Listen to this story.

Penny

Photo by Shelby Wise/Wise K9 Photography

A biodetection dog trained at the Penn Vet Working Dog Center in Philadelphia, Penny participated in a study to demonstrate whether blood samples from patients with canine hemangiosarcoma have a detectible odor that would indicate a disease-specific chemical signature.

Rookie's was a classic case. The 8-year-old German shepherd appeared in tip-top form, showing off her police dog skills during a public demonstration at the Penn Vet Working Dog Center, which trains sniffer dogs. That same week, Rookie collapsed and was rushed to a veterinary hospital. She would never work again.

The diagnosis was hemangiosarcoma, a cancer of blood vessel cells characterized by internal bleeding and aggressive spread. Rookie died four days later.

To any dog lover who's encountered HSA — and there are many — the short span from diagnosis to death is painfully familiar. Notorious among cancers in dogs, HSA develops quickly and under the radar, yielding no obvious clinical signs until it reaches a catastrophic point.

Hence, a dog that seemed hale yesterday could be gravely debilitated by HSA tomorrow. "It's devastating," said Dr. Cynthia Otto, a former emergency room practitioner who runs the working dog center, part of the University of Pennsylvania School of Veterinary Medicine. Owners, Otto said, often think, "My neighbor must have poisoned my dog, because there's no way .... He was perfectly fine five minutes ago."

As a blood cancer, HSA can arise anywhere in the body but usually manifests on or just beneath the skin, in the liver, the heart or, especially, the spleen, according to Dr. Douglas Thamm, an oncologist at the Colorado State University veterinary school and president of the Veterinary Cancer Society. Comprised of "bizarre, malformed blood vessels and prone to tear," the tumors that occur internally are the more difficult to manage, he said.

Even with surgery and chemotherapy, the median survival time for a patient with a ruptured splenic tumor is about six months, with the likelihood of surviving a year less than 10%. An estimated 500,000 to 2,500,000 dogs in the United States develop HSA during their lifetime. (HSA occurs in cats, too, but less so.)

Having seen "way too much hemangiosarcoma" in dogs during 20 years in the ER, not to mention in multiple detection canines trained at the center, Otto, with colleagues, decided to apply their expertise in scent training to the insidious, lethal disease.

Their research found evidence that the blood of dogs with HSA emits a distinctive smell that can be perceived by the superior noses of dogs. The findings are described in a paper published online in November and appearing in the February issue of The Veterinary Journal.

Though it's tempting to imagine "Dr. Dog" being deployed to diagnose fellow canines, the scent dogs' role here is strictly in the research setting — to determine whether the disease has a chemical signature, which it appears to. Next steps are to identify the chemical or combination of chemicals present in afflicted dogs and develop an inanimate screening tool.

"These dogs are amazing biosensors," said Clara Wilson, the study leader, who has a doctorate in applied canine olfaction and behavior. "But we don't necessarily see a path where dogs are doing it on a larger scale ..." she said. "The goal is to encourage the people who are working on the types of machines that could look for those markers."

Otto and others, including a University of Pennsylvania physicist, are founding members of a company aimed at creating an "e-nose," or electronic nose, with multiple applications in disease diagnosis. Canine HSA is not the first illness that dogs have demonstrated an ability to smell. Others, almost all in the human medicine realm, include ovarian cancer, pancreatic cancer, prostate cancer, malaria and Covid-19, to name a few.

Several important questions have yet to be answered before the canine HSA research can translate into a screening tool. For one, the positive blood samples presented to the detection dogs were from patients known to have HSA, presumably at its later stages, since a means of early detection doesn't exist. Accepting that the disease has a chemical signature, at what point in its progression is it detectable?

To answer that, Wilson said, the researchers will tap so-called longitudinal samples, which are drawn from dogs being followed over an extended period.

For the initial study, the question at hand was more fundamental: Does hemangiosarcoma have a singular scent?

Dogs smelled a possible diagnostic error

Sussing out the answer were five dogs who were first trained to distinguish between HSA-positive blood samples and odorless air samples. The dogs earned a food reward if, upon sniffing the positive samples, they gave an "alert," keeping their nose in the container housing the sample for several seconds.

Once trained to alert for HSA, they moved on to sniffing samples from healthy and sick dogs, including dogs with noncancerous diseases, along with HSA-positive samples. For the data used in the final analysis, the tests were double-blinded, meaning that no person in the room with a dog being tested knew which containers had an HSA sample and which did not. (The computerized system knew which samples were positive, however, and sounded a tone for each correct answer, which told the dog that it would receive a treat.)

