, look for ABCA4 first, look for genes that are known and could be known and I'll show that again in my second talk. Otherwise, if you start looking at everything, you'll see too many things. And all these suggestions, the 23andMe and other companies that screen you for a couple hundred bucks and give you your life story is nonsense because you don't want that because you can often get a lot of wrong information and we know that, let's say in AMD that we do. You have these variants in factor H, in the other genes that are definitely linked but sometimes they cause it and sometimes they don't. So you have to be quite careful with that. So yes, it's kind of the advancement and in genetics, it's, I really call it the revolution, you know, the way we can sequence now because the last five years, the advancement is a blessing, but it's also a curse because things get even more complicated. But you really have to kind of work on both ends on clinical morphology and genetics to kind of make sense of all of that. And we see many, many more cases like I showed where you have two genes, indeed involved and you know, you see bits and pieces of each phenotype, but yeah, we're far from diagnosing it very well at this point. Well, the key answer is no, cause that all the longer answer is, I think. Yeah, yeah. We just now know it's not as simple as if you have, the big answer is, is there's merits to appear and it's zero to do anything and all those. I think that goes for hope. No, it is not the case. It's not the case because you have to know that because there is a lot of, you know, this incomplete penetrance cases and you see that a lot. Especially when you work on the families and what I always say here, I think it was the best place to work on the family. I mean, you see the mutation going through the pedigree but you know, some people have early very severe disease. Some people have practically nothing. So then we're surprised that you have a failure. Isn't it awesome? That does confuse the analysis, but. We're in that, that mix right now where we're kind of halfway between these areas and so we don't understand it, but it's up to that and say it's not going to happen. And then the Academy of Physicians, it's not important, you know, if you're going to be geneticist and very short-sighted about that, but it's going to be changing quite rapidly and obviously as this goes along. And we're now digging in more and more of these different diseases and discovering that a lot of this has been incredibly complex, but complex, I mean, it's impossible to cite this. I think this is the golden era, but this kind of thing is a great example of the thought we have. You get this specific genotype, you get this phenotype, there's a specific institution, what the morphologists always think is just not going to be case, it's just not going to be the answer to that. So there's, here's the whole thing. I mean, you know, Greg Hageman sitting here, that's the big one that drives him crazy is that we're getting treatment now for macular generation and we're just calling it morphologically. And then we're treating it, so let's say we're treating complex factor H. Well, we know there's lots of macular generation cases in the various areas, but they have no complex factor H, you have no knowledge. It's not going to do anything. And so then we're surprised that we have a failure associated with it. And we're in that mix right now, where we're kind of halfway between these areas and need to understand it. But, and then the academy came up with this position, it's not important, you know, that's where we know that genetics is. And I'm very sure it's cited, but it's going to be changing quite rapidly, obviously, as this moves along. And we're now digging into more and more of these different diseases and discovering that a lot of us is incredibly complex, but complex doesn't mean it's impossible to site this. And I think this is the golden era, but it's this kind of thing is, Randall gave a perfect example of the thought we have that you get this specific genotype and you're going to get this phenotype and it's going to be very specific, which is what the morphologists had always said, it's just not going to be the case, it's just not going to be that kind of thing. Yeah, in terms of treating of Stargardt, yeah, genetic screening, of course, I think that now we have, since I am at Columbia, in-house genetic screening, but there are many clear certified laboratories that screen right now and actually do whole exome sequencing straight for, and oftentimes it is covered by the insurance. Yeah, so that's another issue that is not always the case and it is not cheap because it's four or 5,000, I think, to do a comprehensive screening, but it is covered. So I would certainly do, in terms of how to treat Stargardt, yeah, that's a good question. I, as I said, we did the gene therapy, so this is putting the working gene in and theoretically for recessive diseases, this should be the cure, but you of course have to do it at the right time when you still have cells left there to where to put it in. Clinical trials are ongoing, but phase one, as you know, it's