Open Letter to Owners and Breeders
from Natasha Olby Vet MB, PhD
Dr. Bell And Dr. Olby's open letters first appeared on the official STCA website. For additional information on Cerebellar Abiotrophy in Scotties, see the STCA website:
Dear Scottish Terrier owners and breeders
As you all know, with the help and support of the STCA health committee, Dr Bell and myself have been working
on collecting blood samples for DNA extraction and storage from dogs diagnosed with cerebellar abiotrophy, and from all of their relatives. It has come to our notice that there is significant confusion over various aspects of this project and we would like to take the opportunity to address some of the questions we have been asked and to give an update on progress so far.
The research grant is entitled, "The Genetics of Canine Cerebellar Degeneration." It was approved and granted
by the AKC Canine Health Foundation, and the STCA Health Trust Fund has contributed to it. The research project involves four breeds; American Staffordshire Terrier, Gordon Setter, Old English Sheepdog, and Scottish Terrier. The two-year grant is to do genome screening in the American Staffordshire Terriers and Old English Sheepdogs (where full family blood samples have already been collected), and to undergo full family blood collection in the Gordon Setter and Scottish Terrier. These breeds have similar autosomal recessive inherited cerebellar disorders. If we find a mutation in one of the former breeds, then we will immediately check the Scottish Terrier and Gordon Setter breeds for the same mutation. If the causative gene is not found in the Scottish Terrier (and we would consider ourselves to be extraordinarily lucky if we found the gene in Scotties in this particular 2 year grant), and full family blood samples are collected and stored, we will apply for funding of a separate research project for genome screening in the breed.
We have been asked several questions but by far the most common is how certain are we that dogs actually
have the disease with which they have been diagnosed? For obvious reasons, people want to be sure that the diagnosis is accurate. From my point of view, this question is central to my research because if you start with a misdiagnosed sample when performing a genome screen, you will be lead away from the defective gene, and
the entire research project may be doomed. Because of this, we have to take into account how the diagnosis
has been reached when we analyze our results, and this can be done statistically without too much problem.
For clarity, we allocate dogs to different groups depending on the diagnostic workup:
1. Dogs in group 1 are the gold standard: clinical diagnosis has been confirmed by necropsy. We are absolutely certain that these dogs had CA. We currently have 6 dogs in this group, but only 3 with blood samples submitted. Painful as it can be, I would encourage all owners of affected dogs to consider having a necropsy done when they die to confirm the diagnosis. The more dogs in this group, the better we can delineate the disease and the more sure we are of our data.
that include showing consistent clinical signs, and a consistent history of slowly progressive disease. They have worked extremely hard a) to review histories and diagnostic video tapes of all dogs to determine that the clinical signs are consistent with CA, and then b) to keep in touch with the owners to ensure that the progression of the disease is consistent with CA. We are collecting samples from all dogs that have been diagnosed clinically in this way. Recent postings on a chat list have suggested that I will not be using these samples: this is not true. I will be using samples from clinically diagnosed dogs, but the way these animals have been diagnosed will be taken into account statistically when performing our analysis. As you can see from the answer to the second question below, we have great confidence in the clinical diagnosis.
clinically is 100%. This may change over time but Drs Bell and de Lahunta have been extremely conservative in labeling dogs as affected and I am very comfortable with this approach, particularly as we definitively confirm disease in more dogs.
The final point we really wanted to address is to reiterate the scope of this project. Performing a genome screen and linkage analysis with the aim of identifying a mutation is a long, complicated and expensive process, and only has a chance of success if the correct samples are available. Indeed, further funding can only be obtained if we can already demonstrate that we have samples from affected and related dogs. All funding agencies will require that we do a statistical simulation of the screening and linkage process with the samples we have collected to show that statistically we have a chance of finding the mutation before they will consider supporting such work. The current project is therefore the very first stage of this whole process: sample collection and correlation with pedigrees. By the end of 2006 we hope to be able to show statistically that we have a reasonable chance of finding the mutation with the samples we have collected. If we cannot collect the samples we will not be able to complete the actual genome screen and linkage analysis to identify a mutation.
We would also like to update you on how things have been going with the sample collection. Dr Bell is an incredible liaison, and as a result of his efforts, several Scottie owners and breeders have done an outstanding job for us in rounding up families of dogs and we have got much further than I anticipated. If we continue to get this type of support, we are hopeful that we will be able to apply for funding to do the actual genome screen some time in the next year. Remember that it often takes 6 to 12 months for funding to come through. Our only chance of being successful in receiving this type of funding is to demonstrate statistically that we can do the work because we have the appropriate DNA samples in hand.
It is not possible today to predict how long it will be until there is an actual DNA test for CA. At the end of this grant cycle and when a new grant is approved, we will be able to speak to that issue more accurately.
This work is being supported by the Canine Health Foundation of the American Kennel Club, and the STCA health committee. Dog owners have also taken on some of the cost by paying for the blood sampling and shipping, and the cost of salaries for people working on the project has been covered by North Carolina State University. Linda Orsborn of the STCA health committee has been invaluable both with respect to her knowledge of the breed and her ability to articulate the problems we have to overcome to get this work done. The health committee clearly has experience in addressing the genetic problems within the breed through their work on identifying the mutation for von Willebrands disease. I have been very appreciative of their practical approach and their understanding that we have to do this one step at a time. I would encourage you all to contact Linda with your questions on this research. If we act through a single liaison, there is less likely to be confusion. It has been and continues to be a pleasure to work with you all as we try to tackle this problem. Dr. Bell and myself hope that this addresses any points of confusion about the sample collection project.
Natasha Olby Vet MB, PhD
Diplomate ACVIM (Neurology)
Associate Professor of Neurology/Neurosurgery
North Carolina State University
how accurate is the clinical diagnosis as made by Drs Bell
and de Lahunta?
We already have some great data:
Every dog diagnosed clinically that has gone on to have
a necropsy has been confirmed to have the disease.
So with 6 dogs necropsied, the accuracy of diagnosis
The next point I want to make is that it is extremely important that a clinical diagnosis be taken seriously. We
must collect samples from all dogs clinically diagnosed now while we can. When doing genome screens to look
for defective genes, the number of samples available, and the familial relationship of those samples provides the power to be able to identify a chromosomal region linked to the disease. It is very difficult to collect samples from parents and grandparents of affected dogs, in addition to unaffected siblings, but that is the type of family that provides the most statistical power. It is likely that we will use samples from all four categories listed above to provide adequate power. As we definitively confirm more and more dogs by necropsy or full workup, the number
of dogs in group 3 increases (i.e. dogs with relatives with confirmed disease). Thus, dogs move from group 4 into group 3 over time. I have worked with the American Staffordshire Terriers for six years now, and we have established the familial basis so well, almost every dog referred to me is directly related to a dog with confirmed disease, even when they come from as far afield as Brazil and Poland.
The second question that is important to address is
2. Dogs in group 2 show consistent clinical signs and history, and have received a full diagnostic workup of MRI or CT of the brain with CSF analysis in addition to the clinical diagnosis Five clinically diagnosed Scottish Terriers have undergone MRI, and three of them had a CSF tap. There is blood collected on 2 of these dogs.
3. Dogs in group 3 show clinical signs and history compatible with the disease (and thus have been clinically diagnosed) but in addition they are direct relatives of dogs in groups 1 and 2. We are therefore sure that they are also affected.
4. Dogs in Group 4 have been diagnosed on their history and clinical signs alone. Dr. Bell and Dr. de Lahunta have developed a set of diagnostic clinical criteria