A Multi-disciplinary Approach to Investigating Canine Diffuse Large B cell Lymphoma

RATIONALE: In humans, DLBCL has been shown to demonstrate genetic heterogeneity and miRNA profiles predicting disease prognosis. In dogs there are limited studies and relatively little genetic characterisation of lymphoma, meaning that it is difficult to find correlation with the human disease. Nevertheless, in both humans and dogs, there is a significant subset of lymphomas that are refractory to treatment.

OVERALL AIM: We propose to better define genetic subtypes of canine lymphoma and characterising response to treatment of the different subtypes through a combination of more accurate phenotyping of lymphomas by integrating a) advanced image analysis b) genome sequencing, and c) total RNA sequencing (including micro-RNA sequencing).

HYPOTHESIS: Canine lymphoma can be more accurately diagnosed, treated and prognosticated through a synergistic molecular, genetics, and image based analysis.

SPECIFIC PROJECT OBJECTIVES

OBJECTIVE 1: Define morphometric characteristics of canine lymphoma through image analysis of digital whole slide images (WSIs) via machine learning and artificial intelligence methods

OBJECTIVE 2: Identify genetic markers for DLBCL susceptibility through genome wide association studies and CNV analyses

OBJECTIVE 3: Define mRNA and miRNA expression profiles in canine DLBCL through mRNAseq and small RNA-seq, and investigate if tumour-associated miRNA can be detected in peripheral blood serum

OBJECTIVE 4: Genetic and morphological signatures from 1-3 will be compared between cases that show good response to chemotherapy and those that are refractory to treatment.