Appendicular central osteosarcoma (OSA) is a common and highly aggressive tumour in dogs. Metastatic disease to the lungs is common and even with chemotherapy the prognosis is generally poor. However, few cases survive well beyond reported median survival times. Current methods, including histologic grading schemes, have fallen short in their ability to predict clinical outcome. MicroRNAs (miRNAs) are small molecules present in all tissues and bodily fluids and are dysregulated in cancer. Previous studies have demonstrated the diagnostic and prognostic potential of miRNAs in canine OSA. We sought to investigate multiple miRNA and multiple variable models for diagnosis and prognosis of canine OSA using plasma samples across three populations of dogs from two veterinary biobanks. Fifty-six miRNAs were analyzed by real-time quantitative polymerase chain reaction. MiR-214-3p was the only miRNA with increased expression across all OSA populations compared to controls. Using a decision tree model for diagnosis, miR-214-3p was the first step in this multi-miRNA model. High expression of miR-214-3p alone was also a predictor of shorter overall survival and disease-free interval across all populations. In both multiple miRNA and multiple variable models, miR-214-3p was always the first decision point with high expression consistently predicting a worse prognosis. Additional miRNAs in combination with low expression of miR-214-3p similarly had a worse prognosis demonstrating better outcome prediction using multiple miRNAs compared to using miR-214-3p alone. Multiple variable models only need to use miRNAs to be predictive although clinical parameters such as age, sex, and tumour location were considered. MiR-214-3p is clearly an important prognostic predictor of canine OSA in plasma as supported by previous studies and across our multiple sample populations. Multiple miRNA models provided superior categorization of patients in predicting clinical outcome parameters compared to the single miRNAs.
Copyright: © 2024 Ludwig et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.