Genetic and clinical variability of the Stargardt disease gene ABCA4, and the implications for treatment

blue star

A huge thank you to our Professor Rando Allikmets who recently joined our Zoom session to talk to us about genetic and clinical variability of the Stargardt disease gene ABCA4, and the implications for treatment.

What is your view on drug therapy - specifically ALK001 and Emixustat?

Both of these approaches try to reduce the accumulation of lipofuscin (and its major component, A2E, which is an aggregate of two vitamin A molecules), in two different ways. Both approaches are scientifically sound, the question is if these would work and how efficient would they be.

ALK001 tries to replace all vitamin A in a patient with a modified form of vitamin A, which forms A2E ~5X less efficiently. The approach suggests a patient to take an oral pill of the modified vitamin A daily. The approach worked well in mice (especially if started very early), but it would be much harder to have an effect in humans due to starting much later in life and the disease process and much more complicated dietary control.

Emixustat is a compound which slows down the most important protein in the visual cycle, RPE65, so the visual cycle will work slower and the amount of accumulated lipofuscin will be lower. Again, it will have some effect, but here the main issue is the proper dosing. If the dose is too high, the drug will cause, not prevent, vision loss. As a reminder, mutations in RPE65 cause early onset retinitis pigmentosa, a much more severe disease than Stargardt. So again, the approach is scientifically sound (I tried myself similar approach at Columbia about 15 years ago) but it has to be applied with great caution. I assume this has been addressed since the drug is already in Phase III clinical trial.

So, both of these approaches are valid, but the effect on the disease has to be seen. Both of these are also “indefinite” approaches, one has to take the drug for the rest of their lives, as opposed to gene therapy (if that would be available), which is usually a one-time treatment. As a reminder, there is approved gene therapy available for the aforementioned RPE65 gene and it seems to be working.

Is it important to know what the 'mistakes' are on your gene?

Yes, then you know exactly what disease you have and can seek specific treatment. This is obviously obligatory for any gene therapy approach.

How does this affect the possibility of our children being affected?

In recessive diseases such as Stargardt/ABCA4 disease, if healthy parents have an affected child, they both carry one ABCA4 mutation. In this case the chances to have an affected child (who gets one mutation from each parent) are 25%. This does not mean that if you have four kids only one will be affected. The 25% chance applies to every child.

If one parent has Stargardt, the child inherits one of the two mutations from that parent. Now the question is – does the other parent have one mutation (i.e., is a carrier) or none? The chances of being a carrier are 1:20, that is 5%, which is a very high fraction for a rare disease. If the other parent has no ABCA4 mutations, then kids will be carriers, but healthy. If the other parent is a carrier, the odds of having an affected kid are 50%.

How can one plan a family ensuring that the disease doesn’t pass on to the children?

If both parents are healthy but a child has Stargardt's, then parents should be tested for mutations found in the child since they are both carriers. If one parent has Stargardt's, then the other should be tested to find out if he/she is a carrier. There is no 100% test available since many mutations are still not known, but this test would be at least 90% accurate. Then, in both cases, healthy embryos can be detected by prenatal testing and used for pregnancy.

Concerning the complexity of ABCA4 gene, recently I've read a publication showing the discovery in intron mutations related to Stargardt’s. We also know that there are many (thousands) of mutations related to Stargardt’s, and that some patients have more than two mutations.

Yes, on one of my slides I gave the approximate fractions of mutation types. About 25% of ABCA4 mutations are introns, either in splice sites, or deep introns. We currently know of about 1500 mutations, but the number is growing. And yes, about 10% of patients have three mutations, meaning that two thirds are on the same chromosome. We call these “complex alleles”, but genetically speaking complex allele still counts as one mutation. 

And about the gene therapies using oligonucleotides (RNA) like the one that have been tested for ProQR, how does that RNA work for Stargardt’s?

Yes, these therapies are also being developed for ABCA4. And you are correct, this approach is complicated due to the number of mutations. It is likely that similar therapies will be first tried on the most frequent ABCA4 mutations.

