Can you smell cancer? Dogs may diagnose with up to 90% accuracy

They sniff out bombs, detect drugs, uncover cadavers, find the lost, and rescue the trapped. Now a special kind of canine can nose out cancer.

Building on a 1989 article in the UK medical journal The Lancet, which published anecdotal evidence of several dogs that discovered melanoma on their owners, researchers around the world have sought to understand — and utilize — these powerful olfactory skills. In a 2004 study, also in the UK, six dogs detected bladder cancer from urine samples at an overall success rate of 41 per cent. In a follow-up study in 2011, that number rose to 92 per cent. In  2011, a single dog caught prostate cancer, also using urine samples, 90 per cent of the time. Since 2013, at the University of Pennsylvania’s Penn Vet Working Dog Center, four medical detection dogs have been identifying ovarian cancer from blood samples with 90 per cent accuracy. Since July 2014, Calgary’s Clever Canines have been pinpointing lung cancer with an average 95 per cent accuracy rate using breath samples. This spring, researchers in Milan, Italy, revealed results of their study using two German shepherds to identify prostate cancer. Accuracy? A groundbreaking 98 per cent. Then there are the tales of dogs pawing at their owners’ irregular moles or snuffling unseen tumours. Who knew man’s best friend could be man’s best chance at surviving this deadly disease?

“I think dogs have probably been doing this for a while and people were like, ‘Stop! Stop! Leave me alone!’ before they put it together,” says Dr. Cindy Otto, the Penn Vet’s study director. “The reason that the ’89 [Lancet] article is particularly interesting is because somebody was open-minded enough to listen and then brave enough to put it into print. There was a potential for being laughed at by skeptics, and those of us who know it works were really proud that they were willing to do that.”

How it works for Dr. Otto and her team is with a consortium of scientists that comprises chemists, physicists, and physicians, as well as cancer patients willing to participate in these trials. Not to mention the furry, four-legged detection devices, of course, who are presented with blood samples and asked to ferret out the unique odours emitted by malignant cancer cells. Despite being buoyed by the dogs’ good work, Dr. Otto says she doesn’t foresee a future in which a hospital is run by Fido and Fifi smearing their slimy noses over our bodies.

“That’s not how it works, and that is definitely not anything we would promote,” says Dr. Otto. “There may be some people who try it but I think it’s fraught with difficulty. Anything that you do to interact with that dog could change [the results]. It might be a false-positive, it might be a false-negative. There are so many factors that could be in play there.”

Tracie Nielson, founder of Clever Canines and lead research facilitator for its canine cancer detection operation — the only one in Canada — begs to differ. The program’s entire goal is to have a functioning clinic using dogs as an early screening tool for patients at high risk for developing lung cancer (the leading cause of death from cancer, according to the Canadian Cancer Society). And she’s almost there: Nielson says the facility is already in place, and predicts the clinic portion will open some time next year for low-cost screening (roughly $350). Unlike other researchers, she and her team have not had to jump through hoops or be stymied by the same obstacles because of a very generous benefactor. Pure North S’Energy Foundation, a Calgary-based not-for-profit organization that funds chronic disease prevention initiatives, approached Clever Canines with the proposal. The 10-year-old company had been primarily involved in dog training, doggie day care and dog walking. “We’d never done anything like this before, but with our years of experience training dogs, we were absolutely up for the challenge,” says Nielson.

With a full-time staff of four and seven dogs in training, Clever Canines’ focus is currently on lung cancer detection, with a goal to eventually include breast, prostate and colon cancer. The team works with one physician from the University of Calgary, who specializes in lung cancer and collects the samples, as well as one research director who compiles the data. The dogs, all of varying breeds, sniff a variety of scents, including those emitted by cancer cells, and learn to zero in on the sample with the cells. “As handlers in the room, we have to be as still as possible and let the dog figure out the right one on his own, with no handler cues” that could confuse him and affect the results, says Nielson.

