My Dog's DM Test Came Back "At Risk." What Does That Actually Mean?

An at-risk DM result means your dog carries two copies of a variant in the SOD1 gene. It does not mean he has degenerative myelopathy, and it does not mean he will get it — not all dogs with two copies develop the disease, and the one published figure — 60% — counted owner-reported signs in 30 dogs rather than lifetime penetrance. What the result changes is the shortlist: if this dog's back legs start to fail at nine, degenerative myelopathy moves to the top of the differential instead of sitting fourth behind hips, knees and discs. That is genuinely worth having. It is also all you have bought, and the useful work starts after the email arrives.

The result answered a question next to the one you asked#

You went looking because of something: a breeder's paperwork, a scuffing back paw, a nine-year-old Shepherd who has started to wobble, or simply the fact that a panel you bought for something else came back with a line on it you did not expect.

The question in your head was about the next three years. The question the swab answered was about the genome, and it answered that one very well. Those are not the same question, and the gap between them has no number in it.

This matters because the result does not arrive looking like a partial answer. It arrives as a bolded phrase — At Risk — on a document from a laboratory, and it reads like a verdict. It is not one. It is a change to a list.

Everything below is about what to do with a changed list.

What the three results actually say#

The Orthopedic Foundation for Animals reports the SOD1 test in three genotypes, and its own descriptions are worth reading in full because they are more careful than most of what gets said about them.^[1]

The mutation itself was mapped in 2009, from 38 affected Pembroke Welsh Corgis and 17 related normal controls. The team found a single G-to-A substitution in SOD1 producing an E40K missense mutation, and homozygosity for the A allele was associated with DM across five breeds: Pembroke Welsh Corgi, Boxer, Rhodesian Ridgeback, German Shepherd dog, and Chesapeake Bay Retriever. Spinal cords from affected dogs carried inclusions much like those in people with SOD1-mutation ALS, which is why the paper concludes that canine DM is "the first recognized spontaneously occurring animal model for ALS."^[2]

Notice the word in the standard clinical review: the mutation "was found to be a risk factor for DM."^[3] Not a cause. The laboratory that identified it says the same thing in owner-facing language: "not all dogs that have the mutation will develop DM so the mutation test is currently a test for risk."^[4]

The carrier result is not quite the free pass it reads as#

This is the part almost nobody publishes, and it comes from the largest survey of the allele ever run. In 2014 a team retrospectively analysed genotypes from 33,747 dogs and examined spinal cord sections from 249 of them. The A allele turned out to be widespread — found in cross-bred dogs and in 124 different breeds — while a second variant, SOD1:c.52T, appeared only in Bernese Mountain Dogs.

Then the sentence that should be on every result sheet: most of the dogs with histopathologically confirmed degenerative myelopathy were homozygotes, but eight dogs with histopathologically confirmed DM were heterozygous carriers, with no other sequence variants in their SOD1 coding regions.^[5]

Eight dogs out of a very large sample is a small number, and the authors' own conclusion is the one to carry: homozygotes are at much higher risk than heterozygotes. In the low-bias follow-up cohort described in the next section, carriers developed signs at 4% (2 of 55) against 6% (3 of 52) in dogs with two normal copies. The authors ran no significance test on that pair and neither will we, but two dogs against three is not a gap you can build anything on — which is why the honest word for a carrier's risk is lower, not low. But "much lower risk" is a different sentence from "cannot happen," and the difference matters at the exact moment it is being used to close down an investigation. A carrier whose back legs are failing still needs the examination.

The one number people quote at you, and what it actually counted#

The question every owner asks next is: what are the odds. There is a number in circulation, it is roughly 60%, and it is worth knowing exactly what it counted before you carry it around.

It comes from the same 2014 survey. The authors knew that most of their genotypes were badly biased — dogs get tested because somebody is worried — so they carved out the one subset that was not: dogs who had been clinically normal and under 8 years old when their sample was submitted, and who were over 10 years old when their owners were surveyed again years later. That cohort was 137 dogs across 17 breeds, and the result was this.^[5]

Read the bottom row before the top one. Three dogs with two entirely normal copies of the gene also developed signs consistent with degenerative myelopathy. They almost certainly did not have it. What that row measures is how many ten-year-old dogs develop a wobbly back end for reasons that have nothing to do with SOD1 — arthritis, discs, lumbosacral disease, an old injury — and it is the built-in control that tells you what the 60% is made of.

So the 60% is real, and it is not lifetime penetrance. It counts owner-reported signs, not histopathology. It rests on 30 dogs. Both of its known biases push the same way, upward: the denominator excludes every at-risk dog who died of something else before ten, and the numerator includes whatever the 6% row is picking up. If it is wrong, it is wrong high. The authors themselves drew only the comparative conclusion — that carriers sit close to normal dogs and far below homozygotes — and that comparison is what the data can carry.

