Dogs can potentially sniff out prostate cancer, French researchers say

Man's best friend may cement his position if early results from French researchers can be replicated. A team of researchers from Tenon Hospital in Paris reported Tuesday at a San Francisco meeting of the American Urological Assn. that dogs can be trained to detect the characteristic odor of unique chemicals released into urine by prostate tumors, setting the stage for a new way to identify men who are most at risk from the cancer. If developed, the test might be more effective than the PSA test now used because it would have fewer false positives.

As surprising as the idea might sound, other researchers have already been studying the use of dogs to detect cancers of the breast, lung and bladder. Many tumors release characteristic chemicals that can be identified by the exquisitely sensitive canine nose. Lung cancer cells, for example, can release such chemicals into the air of the lungs, and they can then be detected on the victim's breath.

Dr. Jean-Nicolas Cornu of Tenon and his colleagues trained a Belgian Malinois — a shepherd breed that has already been used for detecting bombs and in other cancer tests — to identify urine from patients with confirmed prostate cancer, then to differentiate those samples from urine from healthy subjects. Finally, they used one urine sample from a prostate cancer victim and four samples from healthy people, asking the dog to choose the correct one. In 66 tests, the dog was correct 63 times. There were three false positives and no false negatives. That is, the dog correctly identified all the specimens from prostate cancer patients, but misidentified three from healthy men.

The whole training process took about a year, Cornu said, and the team is already training other dogs. The researchers are now attempting to identify what specific chemicals the dog is reacting to in hopes of developing an "electronic nose" that wouldn't require treats and potty breaks.

— Thomas H. Maugh II

Comments

[woof-woof]

"There were three false positives and no false negatives. That is, the dog correctly identified all the specimens from prostate cancer patients, but misidentified three from healthy men."

You sure about that? In the abstract, it states, "In the final phase, dogs were asked to signal a cancer urine among 5 samples containing only one cancer urine." Given that procedure, if the dog picks the wrong urine, isn't that both a false positive (picked a non-cancer sample) and also a false negative (missed the cancer sample)?

Here is the meeting abstract:
http://www.jurology.com/article/S0022-5347%2810%2902517-6/fulltext

[Lung Cancer Stories]

We make sure to provide you the best options for how to survive lung cancer. If you or someone close to you has been diagnosed with lung cancer, then this would be the right place for treatment of lung cancer.

[Jeff Kreisberg]

Sounds like a better way to diagnose prostate cancer than PSA.

[Tee]

Oh . . . is THAT why dogs always have their noses up people's rear end?

[Eve]

Amazing.

[Teeisretarded]

@Tee

Keep your day job, you'll never make it as a comedian.

If you actually took the time to read, you'd know that it's via smelling people's urine.

[RPL]

...The researchers are now attempting to identify what specific chemicals the dog is reacting to in hopes of developing an "electronic nose" that wouldn't require treats and potty breaks.

TRANSLATION: The medical community needs to create "an electronic nose" so they can charge you a $1000 or more for conducting a test that dog could do for free.

[woof-woof]

They could charge even more if they trained a CAT to do that particular Scan.

[donut]

They can also sniff out cookies a mile away.

[Kevin]

RPL - you are a moron. "TRANSLATION: The medical community needs to create "an electronic nose" so they can charge you a $1000 or more for conducting a test that dog could do for free."

I take it you don't own a dog and are ignorant to the extremely large amounts of time, effort, expertise, food, etc... that comes with a dog. A dog would not be free, by any means. Pay a very, very specialized trainer to train a special breed of dog for a year. That alone would run hospitals $50,000 for the dog. And probably even more!

Think things through before making ignorant remarks.

[R'Chard]

I suppose a cat could be trained, too ... but would it tell you? Meh...