Aims: We developed a test to detect hemangiosarcoma (HSA)-associated cells in the circulation of dogs prior to the onset of grossly detectable disease. Our objective is to assign dogs into risk categories for development of HSA, allowing for the rational deployment of eBAT, a drug to target the cancer-stem cell compartment and the tumor niche in these dogs, as an effective means for cancer prophylaxis.
Methods: We used samples from 125 dogs (28 with hemangiosarcoma, 29 with other cancers, 27 with benign vascular pathology, and 41 for healthy) as a training set for machine learning. We used 10-fold cross validation, “leave-one-out” analyses to establish parameters of the test for detection of HSA-associated cells, and to determine sensitivity and specificity. We then applied the algorithms to a prospective cohort of 209 clinically healthy golden retrievers, boxers, and Portuguese Water Dogs older than 6 years of age as a validation set to assess the performance of the test in the early detection setting.
Results: Using 10-fold cross validation, the test achieved a classification accuracy of 85% for healthy dogs and up to 89% for dogs with HSA, with 89% sensitivity and 95% specificity for detection of HSA. In the prospective cohort, more than 50% of clinically healthy dogs were categorized into one of the three conditions indicative of pathology (HSA, other cancer, benign splenic pathology). The test seems to have outstanding specificity and acceptable sensitivity for use in early detection: the false negative rate at 6-months after testing was <1% (1 of 99 dogs). Twenty-one dogs in the prospective cohort have developed cancer or another chronic condition. Nineteen that had a prediction of cancer-associated pathology were diagnosed with cancer (90%). Only two of these 21 dogs (10%) were misclassified. One had a prediction of cancer-associated pathology but died from a nonmalignant condition, and one had a prediction of “healthy” but died of cancer.
Conclusions: We developed a blood test that can successfully detect the presence of HSA-associated cells in dogs with active disease, and that can be used in the early-detection setting to categorize risk of developing HSA. Our results from a prospective cohort of 200 dogs over 6 years of age suggest as many as 50% of these dogs have inapparent disease. The performance of the test is acceptable for use as an actionable component of a cancer prevention platform.
Clinical Significance: To our knowledge, this is the first test that can accurately assign risk for development of hemangiosarcoma to clinically healthy dogs, providing a rationale for cancer chemoprophylaxis. Our study also provides proof of concept for prospective, large-scale trials for early cancer detection in companion dogs.
Citation Format: Taylor A. DePauw, Ali Khammanivong, Jaime F. Modiano. A method for early detection of hemangiosarcoma in dogs [abstract]. In: Proceedings of the AACR Special Conference on Advances in Liquid Biopsies; Jan 13-16, 2020; Miami, FL. Philadelphia (PA): AACR; Clin Cancer Res 2020;26(11_Suppl):Abstract nr B22.