Progress! The State of Usher Syndrome Research After the International Symposium on Usher Syndrome
July 18, 2014
by Mark Dunning
Though he desperately wanted to be there, Bill Kimberling could not attend the International Symposium on Usher Syndrome. Personally, I couldn’t get through one presentation without thinking about Bill. Seems like I wasn’t alone as his name and likeness were on a lot of presentations. His fingerprints were all over this conference and he would have loved to have learned about all the progress that has been made. I’ll do my best to sum it up here, but Bill would do a much better job than I will. It’s just one of the ways we missed him at the conference.
Dr. Bill Kimberling
Bill Kimberling once defined seven steps to finding treatments for a disease:
- Find the disease gene(s)
- Correlate genotypes with phenotype
- Find or develop animal models
- Elucidate the disease mechanism
- Find or develop an effective treatment in the animal model
- Screen the human population to identify people who might benefit
- Test the treatment in these people
I am happy to report that we have made a lot of progress on most of these fronts. Much of what was reported at the conference is new research that has not been published and so has not yet been subject to peer review. Therefore I can only write in generalities about what was presented. All you need to know is that, generally speaking, we’re moving by leaps and bounds and just about everyone left the conference filled with hope and enthusiasm.
I’ll take each of Bill’s seven steps and try to quantify how well we are doing on each of them. In fact, I’ll take a guess at just how happy I think Bill would be about the progress as the measure. We’ll call this the Kimberling Scale. We’ll go 1-5 where one is “Bill would be sad” and 5 is “Bill would be doing cartwheels”. Oh, because I know he eagerly followed the proceedings, don’t be surprised to find Bill in the comments disagreeing with me.
Find the disease gene (Kimberling Scale: 4.5)
Well, actually, we’ve found twelve of them. We’ve also come a long way on finding the specific variants and determining their pathology. In other words, we’re much closer to knowing which specific gene mutations cause Usher syndrome. The sense I got was that almost all the major causes have been identified. The number 74% was thrown out at one point, which I took to mean that we catch nearly three quarters of the Usher mutations these days and more are being found every day. In five years, that number will probably be up closer to 90%. That means just about everybody tested would have a known cause of their Usher syndrome very soon. I imagine Bill would be thrilled by this progress. He found the first Usher gene, after all.
Correlate genotypes with phenotypes (Kimberling Scale: 3)
To review: Genotype is the genetic cause of the disease. For instance, Bella has a mutation in the Myosin 7a gene. Phenotype is how the disease presents itself. In Bella’s case she was born profoundly deaf, has vestibular issues, was night blind before she was a teenager, etc.
We’re doing much better on understanding the relationship between the genotype and the phenotype. Unfortunately we’ve gone from “we have no idea but we have a hypothesis” to “we’ve tested our hypothesis and it’s wrong”. This isn’t a bad thing. Hang on. I’ll explain.
The hypothesis was that specific mutations would behave the same. We used to group people by Usher type and assumed that everyone with the Usher 1b gene, for instance, would a) have the same phenotype and b) that phenotype would match the traditional grouping of Usher types. In other words, people with the same mutation would all have the same hearing loss, the same rate of vision loss, the same balance issues, and the same olfactory issues (Ahh! Didn’t know about that, did you?) Yes, it turns out that the olfactory cells may also be impacted by Usher. Again, more research is needed but interesting just the same. Anyway, to get back to the point. The research shows the hypothesis is wrong. This is good news and bad news.
The good news: The rate of vision loss is all over the place for people with the same genotype. So the whole “you are definitely going to go blind” thing is probably wrong. You are definitely going to have vision problems. The degree to which they occur will differ from person to person, regardless of the genotype. Hopefully we can educate physicians about this so they won’t authoritatively tell people what’s coming now that we realize it’s more complicated to predict. We can give a more refined, qualified prognosis, which, especially in light of other, parallel research advances, leaves a lot more room for hope.
The bad news: We still have no idea about the normal progression of the disease. So the big question all families want to know, “how fast will my vision loss progress?”, still doesn’t have an answer. There was also a lot of data showing that the hearing loss in Usher 2 might be more progressive than we thought. Again, this still needs to be vetted. People lose hearing for a lot of reasons (Hello iPod). But if you use hearing aids, you probably want to get your hearing checked regularly to make sure it doesn’t change.
Overall, though, I think Bill would be pleased but not satisfied by the amount of progress on this front. Four years ago we had a hypothesis but it been barely been tested. Now we’re making progress. Just because it’s not what we expected doesn’t mean it’s not progress. But it’s still too slow. We need much better natural history information and we need it quickly.
Find and/or Develop Animal Models (Kimberling Scale: 2)
Again, good news and bad news here. The good news is that there are more mouse models than ever (some even showing the vision phenotype) and Jennifer Phillips is pumping out mutant Usher fish like some sort of superhero villain. The bad news is that we still lack good models for translational research. The mice don’t demonstrate the vision phenotype well enough or early enough to be good models (or, maybe, we haven’t yet figured out how to measure it well enough). Fish are very different from people. We desperately need an animal model that correlates better to humans. More on this in a moment but I know four years ago Bill felt this was an urgent need. We haven’t made enough progress on this. He would be worried.