always hopeless cases, so they took people that, there is no cells to put it in, but it's safe, they showed. So we'll see, now I hear that it is not done at Columbia. Now I hear they have the, they allow to go into children or at least younger people, so we'll see. Chemical modification I mentioned, yeah, so some people are trying to target that thing, but I said this is not a very universal feature of the disease, although it's the most prominent feature, the A2E accumulation, it's still not in 100%. Then, very simple thing, what I always tell patients is that, avoid light as much as possible because the more visual cycle works, so the more disease you get, there have been some extreme suggestions, let's say the eye patches, and you know, you kind of close one eye, so you keep it, you know, not damaged. And then, before it was known that or thought, and actually this was a paper published by Gabe Travis and group that in the dark, the mice, the mouse model of Stargard did not accumulate lipophysic, that was wrong, it actually does. So now we know that the darkness is really not a cure, I mean it's a difficult suggestion anyway, but I usually say that, you know, try to limit the exposure to light, and the other thing is obvious, vitamin A, just avoid vitamin A as much as possible. And this is again a very controversial statement, because vitamin A is good for vision, everybody says you have to have it. In some diseases it is true that you have, here the problem is that it is not removed fast enough, so the thing is it accumulates. And that is the cause of the disease, so as little as possible vitamin A, so you shouldn't go too crazy, but if you avoid vitamin A in diet, that's enough. However, I've seen many, many cases, and people have asked me that oh, my eye doctor prescribed me 15,000 units per day, and I said yeah, you'll go blind fast, so if you don't stop it now. And then they sometimes argue that, especially the doctors, that Alikmetz is a geneticist, he's not a doctor, he doesn't know anything. But this is one thing that you can. Oh, I'm mixing it with RP. Yeah, so that is, the RP, vitamin A is good for some forms of RP, but that was yeah, when this came out, I even wrote a letter to, when any eye put out that thing that you know, eat vitamin A, I put out, I sent a letter to Seaving, and you know, I said guys don't, they did not publish it. I think they kind of modified a little bit of their statement, but yeah, this is the old Elliott Burson thing, yeah, that you eat vitamin A and cure it. Very controversial. It is, and here we know functionally why it's bad, so actually you shouldn't do it, you shouldn't do it and you should avoid it rather than take any kind of a supplement. So there is a, but many diseases, yeah, it is a, we had a case that I will discuss today where you have mutation in another gene, in the visual cycle there, Chris Balczewski and other smart guys said eat vitamin A, and so there you should do it, because that is a, it depends really where the defect is. Great. So random, the patches of the RP, or people's cells that you just lost, no one, and then we're back. Actually, right now Janet says that they are photoreceptors and not even RP. Well, it is, yeah, that's a good question because now this is the paper, no this is not the paper I'm talking about, this is just a method paper, but we recently came out, Janet I think is the first author there. So there is a bit of a discussion here and quite a lot of animated discussion I would say, because always the theory, and Janet Sparrow was the one to say, we'll see you later, was that the cause of the, although the defect is in photoreceptor, but what happens is that retinoids, Ultrans 116 accumulate, form A to E, photoreceptor is a phagocytosed and then in the RP, the lipophucin accumulates, photo activates, lysis cells, RP dies first, photoreceptor go after. Right now, people are stating that actually the death of photoreceptor in some cases occurs first and this is true with this optical gap that I showed with 1961 mutation and this method allows then kind of disassociate, yeah, RP from the photoreceptor and so what Janet is saying, for example here, so if you overlay these two, these are dead RPE cells, so the back, but there is heavy fleck on that same spot. But she's saying that, yes, there were a few things, some said that there are RPE cells that are, you know, engorged of lipophucin, then there was a discussion that on the edge here, you have them kind of overlaying and that's why it's more intense. Now, the statement is that this is really photoreceptors that give those fleck patterns. Why it is such a structure, it's difficult to say. Well it's so typical, the catchy things that are about three or four or microns or anything. Because now here again, yeah, this is this discussion that I'm pushing the wrong button that, you know, you have lipophucin fleck after the RPE is dead. You don't see it before. So I think it was a year or a few years later was this image done from the same patient. So that's kind of one of these arguments that what actually you see flecks, it's not RPE as much as it is photoreceptors full of lipophucin. Okay.