For me, the concept of dogs sniffing breath samples to detect human cancers early has been the gift that keeps on giving—actually, for about 20 years. Blogs, media interviews, national television news appearances: you name it: it’s a topic that has engaged me periodically for that many years. It is the story that will not go away.
And despite all that continuing interest for me and the larger, more sophisticated scientific community the reality is that so far, it hasn’t worked.
So imagine the deja vu moment I had when reading a recent research report in JAMA Network Open from researchers in China where they claim to have developed a technology that is incredibly sensitive and reasonably accurate in detecting the presence of lung cancer by analyzing breath samples. Maybe this time it might work. Hope springs eternal.
For those interested in the early detection of cancer the opportunity offered by using this type of breath-sensing technology has been a holy grail, although history suggests the journey has been more akin to a rocky, pot-hole filled city street. As appealing as this technology may appear, it simply hasn’t resulted in an effective clinically useful early detection test despite literally decades of effort.
The principle would appear pretty straight-forward, based on reports from the early years of this century: dogs were reportedly able in various studies to distinguish breath samples from patients with cancer compared to those without cancer. And even in some circumstances, they detected cancer recurrence well before it was clinically evident with then-available technologies.
There was a lot of excitement back then about the prospect of your friendly neighborhood pooch being turned into a fearsome cancer-detecting machine. Really: the dogs trained in at least one study were in fact your local home pet, not some specially bred German Shepherd or beagle.
It became apparent pretty quickly that the probable reason these dogs were able to tell the difference between breath samples from patients with and those without cancer was their highly developed sense of smell, far better than what humans can detect.
You can appreciate that lining up bags of breath samples and having pooches smell them for cancer wasn’t at first glance exactly a very sophisticated way of finding cancer, not to mention improbable when it comes to scientific rigor. So off to the lab went the scientists to use their knowledge and fancy machines to try to mimic what the dogs were smelling. I imagine you won’t be surprised to learn that the dogs came out ahead: numerous studies have been performed, but no technology has emerged which has been sufficiently accurate to use in the clinic.
Time has marched on, and so has science. Now we are much more interested in finding biologic markers/proteins in consumers’ blood samples as a means of detecting cancer early and breath samples have taken a (relative) back seat. However, the idea remains intriguing especially given the prospect that it if it did work it would be easy to collect the samples and the analyses would be much less expensive than those currently proposed using circulating tumor DNA or other proteins in the blood.
The most recent credible research employing breath-sample analysis that I am aware of is being run by Cancer Research UK, in—of course—the United Kingdom. Although this has been ongoing for several years, I am not aware of any recent results reported, so don’t know if the trial was successful in detecting lung cancer or other cancers in those at high risk without a current diagnosis of the disease. Based on timing and the fact the trial is now closed to new participants, I would suspect some information will be available soon from CRUK and the company that developed the technology.
Meanwhile, back to that JAMA research report:
The study was performed in China using a well-defined breath sample collection process which basically involved blowing a deep breath into a collection device, then analyzing the breath with sophisticated scientific instruments. The researchers looked at two groups of patients, those with documented lung cancer, and those without. The test was able to detect 100% of the patients with lung cancer (which itself is pretty remarkable) when the test was analyzed “blind,” and was about 96% accurate in discriminating between those with and without lung cancer. Those are pretty incredible numbers.
One needs to keep in mind that this was not a blind screening test, meaning the researchers knew the status of each of the participants. That is not the same as a real world test, where this approach could theoretically be used on millions of people around the world who may be at risk of lung cancer, but who have not had signs of symptoms of lung cancer.
That is the real test of any cancer early detection technology, and this test didn’t go there—yet. No doubt that concept is currently undergoing investigation as I write this.
Time has been a great teacher when it comes to analyzing breath samples for the early detection of cancer. There has been an incredible amount of interest and scientific research directed towards this very intriguing concept that cancers give off small amounts of what we call “volatile organic compounds” which are exhaled in our breath and can be detected by dogs. Despite all of this effort and promise (even a presentation at the American Society of Clinical Oncology several years ago mentioning this approach as promising for the early detection of cancer, nothing has come close to being clinically useful.
Whether the technologies now reported by the Chinese researchers, or the work being done by CRUK, or other research by investigators around the world will bear fruit and open the door to an inexpensive, reliable and repeatable approach to the early detection of cancer through analyzing breath samples remains to be seen.
Today, analyzing breath samples to find cancer early and save lives remains a tantalizing possibility, not a probability. It is 20 years from the time this saga began, about 140 dog years—and yet it still continues to intrigue the scientific community and the public at large.
Time will tell us whether our doggies are destined to become the model of a cancer fighting machine—or just our dear four legged companions that want more belly rubs. No matter: we love them much!