We are very pleased to announce that an improved marker test for Labrador Retrievers is now ready. Much credit for this advance goes to the LR owners and service dog organizations that participated in this research effort by supplying new eye exam reports for their dogs. While this new test is significantly improved over the original test, it is not yet the direct mutation test we all are working toward.
As you know, the original prcd-PRA test for LRs can give false allele (false positive) results for the Pattern B and Pattern C categories. Ongoing research by Drs. Greg Acland and Gus Aguirre and their research team at the Baker Institute of Cornell University focuses on discovering the actual gene and mutation causing prcd-PRA. As part of their ongoing research, they also evaluate new DNA markers for prcd-PRA. Some of these markers have potential use for improving the prcd-PRA marker tests. Earlier, such effort led to significant improvement of the prcd-PRA tests for Portuguese Water Dogs, Chesapeake Bay Retrievers and English Cocker Spaniels.
Success with new LR markers
Two new markers in particular give very reliable new information about prcd-PRA in LRs. These markers were evaluated first at the Baker Institute on their set of research samples from LR, including known PRA-affecteds. When it looked solidly promising, OptiGen tested the new markers further. Application of the new markers was tested on 462 Pattern C dogs and was correlated with their clinical status of non-affected or affected, as reported by eye exams.
Of these 462, 31 Pattern C LRs that also were PRA-affected were tested with the improved markers and all were Pattern C1. (A1, B1 and C1 designate results from the improved test.) There are no LRs known to OptiGen as affected with typical symptoms characteristic of prcd-PRA that do not test as Pattern C1.
Twenty one additional Pattern C dogs stay in the high-risk group of Pattern C1 even with the improved test. Some of these C1 dogs are under 4 years old, which usually is too young to detect development of PRA. However, some of these C1 dogs are between 4 and 12 years old and still have a normal eye exam. The explanation might be that some false alleles remain with the improved test, still to be resolved with furtherresearch. Unaffected C1 dogs should be examined annually, as usual, and report copies shared with OptiGen.
The remaining 410 LRs that were reported previously as Pattern C now test either Pattern B1 or Pattern A1.
Likewise, about half of the dogs that originally tested Pattern B now test Pattern A1 with the improved marker test.
The improved marker test also solves the status of a specific European line of LRs that produced PRA-affected offspring although the original prcd genetic test did not predict it.
We discovered that this LR line (so far, less than 10 dogs are involved) has an altered combination of markers associated with prcd-PRA, and this combination is recognizable with the improved test. This new marker combination is taken into account in the improved test Patterns.
Adjusted pattern distributions
Use of the improved test on all of the LR samples tested by OptiGen so far shows that the number of Pattern A1s doubled, and the number of Pattern B1s and C1s decreased significantly. These trends were expected based on the experience of breeders, and now the improved test reflects this expectation.
Results for OptiGen tested LR samples show the following Pattern distribution with the original and the improved tests.
A B C
Original Marker Test 30% 47% 23%
A1 B1 C1
Improved Marker Test 67% 30% 3%
Good news/bad news
There is an important difference with the LR improved marker test compared to the original test. The original marker test was expected to be 100% certain for Pattern A results, however, the original test also had an unacceptably high proportion of false positive results. This will no longer be the case - detection of a false positive allele is greatly reduced.
However, there is a chance of recombination (see NOTE below) - or dissociation - between the new markers and the prcd gene. Such recombination generates a low but real margin of error for each result Pattern. This information is new. So far, no recombination events have been detected in LR between prcd and the markers used in the improved marker-based test.
However, a possibility of such a recombination does exist, and although such events are rare, this new information must be used to interpret all results in the most cautious way.
Using this more cautious interpretation as a guideline, there is now less than 100% certainty for all Patterns. This theoretical risk of recombination applies to all dogs tested, whether with the original test or with the improved test.
Margin of error remains
As shown in the Table at the end of this report, the improved test has a small margin of error with each result, using most conservative, that is, most cautious, interpretations.
Pattern A1 dogs are statistically normal for prcd-PRA and are not expected to develop this disease or pass it to offspring. No known Pattern A1 dog has developed prcd-PRA or produced prcd-PRA-affected offspring, and no known prcd-PRA-affected dog has tested Pattern A1. However, there is a low theoretical chance that a Pattern A1 dog could have a false NEGATIVE result and therefore be a carrier (risk is less than 0.5%) or even be affected (risk is less than 0.0025%). So far, false negative results have not been observed with the current test. Significance for breeding: Pattern A1 dogs can be bred to any dog and, with at least 99.5% confidence, will not produce pups affected by the prcd form of PRA.
Pattern B1 dogs are not expected to develop prcd-PRA and are classified as carriers of PRA. There is a chance (risk is less than 0.5%) of a false NEGATIVE result, which would make this dog prcd-PRA-affected instead of a carrier. Additionally, there is a similar chance (risk is less than 0.5%) of a false POSITIVE result due to recombination of markers and the prcd gene, which would make this dog genetically normal. So far, neither of these possibilities has been observed. Additional false POSITIVES due to a false allele in the LR population cannot be definitely resolved until there is a mutation test that directly detects the prcd gene mutation. Until then, it is prudent to use the most conservative interpretation and consider Pattern B1 dogs as carriers. Significance for breeding: Pattern B1 dogs have a high probability (possibly as high as 99%) of being a carrier of prcd-PRA and a very low probability of being affected. They should only be bred to Pattern A1 mates in order to reduce greatly the chance of producing affected pups.
