Using sophisticated techniques to sample and analyze airborne molecules in the odors from human skin cells, scientists in the US were able to detect a unique chemical signature for melanoma, the deadliest form of skin cancer.

They suggest their methods, which can spot the difference between different types of melanoma cell and normal skin cells using odor signatures, could lead to a new non-invasive way to diagnose human melanoma in the early stages.

In their study they also show how a nano-sensor device could lead to a portable clinical tool for non-invasive odor detection of melanoma.

The team, from Monell Chemical Senses Center and the University of Pennsylvania, both in Philadelphia, write about their findings in a report published online this week in the Journal of Chromatography B.

Melanoma is a cancer of melanocytes, the pigment-producing cells of the body that give skin its colour. It is responsible for three-quarters of skin cancer deaths, but the earlier it is detected the better the chances of survival.

Currently, early detection relies on visual examination, depending heavily on the ability of individuals to examine themselves and the skilled eye of their doctor.

There is already evidence that dogs, with their highly sensitive sense of smell, can sniff out human diseases. For instance, in 2010, researchers told an annual scientific meeting how dogs can be trained to identify prostate cancer from volatile organic compounds (VOCs) in urine.

Scent-detecting dogs can even sniff out hospital infections from samples of feces and in the air around patients.

George Preti, an organic chemist at Monell, and co-author of this new study, says in a statement:

“There is a potential wealth of information waiting to be extracted from examination of VOCs associated with various diseases, including cancers, genetic disorders, and viral or bacterial infections.”

In their background information, Preti and colleagues note that dogs can also identify melanoma on the skin of patients or on melanoma samples hidden on healthy people. This suggests that VOCs from melanoma are different from those of normal healthy skin.

For their study, the team used advanced sampling and analytical methods and worked with culture-grown cells, to pick out VOCs from melanoma cells at three stages of the disease, plus VOCs from normal melanocytes.

First they put the cells in sealed containers and used an absorbent device to capture the VOCs they gave off.

Then they used gas chromatography-mass spectrometry (GC-MS) to analyze and look for differences in the mix of VOCs captured from melanoma cells relative to normal cells.

They found differences in concentrations of certain VOCs between melanoma and normal cells, and they also found compounds in the VOC mix of melanoma cells that were not present in that from normal cells.

In fact, the researchers could even see differences between different types of melanoma cells from their VOC signatures.

Up to this point in the study, they proved it was possible, using lab-based methods, to distinguish between melanoma cells and normal cells, and even among different types of melanoma cell using the cells’ VOC signatures.

But to make a tool that is useful in clinical settings, the researchers needed something portable. So they turned to a nano-sensor they had used before.

The nano-sensor is made from nano-sized carbon tubes coated with strands of DNA. It can be bioengineered to recognize a range of targets, including specific odor molecules. It could also be turned into a portable device.

When the tested it, the researchers found like their lab-based GC-MS tool, that the nano-sensor could tell the difference between VOCs emitted by melanoma cells and VOCs emitted by normal scells. And it could also distinguish different types of melanoma cells from their VOC signature.

A.T. Charlie Johnson, a professor of physics at University of Pennsylvania, whose lab developed the nano-sensor, says:

“We are excited to see that the DNA-carbon nanotube vapor sensor concept has potential for use as a diagnostic. Our plan is to move forward with research into skin cancer and other diseases.”

The findings are proof of concept that both the lab-based GC-MS and the nano-sensor, which could lead to a portable clinical tool, have the ability to spot biomarkers to distinguish normal melanocytes from different types of melanoma cell.

Preti says the study also shows how detecting VOC signatures could be a rapid and non-invasive way to test for a range of diseases.

“The methodology should also allow us to differentiate stages of the disease process,” he adds.

Written by Catharine Paddock PhD