The Nose Knows: Dog Sniffs Out Colon Cancer

David A. Johnson, MD


February 10, 2011

This feature requires the newest version of Flash. You can download it here.

Hello, I'm Dr. David Johnson, Professor of Medicine and Chief of Gastroenterology in the Eastern Virginia Medical School in Norfolk, Virginia. I've got a news flash for you: canines useful for colon cancer detection by sniffing. It sounds like a joke, doesn't it? But no, it's a reality in a study that was just published online in a very formative journal called Gut.[1] This was a prospective evaluation of the ability of dogs to detect colon cancer. That sounds pretty hypothetical. Certainly we've all seen dogs sniffing in airports, looking for illicit substances, but this is a real thing: looking for cancer.

So I looked at the background on this, although it sounds very far-fetched (no pun intended). This actually dates back to previous work with canines being able to sniff breast samples, looking for breast and lung cancer. Some work with ovarian cancer tissue sniffing has also been done. This was a prospective study from Japan where they trained a dog at their cancer-sniffing center. They developed a training program for an 8-year-old Labrador. The dog's name was Marine (for those who are interested in fine details). The dog was trained with a tennis ball (go figure, those Lab lovers will know what I'm talking about). The dog was ultimately trained to recognize the scent of colon cancer in breath and stool specimens. The dog would sit in front, and wait patiently, and if it was a successful evaluation, the dog was rewarded with a tennis ball. That was how way the dog was trained.

Ultimately, the sensitivity of detection in 36 samples was very high -- 91% for breath samples, and essentially 100% (97%) for stool samples. The specificity was 99%. So what does this mean? What are we going to do with dogs? Is this going to replace colonoscopy? I don't think so, but it does raise some very interesting prospective questions. This was an in vitro study. What will happen when this is applied in vivo (meaning the patient themselves)? Several questions come to my mind right up front.

One, will the length of the nose matter? Will a Boxer be disadvantaged relative to a Greyhound? Will height matter in the logistics of getting to the area for sniffing? Will Chihuahuas and Dachshunds have a disadvantage compared with Great Danes? We won't know that until they have been formally tested. We also don't know the implications of laboratory spills, whether food or meat would be a distraction, one more so than another, and whether that would influence the scent. What we're looking for as we go forward is the application of commands to the dog. Will, potentially, the command "fetch" turn a diagnostic exam into a therapeutic polypectomy? Wait and see, we have a ways to go before we look at this.

But, what does this mean, all jokes aside? We do recognize that there are volatile organic compounds in cancers and the ability to recognize these, be it sniffing by a dog, or potentially, some type of computer-generated analysis, it certainly raises a potential for new applications for diagnostic screening. Furthermore, if we understand what these organic volatile compounds are, maybe there are some therapeutic implications for cancer.

In the meantime, this is another reason to be kind to your 4-footed friend. A dog may really be man's best friend for lots of reasons, but until we know more, I'd keep them close at hand. You may need them for good screening exams. Dr. David Johnson, I'll leave it to you as to the best application for your dogs at home. I look forward to chatting with you again soon.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.