Originally Posted By: DM Message Board
All breeds can get a degenerative condition of the spine, which is both chronic and progressive, called Degenerative Myelopathy. However, the Degenerative Myelopathy of other breeds is not the same disease German Shepherd Degenerative Myelopathy. German Shepherd Dog Myelopathy (GSDM) is unique.
DR Coates has worked under the theory that DM is ALS. Amyotrophic Lateral Sclerosis and related diseases are motor unit diseases where the nerve cells in the body responsible for controlling movement die off leaving the patient weak and with varying degrees of Lower Motor Neuron dysfunction (loss of reflexes and flaccidity) or Upper Motor Neuron dysfunction (hyperactive reflexes and spasticity). Those causing LMN disease affect the EMG early in the course of the disease. Those causing UMN disease result is selective shrinkage of the motor cortex visible on MRI. Neither of these conditions exist in GSDM.
Immune diseases like MS attack varying parts of the nervous system and one of them, Primary Progressive MS, specifically targets the myelin and axons of the spinal cord leading to UMN signs but without affecting the cell bodies of the neurons (which is what is seen in GSDM on histopathology). The CSF protein is usually normal in ALS, but abnormal in MS. Oligoclonal bands of IgG are common in MS and uncommon in ALS. The recessive forms of ALS are extremely slow in development and do not result in shortened life-span. Even the one motor unit disease known in dogs, Spinal Muscle Atrophy in Brittany Spaniels occurs in young dogs with progressive EMG changes leading to death. That might be more consistent with the “early onset DM reported in the GSD which is not the same disease as GSDM on histopathology.
ALS diseases cause motor problems but not sensory ones. That is they do not cause CP deficits or hypermetria (ataxia in which movements overreach the intended goal.). People do not knuckle and scrap their toes when they walk, they only show weakness. Most of them are painful because of muscle spasms. (Does that sound like GSDM?....NO!)
So, even if there is a genetic change in SOD1, Dr Clemmons believes that a change must also be explainable based upon the clinical signs. If not, it may just be a casual relationship not a causal one. GSDM as a pure motor unit disease just does not fit all of the available data. (Not just DR Clemmons…everyone's!)
To simplify- lay person's explanation:
DM Corgis, Boxers, : motor unit disease
DM GSD: Auto-immune disease
DM Corgis, Boxers : Protein is normal in the AO CSF
DM GSD: Protein is normal in the AO CSF but Protein is elevated in the Lumbar CSF
DM Corgis, Boxers: Oligoclonal bands of IgG are uncommon
DM GSDS: Oligoclonal bands of IgG are common in MS
DM Corgis, Boxers: affects cell bodies of neurons
DM GSDS: Does not affect cell bodies of neurons
DM Corgis. Boxers: muscle spams
DM GSDS: no muscle spasms
DM Corgis, Boxers:EMG is affected early in the disease
DM GSDS: EMG is normal
From DR Clemmons:
“They are [similar, the DM Flash test and the gene test]. The main difference is that without the resources of MIT behind me, I had to look to see if there was a genetic relationship to GSDM and using what we had that was reasonable, we use RAPD (random access repeat primers) to look at the genome. Using these, we found a pattern that was consistent in GSDM (and other forms of DM).
While we initially looked at a specific set of primers that we looking at the IRB1 region and found a change there that was consistent in many GSDM patients, it was not as consistent in all of them as the change in the RAPD analysis. Although we did not continue to call the test in other breeds the DM Flash test, we continued to call the test in the GSDs that in honor of Jack Flash. We also have seen several other changes that exist in GSD patients that are consistent, but we track the DM Flash test to confirm the diagnosis.
Of course, since demonstrated the potential to find the gene associated with GSDM (and other DM cases) and that developed the interest in other researchers who have now found the change in the SOD1 region of the canine genome. They were able to use the 18000 SNiP array (we could not afford them since they were $2400 each), but MIT had developed these when they did the canine genome project.
The major difference between SNiPs and RAPD is that SNiPs let you put a gene name to the product and the RAPD does not (as easily). RAPD also may find something that SNiPs do not and vice serse, but it does appear they have found the same thing here (or there are more than one change which is still possible). We took a different tact in using the RAPD, which was to determine the incidence of the gene change so we determine it effectiveness as a diagnostic.
As such we found that 25% of the dogs care the trait we track, but only 10% of these ever develop GSDM. So, it makes a poor screening test. Yes, 91% of the reason for developing GSDM is due to the genetics and patients who are positive in the DM Flash test have a 12 fold increase risk of developing GSDM, it cannot be the soul factor as to why they eventually develop GSDM.
On the other hand, if an older GSD has clinical signs, it is a great diagnostic test with a 96% sensitivity and 99% specificity. It correctly identifies those dogs who have GSDM from those (proportionately more) who do not have GSDM once clinical signs develop. Probably the gene test does that, but we do not know for sure, since they do not have that information about their test yet or at least have not presented it. The gene test might do one thing that we currently do not (although future versions might) in that it may be able to tell who is a carrier of the risk factor. We have only tracked those who are at risk and are homozygous for the trait.
However, since the gene is not the soul factor for developing DM, I am not sure that genetic testing and elimination of carriers and at risk patients is appropriate. Based upon our work, that could mean removing 75% of GSDs. That is foolish.
Dr. Coates has stated that since the gene involved in their work is the SOD1 gene, that DM is a motor unit disease and an animal model of ALS. In that, I still believe she is wrong. We have never been able to demonstrate selective motor unit involvement even in late cases. ALS does not have sensory abnormalities which would include proprioceptive deficits which are a predominate feature in DM. We do not think you can build a puzzle by starting with the last piece; the puzzle is built from all the pieces. Only when all the pieces fit together, can the whole picture be seen.