Which is why the institutions decline to publish a headline figure. The OFA says research is ongoing to estimate what percentage of dogs testing A/A will develop DM within their lifespan.^[1] Cornell's canine health centre goes further and says that even in a genetically predisposed dog, the overall chance of becoming clinically affected "may still be low."^[6] Doing it properly needs a large group of homozygous dogs followed to the end of their lives with confirmation at the end, and that has not been done.

Two things make it harder still. The disease is late in onset — the OFA notes that some affected dogs did not begin to show signs until they were fifteen^[1] — so a homozygous dog who dies of something else at eleven counts as "unaffected" in a way that quietly inflates the apparent good news. And DM cannot be confirmed in a living dog at all, which means the denominator and the numerator are both soft.

What does exist is a study that proves the variability is real rather than just unmeasured. In 2016 a genome-wide association analysis took Pembroke Welsh Corgis who were all homozygous for the SOD1 mutation and compared the ones who developed DM against the ones who did not — a study design that only makes sense if plenty of homozygous dogs never get the disease, and the paper says so in its abstract, up front: "not all dogs homozygous for this mutation develop disease." A haplotype within the SP110 gene was present in 40% of affected dogs compared with 4% of unaffected dogs (P = 1.5 × 10⁻⁵), and was associated with an increased probability of developing DM (P = 4.8 × 10⁻⁶) and an earlier onset (P = 1.7 × 10⁻⁵).^[7]

Forty percent against four percent. Same disease, same two copies of SOD1, and a large part of what separated the corgis who got it from the corgis who did not was sitting in a completely different gene. That finding belongs to Pembroke Welsh Corgis and has not been shown to transfer — but it is a demonstration, in one breed, that an at-risk genotype is a statement about one locus in a system with more than one.

So should you test? Four readers, four answers#

You are choosing a mating. Yes, test, and this is the use the test was built for. The 2014 survey's own conclusion is that "breeding to avoid the production of SOD1:c.118A homozygotes is a rational strategy."^[5] Note what that sentence does and does not say: avoid producing homozygotes, not avoid breeding carriers. The OFA is explicit that breeders should factor the result in but "should not over-emphasize the test results," and that an at-risk dog is a more serious fault than a carrier rather than an automatic exclusion.^[1] In a breed where the allele is common, culling every carrier costs more genetic diversity than it buys.

You are buying a puppy. Ask the breeder for both parents' results rather than testing the puppy. Two clear parents cannot produce an at-risk puppy. One clear parent cannot either — a clear-to-carrier mating produces carriers at worst. It is carrier-to-carrier and anything-to-at-risk that can produce a homozygote. That is a question you can ask in an email before you have paid a deposit, and it costs nothing.

Your healthy middle-aged dog is a listed breed and you are curious. This one is a genuine judgement call and it depends on what you would do with the answer. There is no preventive treatment to start, no screening imaging worth doing, and nothing about a five-year-old's care that should change. What you would get is advance notice — and if you are the sort of person who would use that to establish a baseline now and to book an examination early rather than late, the sixty-five dollars is well spent. If you are the sort of person who would spend the next four years reading the result into every stumble, it is not.

Your dog's back legs are already going. Test, and expect it to be one input rather than the answer. It belongs alongside a neurologic examination, not instead of one. And be clear which result you are hoping for: the useful one is normal, because it argues for something else, and some of the something-elses can be treated.

dogorthopedic.com

Three tests called DM, at $65 each, and you probably need one#

Here is a piece of purchase intelligence nobody writes down. The fee for all OFA/University of Missouri DNA tests is $65 per test, taken on two buccal cheek swabs the OFA mails you, with automatic registration of the result in the OFA database included.^[8] And the OFA lists three separate degenerative myelopathy tests.^[9]

The DM2 variant is the one the 2014 survey found exclusively in Bernese Mountain Dogs.^[5][10] If you own a Bernese, the standard DM test on its own is an incomplete answer and the second $65 is real. If you own anything else, it is $65 for information about a variant your dog's breed has never been shown to carry.

The risk modifier has the sharper trap in it, because the sequencing matters. It is sold for Pembroke Welsh Corgis who are already homozygous for the SOD1 mutation.^[11] Order both at once for a corgi and you will spend $130; order the SOD1 test first and, if it comes back clear or carrier, you have saved the second $65 and learned everything the pair could have told you. For every other breed, including German Shepherds and Boxers, the modifier has not been shown to apply — the 2016 authors wrote that further study is warranted to clarify the effect across dog breeds, and that study has not yielded a second breed-specific test.^[7]

We earn nothing on any of this. There is no affiliate link on a cheek swab from a university laboratory, which is one of the reasons this article can be blunt about which of the three you need.

“Because your dog can't tell you how much pain he or she is in.”

