00990-A: Prognostic Indicators for Anal Sac Gland Carcinoma in the English Cocker Spaniel and Preparing to Map Predisposition to this Tumor

Grant Amount: $12,678

David Sargan, MA, PhD; University of Cambridge

July 1, 2007 - December 31, 2007

Sponsor(s): Samoyed Club of America Education & Research Foundation

Breed(s): English Cocker Spaniel

Research Program Area: Oncology

Project Summary

The goal of this project was to perform a whole genome scan in the English cocker spaniel in order to identify the regions of the genome that contain genetic variants that predispose this breed, but not others, to anal sac gland carcinoma. The whole genome scan involves comparing the genome of English cocker spaniels that presented this carcinoma, to the genome of age-matched controls (e.g., cocker spaniels aged 9 and above without any history of this type of tumour). Whole genome scan studies for complex diseases, such as cancer, usually include at least a hundred cases and the same number of age-matched controls. At the moment there are 61 cases. The project short of control dogs aged 9 years and above; at the moment there are have 5 such dogs plus 32 additional samples corresponding to young non-affected dogs or to dogs with no age information. While the number of samples required to perform the whole genome scan are collected, a candidate-gene association study has been performed. Three genes that have an important role in the immune system have been chosen for the anal sac gland carcinoma in the English cocker spaniel. They belong to the dog leukocyte antigen system (the equivalent of the Human Leukocyte Antigen system), and code for proteins forming antibodies, this means that they are highly variable or polymorphic. The variable portion of 3 of these genes (DLA-DRB1, DLA-DQA1 and DLA-DQB1) have been sequenced in 61 cases and 38 controls (most control dogs younger than 9 years or with an unknown age). For one of these genes the experiments have been completed and resulted in a statistically significant difference in the distribution of alleles in cases and controls (p=0.000029). It is important to stress that theses result points to an association, but the causal relationship has not been proven. This is because, as mentioned for the whole genome scan, other genetic variants in close proximity to the ones studied might also show a statistically significant association so it will be necessary to distinguish the "passive associations" from the causal ones.

Publication(s)