While Clever Canines trains living, breathing cancer detectives, the other studies’ objective is to develop a kind of electronic sensor (two B.C. researchers, Dr. Annette McWilliams and Dr. Haishan Zeng, are working toward a similar goal) that replicates the highly tuned sensory skills of these special pooches, but so far the dogs are winning the race. “Dogs are the Where’s Waldo of the smell world,” says Penn Vet’s Dr. Otto. “They can pick out a single, unique scent amidst a whole confounding background. Machines have more difficulty [separating similar smells]. Dogs don’t seem to have that problem with the overlap. I think the dogs do have the edge on sensitivity but for screening thousands, the machines definitely have the edge. So we may go with a less capable system just to screen the numbers. There may always be a role for the dogs. I could imagine a time where the machine wasn’t sure, so we’d check it through the dogs.” Kind of a four-legged second opinion, if you will.

Despite the exactitude of these dogs’ nostrils, there is skepticism in the medical community — at least in the U.S. New Jersey-based science journalist Arlene Weintraub, who literally wrote the book on the subject in Heal: The Vital Role of Dogs in the Search for Cancer Cures, says there’s a good reason for that. “I think there’s pretty widespread agreement that early detection is a potential cure for a lot of cancers and that we need better early detection technologies,” says Weintraub, who notes that traditional screening methods, such as mammography, are also fraught with false-positives, and the subject of ongoing controversy. “But by the same token they want these detection methods to be very, very accurate and I think that’s where some of the skepticism comes in. [Medical professionals] want to see proof that they’re going to work, and that they’re not going to have a lot of false-positives.”

For their part, Nielson says they’ve encountered very little skepticism in the course of their research, attributing that to their open, transparent approach. “We talk with professionals in the medical community, we share with them what we’re doing, and we’re open to their feedback. We even have an observation area where people can come in and watch us work.”

Weintraub’s interest is personal. She lost her sister to gastric cancer, which, like ovarian cancer, is impossible to detect until it’s too late. But it’s highly curable if caught early. That’s where the dogs come in, and not just any old mutt, but a very special kind. Dr. Otto’s team at Penn Vet uses three breeds with long histories of using their noses for work — a lab, a springer and a shepherd. Beyond pedigree, she says the difference between a good dog and a great dog is his ability to focus.

“We don’t know exactly which [breeds] are the best,” she says. “Dogs have such incredible ability to smell, but what we’re asking them to smell is an odour that’s present in a single drop of blood. We find that the dogs that do well are those that can focus so incredibly that everything else disappears. They’re Zen master dogs; they can really get in that zone and pay attention. You can see the little wheels turning: ‘It’s not this, it’s not that, oh there it is, I got it!’ It’s a very subtle odour that these dogs are finding. It may be that other [studies] that use urine, where there might be more of [the sample], it’s not so hard for [those dogs]. But for us, we’re looking for the super sniffers, like the people who do wine tasting or fragrance sniffing.”

These furry connoisseurs don’t work cheap: Dr. Otto says it costs $36,000 to train just one dog (Tracie Nielson says it’s closer to $50,000 for them) and money is elusive. Unlike the deep pockets up north, Dr. Otto’s team works without grants or subsidies, relying instead on funding from smaller foundations. “I think this is a little outside the norm, so people are having a bit of trouble deciding if this is a fluke or something real,” she says. “Certainly our evidence from all aspects — the dogs, the chemistry and the nano-sensing — says it’s real.”

There are also red-tape hurdles to overcome. Arlene Weintraub says regulatory agencies want certainty that results will be right most of the time. “Ninety per cent accuracy is great but the fear of false-positives is a concern to regulatory agencies. Still, I’m definitely seeing the advances being made. Technology is getting better and better every day, and there are a lot of smart people working on this problem.”

But smarts mean nothing without the funding to apply them. Dr. Otto says if she and her team had access to more cash, it would allow them to focus more fully, to devote more hours a day, which within two years would lead to a prototype machine. “We have the proof of principle, we know we can do this. We now need to identify the chemicals so we can help develop the electronic sensing, find out how early we can detect cancer, and whether we can use other samples besides blood. There are so many questions that we would love to be able to answer.”

Her ideal goal is an “e-nose” machine affordable enough to install in every hospital, which would provide instant test results. “My wish would be that every woman could be screened at least annually and we wouldn’t have these cases that are missed.”

Back at Clever Canines, Tracie Nielson and her pack are on track to fulfill their own wish for an early cancer detection facility in less than a year, using shiny, wet noses instead of cold, hard steel.