Skip to main content
Login | Suomeksi | På svenska | In English

Browsing by Author "Loukola, Johanna"

Sort by: Order: Results:

  • Loukola, Johanna (2021)
    Cancer is the second leading cause of death worldwide. Cancer is believed to emit volatile organic compounds (VOCs) which dogs may be able to smell. This literature review brings out previous studies about cancer detection dogs, observations made based on them and their future possibilities. There is also some information about training of the dogs, proper test settings and validating a diagnostic test. Detectable odour emissions from a patient with neoplasia, or “smell of cancer”, is an interesting topic, but it has a minor role in this work. Training of the dogs should be done with samples which disease status is confirmed so that the dogs are not rewarded for indicating wrong samples. The trainer must be able to read the dog’s signals and take care of its basic needs in order not to continue training after the dog’s concentration has deteriorated. Training should happen gradually and through positive reinforcement. There should be enough samples and they must not be the same in the testing phase as in training. In validating a diagnostic test, the main goal, materials and methods of the study must be defined and the test should be compared to a golden standard. The smell of cancer is yet unknown although there are studies of the topic. Particular VOCs, which presumably constitute the cancer smell, exist in the body’s secretions. Further studies are needed in order to understand what happens at a molecular level in cancer. There are many studies about cancer detection dogs where they try to discriminate a cancer sample among controls. In these studies, the number of dogs has varied between one to six, there are diverse breeds and dogs have different backgrounds. Good results have been achieved with only two to three weeks of training. Studies have been made of nine different cancer types and the most studied ones have been lung and prostate cancer. Urine and breath were the most common samples used. The sensitivity of cancer odour detecting dogs has varied between 18 to 100%. Previous studies have had different limitations, one of which can be considered having only one cancer sample with controls. This leads to that specificity cannot be held valid. There have been also some limitations in using dogs like short usage time because of their limited life, small number of dogs and wrong rewarding system. There should not be any systematic differences between control and cancer samples including the patient’s age and the handling procedures of samples in the study. Way of life and medications may also have an effect on the dog’s choice, but they were not taken into account in all of the studies. A small number of samples was a problem in some cases. Cancer detection dogs could possibly be used in recognizing the cancer smell, mass screenings, and surveillance after treatments, in addition to laboratory diagnostics and in the developing electronic diagnostic methods such as an artificial nose.