Osteosarcoma is the most common form of primary bone cancer, largely occurring in teenagers and young adults. However the overall incidence of the disease is low and accounts for only 0.2 percent of all malignant tumors. The common thread within this otherwise heterogeneous group of bone tumors is the production of immature bone by the tumor cells. The mechanism underlying this pathological process is unclear although genetic and/or epigenetic events that disrupt osteoblast differentiation are now thought to be involved. The challenge of studying the genetics of such a rare human condition has led scientists to look at applicable animals models. Kerstin Lindblad-Toh from the Broad Institute of MIT and Harvard, USA, and colleagues demonstrate how the dog model of osteosarcoma provides novel insights into the genetics of the disease that may prove beneficial for both humans and dogs. Here Lindblad-Toh tells us more about the results and implications of their genome-wide analyses published in a recent Genome Biology study.


What are the benefits of using the dog as a model for osteosarcoma?

Osteosarcoma is rare and relatively heterogeneous in humans making it hard to study the genetics. In dogs osteosarcoma is quite common in specific breeds. The canine breed structure, where dog breeds have been recently formed from a small number of founder dogs, makes it much easier to map genetic risk factors for disease. In fact, a smaller number of risk factors tend to have accumulated within each breed allowing us to determine what genes and disease mechanisms drive the disease in each breed.


Why did you choose the three particular dog breeds for your study?

We chose three dog breeds where osteosarcoma is common. In greyhounds, Rottweilers and Irish wolfhounds 15-25 percent of dogs get osteosarcoma, making it relatively easy to collect blood samples from enough dogs with osteosarcoma.


What are the most interesting findings of your genome-wide analyses? How do these results relate to the human condition?

We find a total of 33 risk factors, which are different in different breeds, suggesting that disease mechanisms can be diverse in dogs as in humans. In humans very few genes have been found so far. These genes point to pathways that all have a common theme of bone growth and differentiation. Still the mechanisms resulting in disease vary slightly between breeds suggesting that different genetic mutations can result in disease in different ways.

We also show that the inherited risk factors discovered here appear to work together with additional mutations in the tumor cells to cause disease. Importantly, the inherited mutations and somatic mutations hit the same pathways. Some are well-known such as p53, and others are novel findings.

Finally, we show that many of the mutations appear to act through aberrant regulation of genes, not through mutations in the actual proteins. This is intriguing and could also explain why it has been harder to find mutations in human osteosarcoma using for example exome sequencing.


How do you think the results of your genome-wide association study in dogs could be translated into humans?

To get a better handle on which of the canine disease genes and mechanisms are relevant in humans, we plan to sequence the genes and pathways found in the dog in human patient samples.

We think we can learn a lot about disease mechanisms in human patients by more carefully studying the mutations and disease mechanisms found in dogs. A better understanding of the disease biology could potentially lead to better treatment options in the future for both dogs and human patients.


What’s next for your research? Are there plans to further your studies in dogs or humans?

We plan to continue to look for genes in more dog breeds, to get an even more complete picture of the genes that are affected in osteosarcoma. We are also sequencing tumors to find somatic mutations to further study how inherited risk factors connect to tumor mutations.

We are currently studying the disease mutations and their functional effects in normal cells and in tumors to understand what goes wrong. This could lead to better treatment strategies and can also tell us if dogs with specific mutations respond better to specific treatment regimes.

In parallel, we will sequence the same genes in human patients and compare the genes and disease mechanisms in dogs and humans.


What other genetic studies have you carried out in dogs?

We are working on a large number of diseases in dogs including multiple cancers, multiple immunological diseases such as lupus, and several behavioral conditions including obsessive-compulsive disorder.

Dogs are excellent models of human disease based on the breed structure, but also as they have a similar set of genes, similar disease presentation, share our environment and receive health care in a similar fashion to humans.


Can studies into human diseases using the dog as a model, also prove beneficial for the equivalent conditions in dogs?

Dogs and humans are similar enough that findings about disease in one species are also important for the other. Learning about the disease biology in dogs also provides opportunities for better diagnostics and treatment for the dogs. In this context it is important to note that dogs get treated primarily with medicines developed for use in human patients and often these medicines have about the same effect in dogs as in humans. Thus learning about canine disease will benefit both dogs and humans.


Questions from Maria Kowalczuk,  Deputy Biology Editor for BioMed Central.


More about the researcher(s)

  • Kerstin Lindblad-Toh, Professor of comparative genomics, Uppsala University, Sweden.

    Kerstin Lindblad-Toh

    Kerstin Lindblad-Toh is a professor of comparative genomics at Uppsala University, Sweden, the Co-Director of SciLifeLab in Sweden and the Scientific Director of Vertebrate Genome Biology at the Broad Institute of MIT and Harvard, USA. Lindblad-Toh did her PhD in human genetics at the Karolinska Institute, Sweden. She then continued her research career developing mammalian… Read more »


Highly AccessedOpen Access

Genome-wide analyses implicate 33 loci in heritable dog osteosarcoma, including regulatory variants near CDKN2A/B

Karlsson EK, Sigurdsson S, Ivansson E, Thomas R, Elvers I, Wright J, Howald C, Tonomura N et al.

Genome Biology 2013, 14:R132

Go to article >>

Related posts