What the result is actually good for: a monitoring plan, and it is free#

Here is the reframe worth the price of the test. An at-risk result does not tell you what will happen. It tells you that if something starts happening, you already know which examination to ask for — and it hands you years of lead time to build the only thing an owner can bring to that appointment that nobody else can: a record.

That is the whole value, and the delay this site exists to shorten is made almost entirely of that lead time going unused. The months between the first quiet sign and the day someone names it are not lost to the disease. They are lost to nobody having written anything down.

So do this, once, this month, while the dog is well:

• Film ninety seconds of him walking away from you on a non-slip surface, in good light, from directly behind. Save it with the date in the filename. Asymmetry in the hind end is far easier to see against a two-year-old video than against a memory. Do it again every six months.
• Photograph both hind paws from behind, nails visible. Worn or flattened nail tips on one side are the physical evidence of a paw that is not clearing the ground, and it is one of the few signs of early DM that leaves a mark you can point at.
• Write down one dated sentence about what he can currently do: the stairs, the car, the length of the walk, how he gets up off a hard floor.
• Weigh him and write that down too. Weight is the one variable in an ageing dog's mobility that you fully control, and it matters whichever way the genotype goes.

None of that costs anything, and it converts a result you cannot act on into a baseline you can.

The signs to watch for are not the ones most owners expect. DM starts as unsteadiness rather than unwillingness, it is usually worse on one side first, and it does not hurt — which is exactly why it gets filed as "his hips." The full description of the disease and its stages is on its own page, and the distinction between a limp and a weakness is the single most useful thing to have straight in your head before the appointment. If you find yourself measuring rather than watching, thigh circumference is the most sensitive home measurement there is.

dogorthopedic.com

What to buy: probably nothing, and here is the exception#

For a healthy dog with an at-risk result, the honest answer is that you do not need to buy anything. There is no supplement, brace, bed or harness that changes the probability in that email, and anyone selling you one for that reason is selling you a feeling.

The free fix comes first and it is the rugs you already own. A dog who is unsteady behind — for any reason — loses confidence fastest on hard floors, and moving the household's existing runners onto the routes he actually walks costs nothing. Our flooring article has the specifics.

When rugs are not enough, the smallest useful purchase in this whole category is nail-grip sleeves.

The angle that belongs specifically to this article, and it cuts against the product: traction changes the thing you are monitoring.

Everything in the section above rests on nail wear as evidence. ToeGrips sit on the nail tip, which is precisely the surface that contacts the ground when a hind paw is not clearing it — and the manufacturer's own claim is that a bag lasts one to two months, which is another way of saying the grips are what wears. We have not tested this, and it follows from geometry rather than from a study, but the implication is clear enough to act on: once grips are on, your six-monthly nail photograph stops recording what it used to record.

The fix is free and takes a minute. Photograph the bare hind nails before the first set goes on, and after that photograph the grips instead — a set worn visibly flatter on one hind foot is the same evidence in a different material, and arguably a better one, because it accumulates over a known interval rather than over an unknown one.

The honest costs, none of which are small, and all of them from the manufacturer's own figures as recorded on our product page. At $39.99 a bag lasting one to two months, this is a $240–$480 a year running cost, not a $40 purchase — $204–$408 on the manufacturer's subscription price. Application needs isopropyl rubbing alcohol, which is not in the box and is the commonest reason a first attempt fails. And there is a mechanical limit specific to this disease: grips do nothing at all for the moment a knuckling dog puts weight on the top of the paw, because at that instant the nails are not touching the floor.

And the exclusion that matters most: our own product page says not to buy traction as a substitute for a diagnosis, and for a dog with new hind-end signs that stands absolutely. Book the examination first. The rugs cost nothing in the meantime.

dogorthopedic.com

What to do this week#

If the result is normal, put it in the folder and stop thinking about DM. If this dog's back end fails at ten, it will almost certainly be something else, and that is a useful thing to know in advance.

If the result is carrier, the same, with one asterisk: it lowers DM on the list rather than removing it, and it does not excuse skipping an examination if he ever starts to wobble.

If the result is at risk and he is well, spend twenty minutes this weekend on the baseline — the video, the paw photographs, the dated sentence, the weight. Then close the laptop. There is nothing else to do, and nothing you can buy that changes the odds.

If the result is at risk and something has already started, the swab has done its job by moving one name to the top of a list. Book a neurologic examination and take three things with you: the video, the paw photographs, and the sentence that separates this disease from the four things it gets mistaken for — he is unsteady, not unwilling, and I cannot find anything that hurts.

The delay this site was built around is not usually caused by a hard diagnosis. It is caused by months of a household explaining a quiet sign to itself. An at-risk result is one of the very few pieces of paper that can end that argument early, and that is worth more than the certainty it does not contain.