Man has been exploiting dogs’ remarkable sense of smell for some time now, using them to detect illegal substances, missing people, explosives, wildlife, blood and even things like banned electronics in prisons. But it seems that we may have found yet another use for those soggy noses, as recently there has been an increasing amount of evidence to suggest that they could help us detect various different human cancers with remarkable accuracy.
Just last month, scientists announced that a German shepherd-mix called Frankie could detect thyroid cancer in urine samples with an 88% success rate. Now, two more German shepherds have been trained to detect prostate cancer, and they got it right more than 95% of the time. These findings can be found in The Journal of Urology.
Prostate cancer, which is one of the leading causes of cancer death among men, is currently detected by a blood test that looks for a protein called prostate-specific antigen (PSA), which is produced by cells of the prostate gland. Although this is normally present in small amounts in the blood, raised levels can indicate the presence of prostate cancer. However, other things such as inflammation or a urine infection can also cause an increase in blood PSA, which is why it’s not reliable enough to be used on its own and must be followed up with physical examinations and invasive tests like tissue biopsies.
Since there is evidently a clear need for better prostate cancer tests, scientists are looking for new and more accurate ways to pick up the condition, which is why some have turned their attention to world-renowned detectors: dogs. Previous studies have shown that dogs can sniff out certain chemicals present in the urine of men with prostate cancer, so scientists from the Humanitas Clinical and Research Center, Milan, decided to take this further by testing out how accurate trained dogs could be.
For their investigation, they used two female German shepherd dogs who had been previously taught to sniff bombs and trained them to identify prostate cancer-specific compounds in urine samples. They then tested them out on 362 patients with prostate cancer and 540 controls who either had no cancer or nonprostatic tumors. They found that one dog picked up 100% of the prostate cancer samples and only falsely identified 7 negatives. The second dog correctly identified 98.6% of positive samples and wrongly identified 13 non-prostate cancer samples as positive.
Although these results are impressive, the study has important limitations, which means that we probably won’t see dogs in clinical settings just yet. Presently, it is unclear which cancerous chemicals the dogs are detecting, and if we do work this out, then it might be possible to develop lab-based tests for them. Furthermore, as the NHS points out, the study only involved men who had already been diagnosed with prostate cancer, so further studies are needed to see if they can pick up cases that have not yet been diagnosed, or to monitor men with high PSA but negative biopsies.