Pattern C1 dogs are homozygous for prcd-PRA markers and are at high risk for developing PRA. There is a small theoretical chance, less than 0.5%, of a false POSITIVE result due to recombination of markers and the prcd gene, meaning that a Pattern C1 dog could be a carrier, or with even lower chance, less than 0.0025%, could be normal. So far, false positive results due to recombinations have not been observed with the current test.
Additional false POSITIVES due to a false allele in the LR population cannot be definitely resolved until there is a mutation test that directly detects the prcd gene mutation. Until then, it is prudent to use the most conservative interpretation that Pattern C1 dogs will eventually develop prcd-PRA, if they live long enough. Cases of extremely late onset PRA in the LR are known, suggesting that some Pattern C1 dogs might not be diagnosed with PRA during their lifetime. Significance for breeding: Pattern C1 dogs should only be bred to Pattern A1 mates in order to reduce greatly the chance of producing affected pups.
Limits to all genetic testing
There are certain limits for any and all DNA genetic tests: whether a test is mutation or marker-based, it identifies only the specific mutation being tested or the association between a specific marker set and the disease. For example, a mutation test will detect one specific mutation in one specific gene. If there are several different mutations or several different genes that can cause the same condition, one must discover and then test for each mutation and each gene. Likewise, a marker test will use one marker or set of markers to define a specific condition. If the condition is associated with several different marker combinations, one must discover and then test for each marker combination. It can be difficult or even impossible to know how many mutations or how many marker sets exist in all the members of a specific breed. As more and more dogs are tested, previously unknown variations will come to light.
Also keep in mind that not all retinal disease is PRA and not all PRA is the prcd form of PRA. Accurate diagnosis is essential. A dog can test as Pattern A1 or B1, yet have a different type of retinal disease. Although other types of retinal degeneration occur in the LR, by far the most common is prcd. Even though the only PRA disease that has been fully defined and reported for LRs so far is the prcd form of PRA, we are interested in knowing about other retinal diseases. This information permits us to examine the samples in the research lab to determine if other genes are also involved.
Re-tests and new tests
All LR prcd-PRA tests starting July 14, 2003 will be reported based on the improved marker test. There will be no need to request a re-test for test reports dated July 1, 2003 and later. The cost of this test will remain the same even though the testing has been expanded with new markers.
Re-test reports, and certificates where eligible, are being sent at no charge to owners who provided follow-up eye exam reports on Pattern C dogs for use in this research.
If your dog is Pattern A based on the original test, you do NOT need to request a re-test report unless you want it to show the result as Pattern A1.
For all other dogs tested before July 1, 2003, there is a charge of $25 to receive a re-test report. A certificate of normal for Pattern A1s is included in this fee and will be mailed shortly after you receive the report. You do not need to send another blood sample.
The request form can be printed from OptiGens site, and mailed or faxed to OptiGen along with your payment. Turn-around for re-test reports will be about 2 weeks.
*NOTE Recombination of markers and the prcd gene can come about in two ways.
1. The chance is approximately one in 200 that such a dissociation event would happen for the first time during meiosis (formation of egg or sperm in the parent). In scientific terms this is calledrecombination. If that parent is a prcd carrier (heterozygous), the outcome would be that the set of normal markers recombined with the prcd disease gene, and the prcd gene therefore goes undetected. Thus, the Pattern A1 or B1 dog that has a carrier parent has a 1 in 200 chance of inheriting a newly recombined chromosome from a carrier parent.
However, if a recombination occurred in a homozygous clear parent or in a homozygous affected parent, the recombination would be of no consequence. So far, one new recombination between markers and the prcd gene was found in the Toller breed and one in the mixed breed research dog colony.
2. A dog could be a false negative Pattern A1 or B1 if a recombination occurred in the distant past and the recombinant chromosome was transmitted to subsequent generations, creating a small pool of false negative chromosomes. Although theoretically possible, it is highly unlikely. If there were false negative dogs already existing, prcd-PRA affected Pattern A1 or B1 dogs should have been reported to OptiGen by now. So far, no LR fits this description. Currently, there are two possible instances in all 7 breeds being tested for prcd-PRA: a Labrador Retriever line in Europe and a small subset of another breed in the US. Both cases are still under study.
Possible results using the OptiGen prcd test - with most conservative interpretation
Pattern/ Risk Group Significance For Breeding Risk For Developing prcd
A1 I. Normal Can be bred to any dog, Not expected to develop
with 99.5% confidence of prcd, risk is less than
producing no affected pups 0.0025%
B1 II. Nonaffected Probably a carrier of prcd Not expected to develop
(99.0% confidence)and should prcd, risk is less than 0.5%
be bred only to Pattern A
C1 III. High Risk Probably homozygous for prcd Probably will develop prcd,
(99.5% confidence)and should be with only 0.5% chance of not
bred only to Pattern A being affected.
*Probability risks are based on estimated frequency of recombination between the improved prcd markers and the disease gene.