Too much data says that GSDM is an immune mediate chronic neurodegenerative disease associated with demyelination and axonal loss. This is most similar to PPMS rather than ALS. The fact that the SOD1 is involved complicates things because it is means that DM may be a “ child” where the change in SOD1 triggers the immune disease and attack of the nervous system and so it looks like a cross between ALS and PPMS. As such it is not a pure model of either. However, the truth is the important thing. With that, we can devise appropriate strategies to combat it.
Dr. Coates has made a few mistakes in her work on GSDM which is a bit surprising, but they are real.
One study was nullified since the case selection led to only 3 dogs in the study of 12 that actually had GSDM. On the other hand, this confirmed our work and that clinically only 25% of the patients who present with signs of posterior paresis actually have DM.
She suggested that all dogs have oligoclonal bands and, therefore, they were not important in DM. This is not true and we found that 60% of GSDM patients do have more than one unique clonal band in lumbar CSF which is the definition of oligoclonal band positivity.
Finally, she keeps looking for CSF changes in DM in the AO CSF which we have repeatedly shown to be normal. CSF changes in DM occur in the lumbar CSF and if there are changes in the AO sample, there is something other than DM.
Her work provides pieces of the puzzle, but only if the rest of the puzzle is considered. We have tried to provide those pieces. For years, I was the lone voice and criticized by many. Now, I have a chance to do the same, which is refreshing.
Sadly, they have probably effectively dried up our sources of funding which may make it very difficult to continue our program to help GSDM. I still think that gene imprinting at puberty is what establishes the true risk of developing GSDM. While the SOD1 gene (or the RAPD based DM Flash test) may be important, they are not the sole determinate and to not continue to pursue the other factors would be very sad.
In summary, I embrace the MIT findings, since they help us explain the disease. It has provided us with perhaps a way to confirm what we found with the RAPD based DM Flash test and from whence that change is coming. We understand that we cannot compete genetically with MIT, but there is room for us both.”
If you only look at the gene test and do not advance beyond that, you are doing a disservice to the dogs. First, if the SOD1 association is confirmed (DR Clemmons is working on that), it still does not explain the disease nor why only a very limited few get DM. So, we have a lot of answers that have not been discovered and they will not be if we do not continue to do research. Personally, I do not trust Dr. Coates to lead the research for GSDs. I would prefer to have a researcher work on the type of DM that affects our breed- as GSDM is different than the DM of Corgis and Boxers! They are 2 entirely different diseases.
Neither the gene test or the Flash Test can tell you which dogs to eliminate from the gene pool- it is way too early in research to be able to do that. Neither test can predict which dogs will develop DM! However, since the gene is not the soul factor for developing DM, I am not sure that genetic testing and elimination of carriers and at risk patients is appropriate, at this stage of either research, as Based upon our research work, that could mean removing 75% of GSDs. That is foolish. We need much more info, since many factors not just one, cause a dog to develop DM.
Here is a final thought for you. If the gene test’s advantage is that they can find the carriers as well as the hnomozygous dogs, then the predictive value of a positive test is only 3%. If they only report the homozgous dogs then the predictive value is 10% just like the DM Flash test. Neither test is a good screening test, but we have shown that the Flash test is a good diagnostic test in clinically affected dogs.
In my discussions with Dr Clemmons, he has told me that If carriers are detected and the incidence is 75% of the breed while only 2% get the disease, there is no way to eliminate the disease from the DM Flash test or the gene test alone. You would have to remove between 25 and 75% of the dogs; to prevent 2% from getting the disease. That is why continuing our DNA work looking at the other regions we have already demonstrated as associated with the disease is important.
Only by finding the “imprinted” genes which allow the other genes to activate will we be able to find the dogs who will end up with the disease. Finding the DM Flash test or the gene test is just the next step in the process of understanding the disease and how to prevent or treat it. If the rest of the research is not done because everyone thinks this is the final piece in the puzzle, it will be a sad day for GSDs.
In discussions with Dr Clemmons I have learned that a test’s significance is not just based upon the sensitivity and specificity of the test, since the value of the test is also a function of the incidence of the disease in the population. So, some feel that the predictive value of a positive and predictive value of a negative test which accounts for both of these factors may be a better way to determine usefulness of a test.
For a screening test to be useful, it should have a very high predictive value, the disease should have a treatment, and the test should be cost effective. Based upon these criteria, and particularly the lack of treatment and low predictive value of a positive test, no reasonable epidemiologist would recommend performing the DM Flash test or the gene test as a screening test. We do know that the DM Flash test is a good diagnostic test, but that is because the incidence of DM in clinically affected dogs is around 25%, not 2% .
I do not know if the gene test is designed to detect both heterozygous (carriers with a single copy) and homozygous (2 copies) or just the homozygous affected dogs like the DM Flash test. Either way, neither should be used as screening tests for the disease in normal dogs, because the number who will actually come down with the disease is only 10% of the homozygous affected dogs, at best. Over all only 2 % of GSDs get GSDM. However, probably 25% of all GSDs are homozygous for what we detect in the DM Flash test.
On the other hand, almost 100% of the dogs who have GSDM are positive in the DM Flash test so that the presence of the homozygous trait does account for 91% of the reason why a give dog will develop GSDM. That is still only 2% of all GSDs.
I dont think any test has a 100% accuracy. I know the Flash test has a 96% sensitivity and 99% specificity. I dont know the sensitivity and specificity of the gene test or if they even released those figures. No test is 100%, for sure! Our test is broader and may look at things they do not. Even so, their test and ours may look at different aspects of the disease. Even if they looked at the same thing, there will be some differences.