Global Gene Expression Analysis of Canine Cutaneous Mast Cell Tumour

Introduction: Currently, prognostic and therapeutic determinations for canine cutaneous mast cell tumors (MCTs) are primarily based on the histological grade. However, the prognostic value of this latter is still highly questionable. In the present study, MCT transcriptome was characterized to identify a set of candidate genes potentially useful for MCT classification and prognosis prediction. Materials and Methods: Fifty-two canine MCT biopsies were enrolled in the study. Isolated and purified total RNAs were individually hybridized to the Agilent Canine V2 4x44k DNA microarray. Normalized data were analyzed by using SAM (Significant Analysis of Microarray), PAM (Prediction Analysis of Microarray) and TMeV tools. Furthermore, a Functional Annotation bioinformatic tool (DAVID) was used to classify modulated genes. Results: PAM identified 14 transcripts providing the greatest accuracy of class prediction into two classes (mostly referable to High and Low grade MCTs) with a misclassification error equal to 0. The functional analysis of genes differentially expressed (597) between the aforementioned two groups provided evidence that they were involved in cell cycle, DNA replication, p53 signaling pathway, nucleotide excision repair and pyrimidine metabolism. The PCA of all samples, made by using this same panel of genes, clearly identified two clusters (the first two components accounted for the 90.9% of total variance). Conclusions: The molecular characterization of canine MCT transcriptome allowed the identification of a gene set that clearly separate differentiated and undifferentiated MCTs. This might potentially be helpful for MCT classification and prognosis. Supporting grants: RC IZS VE 04/10