Abstract: Background – Certain dog breeds appear more likely to have adverse vaccine reactions, suggesting genetics may play a role. Hypothesis/Objectives – To evaluate genetic factors associated with adverse events recorded within three days of vaccination. Animals – 798,322 dogs vaccinated at Banfield Pet Hospital from January 2016 – April 2023 that had DNA samples collected via Wisdom PanelTM. Methods – Electronic medical records were used to identify dogs with (cases) and without (controls) possible vaccine-related adverse events. Control dogs were randomly selected to be five times the number of cases. Additional medical record data (e.g., signalment, number of vaccines given) were collected. Genetic breed was determined using the Wisdom Panel breed detection algorithm. Dogs were genotyped using a custom 100k Illumina Infinium XT SNP microarray and standard quality control was performed. A mixed linear model was used to calculate odds ratios and p-values for the remaining 7506 cases and 39072 controls, adjusting for a centered relatedness matrix using GEMMA software as well as sex, weight, and number of vaccines given. Genome-wide significance was based on p-value Results – Two variants on chromosome 11 reached genome-wide significance for an association with vaccine-related adverse events (p=1e-13). The associated region includes genes encoding interleukins 4 and 13. Whole genome sequencing is currently underway to identify the locations and predicted impact of variants in this region. Conclusions and clinical importance – Pre-identification of dogs at increased risk for an adverse vaccine reaction could allow proactive intervention to mitigate patient risk.
Learning Objectives:
...understand the methodology used to identify vaccine reactors and non-reactors, allowing further analysis into potential genetic mutations increasing susceptibility to adverse events
....have a high level understanding of genome wide association studies and the information these studies may provide
....see the potential capability of predictive modeling to pre-identify at risk populations