Across 423 blinded trials, the dogs' collective accuracy was 70%.

And it's possible the accuracy was slightly better than that. With one particular sample deemed HSA-positive, the dogs had a lot of trouble, identifying it as positive only 39% of the time. The unusually high failure rate led the researchers to follow up with the sample provider. They learned that the diagnosis of splenic HSA was, in fact, under review, owing to a difference of interpretations among several pathologists. A final determination is pending, with the "likelihood that the HSA classification will be overturned," the study authors wrote.

If that happens, it means the dogs who did not identify the sample as HSA-positive weren't wrong. Some had steadfastly maintained their answer, despite not receiving a reward for their decision, Wilson recalled, amazed.

"I was basically giving them all negative line-ups time after time after time," she said, explaining that each dog was presented with three samples matched for age and sex — one healthy, one diseased and one supposedly HSA-positive. "Some of them were saying, 'I don't think this is right, but if you keep telling me it is, I can switch,' you know, within that learning period. But some of them were like, 'This is definitely not what I've been taught to find.' "

The chemical or chemicals comprising what the dogs appear to be detecting are believed to be volatile organic compounds, or VOCs. Exploring VOCs as targets for disease detection is "a really hot topic in human medicine right now," Wilson said. "We hope that this interest in VOCs can now begin to extend into veterinary medicine, too."

Perhaps more commonly known as pollutants, VOCs are a broad group of organic chemicals that evaporate easily at room temperature. VOCs are the basis of scents. Some are human-made, toxic compounds, and some arise naturally from metabolic processes.

It is the latter property that medical science is delving into. For example, a paper published in 2025 refers to VOCs in exhaled breath, skin, stool and urine, as well as blood, and the potential to diagnose and monitor Parkinson's by analyzing VOCs.


Lulu

Photo by Dr. Katie Linderman

Dr. Katie Linderman saw a dismaying spate of canine patients in 2018 afflicted by hemangiosarcoma, including her own dog, Lulu, who died at age 10.

Pursuing answers on all fronts

Beyond the hunt for its signature scent, HSA is the subject of an array of research efforts to improve understanding of the heartbreaking disease. One organization giving a concerted push to the science is the nonprofit Morris Animal Foundation, which in 2023 started the Hemangiosarcoma Initiative to raise money for research on preventing, detecting, treating and, potentially, even curing the condition.

The campaign has raised $1.9 million, bringing to $3.3 million the total the foundation has put toward HSA, according to Dr. Kelly Diehl, senior director of science communications at the foundation. Grants funded by the initiative have gone to 12 projects, including an investigation into the genetics of dogs with HSA; studies of markers in and on HSA cells as targets for therapy; and further work by the Penn Vet Working Dog Center team on early detection.

The foundation also supported that team's initial work by providing blood samples from its Golden Retriever Lifetime Study. The purpose of that study, begun in 2012, is to identify the nutritional, environmental, lifestyle and genetic risk factors of diseases in dogs.

Diehl said cancer kills six out of 10 golden retrievers, and HSA accounts for two-thirds of the cases. She is optimistic that the attention given to the disease will produce welcome results. "I feel we're on the cusp of making some big advances, because whenever you throw a lot of resources [at a problem], that usually moves the needle," she said.

To Dr. Katie Linderman, progress on HSA can come none too soon.

The veterinarian encountered a rush of cases in 2018, leaving her feeling desperate. A solo practitioner in Tennessee at the time, Linderman posted to a message board of the Veterinary Information Network — an online community for the profession and parent of the VIN News Service — in hopes that colleagues could point her to earlier detection, more effective treatment or, best yet, a means of prevention.


"What will block the formation of this neoplasia?" she asked, in vain. "So frustrating ... I have at least 10 canine patients who have died or are dying this month ... "

The sense of helplessness became acutely personal when her own dog became Linderman's 11th case during that terrible period. Lulu was 10, a mix of maybe pit bull and Labrador, and a pure sweetheart. The first signs that something was amiss with Lulu were excessive drinking and urinating and being slower to get up and lie down. "She suddenly seemed much older," Linderman said.

That was in August that year. By October, Lulu was diagnosed with HSA and had surgery to remove her cancerous spleen. She did not survive long enough to complete follow-up chemotherapy, dying in Linderman's arms at Thanksgiving in a seizure.

More than seven years later, the ache of loss lingers. "I'm kind of scared to get as close to a dog as I was with Lulu because of how painful it was," said Linderman, who is understandably eager to see the research advance. "Anywhere they get with this will be better."