What does science really know about the relation between the mutations and the progression of the disease? Which mutations are severe, etc? Do we effectively know that?

Yes, I described this in my talk. We know a lot about the severity of mutations and the disease prognosis associated with different classes of mutations (deleterious, severe, average, mild, very mild, etc.).

What is the benefit of being genetically tested? Does gene therapy improve or just stabilise the condition?

Genetic testing gives you the correct diagnosis. Clinical diagnosis can be wrong, I discussed it in my talk. Many other eye diseases look clinically like Stargardt.

We do not know before we do the trials. It should certainly stabilise, but may also improve. Patients treated by the only currently available gene therapy treatment for the RPE65 disease saw some, or even good, improvement. Every disease is different, so hard to predict.  

Is it at all possible to treat Stargardt’s disease? Is there any difference in treatments for patients from India / Asia

Yes, I think many treatment options will work and are being tried. There is no difference for people of any ethnicity.

I lost most of my central vision at age 29 during pregnancy (of triplets). Now, at 47 I am losing small dots in my peripheral vision (maybe hormonal / menopausal). Is this common? Should patients with a clinical diagnosis of Cone Rod and 2 ABCA4 mutations be considered as Stargardt patients?

There is no evidence that hormonal imbalance causes any complications in Stargardt disease. You are probably experiencing disease progression. I am not an MD, so cannot give very specific clinical advice, but it is known (and I described it in my presentation) that a fraction of Stargardt patients will also experience loss of peripheral vision. This is especially true in late stages of cone-rod dystrophy, which is an advanced stage of the ABCA4 disease. The important part is genetic testing, if a patient has two mutations in ABCA4, then they know the cause of the disease, whatever the stage is called.   

How do we explain or understand that multiple different genes with very different functions, can cause diseases that all look the same clinically?

The detailed answer would take too long and often it is not known. It is usually the case when genes / proteins are in the same location or in the same biological pathway while performing very different functions. Genetic testing provides the exact answer, clinical testing is often too descriptive to provide a definitive answer.

I have three questions: 1) My cousins (brother and sister) are married to non-relatives. One of the spouses is even from a different country. Each of my cousins has one daughter who were diagnosed with Stargardt’s at around 15 and 18 years of age. I know that the gene has to come from both the mother and the father. Is that possible that both my cousins had the gene and also their spouses had it too? 2) Are they good candidate for gene therapy at this age? Or does it have to start at an early age? 3) What is your opinion with regard to acupuncture therapy? They tried it in Turkey and they were told if it is going to help then at least it will prevent deterioration.
  1. This is the only possibility, your cousins were carriers of the same mutation and their spouses were also carriers of likely different ones. The carrier frequency of AbCA4 mutations is much higher than appreciated, about 1:20 people carry a possibly pathogenic ABCA4 mutation. In this case they seem to have a really bad luck, which is impossible to predict. If a person is healthy they do not get screened for a disease…
  2. I see no reason why they should not be good candidates. However, clinical trials are rarely allowed in children, so earlier than 18 would likely be impossible now. Once the treatment is approved, this may change.
  3. The acupuncture therapy is a hoax. I know about this one in Turkey and there are many other similar hoaxes out there. These usually push “stem cell treatment”. These places should be closed and “providers” arrested since they prey on patients who will grab at any opportunity. They just want the money — the treatments have absolutely no scientific basis.
If a child is diagnosed in their teens does that mean their progression will be more rapid?

Often yes, but not necessarily always. It depends on specific mutations.

How much collaboration or communication is there between universities and organisations doing research?

There is a lot. All research these days is collaborative. Unfortunately, there is also a lot of money-driven competition in science today, which often hampers the progress instead of promoting it. Science today is, unfortunately, very different from when I started almost 40 years ago. 

What institution was conducting the gene therapy trial which used the Oxford Biomedica / Sanofi trial?

It was done at several places. In Europe it was in Paris. The main centre was in the Oregon Health and Science University in Portland, Oregon, US.