We all know that certain types of cancer are associated with certain breeds - German Shepherd Dogs with splenic haemangiosarcoma, Boxers with brain tumours, Cocker Spaniels with anal sac adenocarcinoma and Irish Wolfhounds with osteosarcoma to name but a few. But why should this be?

Cancer is a disease of DNA. Cancers arise when there has been damage to cellular DNA, particularly damage such that cell cycle regulation is disregulated, and this is then followed by other mutations which allow the cancerous cell to survive within the host and spread.

Initial damage can occur through:

 Chance - it is estimated that spontaneous mutations occur at a frequency of 10^-6 to 10^-7 per cell per generation. The mutation can be in an important or less important piece of DNA. The damage can either undergo repair, cause the cell to undergo apoptosis, or slip through the net

 Exposure to mutagens will increase the odds of an important mutation arising. Examples are tobacco smoke, ionising radiation, ultraviolet light, anti-cancer drugs

 Introduction of viral oncogenes can lead to cancer arising.

However, it is also known that some human families carry various gene mutations which have probably occurred randomly in a germ line cell in the past which means that right from birth they are at an increased risk of developing cancer. The gene mutations known about are genes with high penetrance and are inherited in a straightforward autosomally dominant/recessive manner. They have historically been identified in cancers that are rare in the general population but common in a particular family. For example male breast cancer is associated with mutations in a gene called BRCA2. These individuals tend to have early-onset disease - one gene is non-functional due to mutation so there is only one left (on the chromosome from the other parent) to knock out by chance. As all cells carry the mutated gene - it's not just one odd cell involved - individuals tend to get multifocal disease, or different types of cancers.

There are individuals with increased susceptibility or resistance to cancers (why do some heavy smokers survive until 90?) who probably will have more complex polyvalent genetic patterns with less complete penetrance than the classic gene mutations we have found to date.

Do Dogs Have Cancer Families?

The obvious extended family in the canine world is the breed, and previously stated, we do know that there are breed predilections for cancer.

Sometimes it might be thought that the predisposition may be due to the anatomical configuration of the dog - for example osteosarcoma of the long bones of dogs is a disease of large breeds. These breeds have growth plates that have cells that need to multiply more to make longer bones as these breeds grow bigger than the average dog. It could be argued that an increased number of cell multiplications increases the risk of genetic mutations to occur randomly. These growth plates are subject to greater weightbearing which may predispose to wear and tear repairs and again for random genetic mutations to occur. However, it cannot completely account for the fact that certain of the larger breeds seem to have very high incidences of osteosarcoma in comparison to others.

There have been reports of familial lymphoma in Bull Mastiffs; a study followed 59 dogs from three households prospectively in 1984. Two of the three households had a history of lymphoma cases. During the survey period nine of 59 died with lymphoma, with the distribution of the cases displaying a familial pattern.

Since the elucidation of the canine genome in 2004 and the explosion of technology that allows relatively easy analysis of canine genetic aberrations there is a whole raft of research looking at the genetic damage that allows cancer to arise.

Genetic Testing in Veterinary Oncology - Where Are We?

We are not yet able to provide 'off the shelf tests' that will either identify an individual at risk of cancer or predict an outcome for an individual with cancer, but research is being undertaken with these aims in mind. It also may turn out to be not as straightforward as was initially believed.

Several strategies are being employed looking at different outcomes. There are different techniques that can be used depending on what is being studied. Firstly 'genetic testing' can be used to identify genes that are important in specific cancers. By taking a cohort of individuals with a specific tumour that belong to a breed that has a high risk of developing the tumour and another cohort of healthy individuals of the same breed and performing a genome-wide analysis (GWAS) on peripheral tissue such as blood or cheek swabs it is possible to identify genetic differences between the two groups. As both groups are from the same restricted gene pool the likelihood that the differences are directly related to the increased risk of developing the cancer studied is high. This approach may identify genes that are important in developing that cancer in any dog regardless of breed or indeed any human. This information may allow us to breed to eliminate the increased risk or even to identify genetic targets for therapeutic intervention. Both the affected gene and the mutation that is present within a particular gene identified as being important in a particular cancer may be the same or vary between breeds.

In addition, by performing a biopsy of a tumour once it arises, the genetic mutations (some present in the germ line DNA and others likely to have arisen in the cancer cells during development and progression of the tumour) can be analysed to look for a 'malignant signature'. For example in humans it is known that lymphomas can be classified according to their genetic aberrations and these classifications are associated with different outcomes. Currently work is being done on several canine tumours where there is a wide range of behaviour, such as mast cell tumours and soft tissue sarcomas, to identify similar signatures associated with a good or bad outcome.

In conclusion, therefore, it can be said that 'genetic testing' in cancer is a valuable research tool which is helping us understand how cancers arise and how they may behave within an individual, and allows us to identify potential therapeutic targets. In human medicine genetic testing of peripheral blood has allowed identification of familial cancers associated with single genes with a high penetrance. In dogs (and most human cancers) there are likely to be multiple genes with low penetrance associated with an increased risk of cancer. Therefore it may be more difficult to identify a single gene test that gives a clear yes or no answer. More likely a 'gene profile' may be developed in the future that identifies a quantifiable risk of cancer in a particular individual.

References

1.  Breen M. Update on genomics in veterinary oncology. Topics in Companion Animal Medicine 2009;24(3):113–121.

2.  Onions DE. A prospective survey of familial canine lymphosarcoma. Journal of the National Cancer Institute 1984;72(4):909–912.

3.  Shearin AL, Ostrander EA. Leading the way: canine models of genomics and disease. Disease Models & Mechanisms 2010;3(12):27–34.