Elucidate the Disease Mechanism (Kimberling Scale: 4)
Here is all I can say about this: there were some very passionate molecular biologists at the conference. I didn’t understand what they were talking about most of the time, but I know beyond a shadow of a doubt that they will not stop until they fully understand the Usher protein interactome. I also know that they have made great progress. We are really close to understanding the jobs of the Usher proteins in both the eye and the ear. I would say that more progress has been made on this front than any other. Bill would have loved this stuff if he had seen it and I believe he would have been very pleased with the progress.
Find or develop an effective treatment in the animal model (Kimberling Scale: 3)
Again, good news all around on this front but with a big caveat which I will get to in a moment. There are drug therapies, gene therapies, stem cell therapies, and prosthetics for the eye, the ear, and the vestibular that are being tested in animals RIGHT NOW. That’s right, I used capital letters. Many of these look promising and more are coming. If we were mice, we’d be psyched because treatments are already showing success in mice. Unfortunately, we are not mice and that’s the problem I mentioned earlier. We don’t have a good model to get us from the animals to people.
Just stop and think about this for a second. We don’t yet have mouse models with a testable retina phenotype for all types of Usher (though there was progress on this demonstrated at the conference). So what if I could only tell you that a treatment was tested in a mouse and it helped its vestibular, would you be willing to have that same treatment tested in your eye? Can the hearing or the vestibular function of a mouse correlate to the human eye? And would you, as a possible clinical trial candidate, feel confident having such a treatment? That’s not enough for most people. Now how about the eyes of a monkey? What if a monkey were given the treatment and it rescued that monkey’s vision? And it did so over a period of years without side effect? Well that’s different, right? That’s where we are. There are treatments that help mice but the impact on the vision is hard to measure. We don’t have other models more applicable to humans. We need better models.
There is something very cool that is in the early stages of development that might mitigate the need for a better animal model. Jennifer has written about this in the past. Stem cells, your stem cells, can be used to create retina cells. We could soon be able to recreate your retina in a petri dish and test treatments on it. And when I say yours I mean yours. It would be your specific DNA. So how would you feel about that? Your eye, treated in a dish, not in your body. Would that make you feel confident enough to join a clinical trial? What if it was just a few retinal cells and not the whole eye? What if that were combined with safety testing in an animal model (ie, just making sure it doesn’t have obvious side effects rather than testing how well the treatment works in an animal)?
My guess is Bill’s reaction to this would be both eagerness and apprehension. We are THIS close to finding viable treatments but they still feel a long way from getting out of the lab and to you because of the lack of good models. Translational research models needs to be one of our focal points coming out of this conference.
Screen the human population to identify people who might benefit and Test the treatment in these people (Kimberling Scale: 1)
I’m going to put these last two steps together. Remember, the bridge from treatment in a mouse to treatment in the clinic is not just viable animal models but clinical trials in humans. We test the treatments in humans. The treatments appear to work. The FDA approves and ophthalmologists around the world can start dispensing those treatments. It takes a lot of humans to find the right candidates for clinical trials. Not everyone wants to be involved in trials. Not everyone lives near the testing center. Not everyone meets the right criteria. You need a large pool to make it work.
Here’s the good news. Conservatively there are around 400,000 people in the world with Usher syndrome. That’s conservatively. That’s more than enough to find viable candidates for clinical trials.
Now here’s the bad news. The Usher Syndrome Registry contains around 600 families. 600. There are other registries around the world. The largest I know of contains around 2,000 families and that is a combination of a number of other registries. Combined throughout the world, I would guess we maintain good connections with 3,000 families maximum. That’s for all types of Usher. And only a percentage of those families have been genotyped. In other words, we don’t know what type of Usher they have and some of them may turn out not to have Usher at all. We probably have genotyped, globally, less than 1,000 families. That is simply not enough and it is the greatest risk we have to bringing treatments to families. The greatest fear expressed by researchers at this conference was not that we would fail to find treatments. They are confident that they will. It’s that they would fail to get them through clinical trials because we are not in touch with enough Usher families.
Bill would consider this a tragedy. He has dedicated his life to finding Usher families, to meeting Usher families, and to helping Usher families. He knew from the start that was the key to finding treatments. To fail because we haven’t done enough on that front would be tragic.
BUT (those are happy capitals), this is a resolvable issue. We need to get everyone with Usher syndrome to register in the Usher Syndrome Registry. Then we need to genetically test them all. It is a simple problem and a simple solution but it will be a lot of work. And it is work that we will all need to do.
The number one priority for the Usher Syndrome Coalition coming out of the meeting is to resolve this problem. We need to find everyone with Usher syndrome and we need to find them now. This will take the education of doctors, support organizations, and Usher families. And that education will need to come FROM Usher families and researchers with the support of the Usher Syndrome Coalition. That’s right, I’m looking at you. We are going to develop the educational materials. You are going to deliver them (with our help, of course). And together we are going to find everyone with Usher syndrome in the world.
Then we are going to work with the labs around the world and the governments around the world to get all of those people genetically tested. While this is going on, the research will continue with a focus on developing those models that will get us from the lab to the clinic. We are still years away from clinical treatments for Usher syndrome, but we know how to get there. And I know that, more than anything else, would make Bill smile.