A 1989 short letter in The Lancet describes a woman who became aware of a suspicious mole on her leg after her dog began sniffing at it daily and eventually tried to bite it off.1 The dog likely saved her life, as doctors then identified and treated her malignant melanoma early. This event also prompted scientists to hypothesize that tumors emit a specific odor. In the years that followed, researchers explored the use of sniffer dogs to detect cancer. Despite some success, using canines as a diagnostic tool remains limited—even well-trained dogs with exquisite noses cannot be standardized or scaled to meet clinical needs. They can, however, inform scientists’ development of electronic noses that sniff out cancer.

In a recent study published in International Forum of Allergy & Rhinology, a research team from the Monell Chemical Senses Center and Thomas Jefferson University Hospital in Pennsylvania, USA addressed the need for new minimally-invasive diagnostic methods to detect and track cancer based on volatile metabolite signatures—the unique odor profiles that tumors emit as waste products of abnormal metabolism.2 More and more studies are showing that certain cancers can be discriminated from healthy controls via the volatile metabolome using instrumentation or trained animal olfaction,” said Bruce Kimball, co-principal investigator of this study and a chemical ecologist at the Monell Chemical Senses Center.

Dogs under blanket together stock photo
Researchers are getting closer to creating cancer-detecting electronic noses that are comparable to those of dogs.

Kimball’s team studied individuals with and without sinonasal inverted papilloma (SNIP), a benign yet aggressive nasal tumor that can turn cancerous. They used a technique that analyzes the vapor in the space above a clinical sample to evaluate the odor signatures produced by these individuals. They collected plasma and nasal secretions from individuals with and without SNIP and compared the samples’ volatile metabolites. Kimball’s team identified a pattern of metabolites in the plasma and nasal secretions of SNIP patients that differed from that of healthy individuals. 

Surprisingly, the researchers found that the toluene concentration in SNIP plasma samples was lower that that of healthy participants. Higher toluene blood levels are associated with a higher risk of developing certain cancers and may indicate exposure through unhealthy activities such as smoking. “It would not have been all that surprising if toluene concentrations were higher in SNIP patients,” Kimball said. He hypothesized that abnormal tissue metabolism may use toluene as an energy source for growth, or that a unique protein made by tumor cells may bind free toluene. 

“It’s not intuitive, but it’s not without precedent,” Robin Couch, a biochemist at George Mason University in Virginia, USA who was not involved in this study said. “The pathology of disease can disrupt biochemical pathways or reroute metabolites into alternative pathways, resulting in the depletion of certain metabolites.” 

This study provides proof of principle for using volatile metabolite signatures to diagnose certain benign tumors before they become malignant. It also provides insight into the broader development of other sensing technologies, such as electronic noses, for non-invasively identifying and tracking a wide range of cancers. According to Couch, creating cancer-detecting electronic noses that are comparable to those of dogs is feasible. “The electronic nose is going to become a reality. The scent prints of a sample will be used much like a fingerprint as a unique identifier of disease.”

When asked how our loyal canine companions fare as diagnosticians compared to humans using instruments, Kimball said, “I have no doubt that in some cases, they would do better than my instrumentation. Dogs are looking at probably the same suite of volatiles that I am, plus a number of things that I can’t see. And I think humans, in many cases, can be just as good as dogs. We just don't train our noses to do such things.”


  1. H. Williams, A. Pembroke, “Sniffer dogs in the melanoma clinic?” Lancet, 1(8640):734, 1989. 
  2. M. Chaskes et al., “Unique volatile metabolite signature of sinonasal inverted papilloma detectable in plasma and nasal secretions,” Int Forum Allergy Rhinol, 2022 Epub ahead of print.
Research Resources Logo