June 1 (UPI) -- Scientists have developed a new electronic nose capable of sniffing out the molecular signatures of hard-to-detect pancreatic and ovarian cancers in the vapors emanating from blood samples.
The novel odor-based test relies on a machine-learning algorithm to identify unique combinations of volatile organic compounds, or VOCs, in the vapors wafting from blood plasma samples.
In the lab, the technology was able to detect and identify elusive cancer cells with 95 percent accuracy.
Researchers are scheduled to detail the technology's diagnostic capabilities on June 4 at the annual meeting of the American Society of Clinical Oncology.
Though the odor-test technology is in its infancy, its developers suggest the device has great potential.
"The data shows we can identify these tumors at both advanced and the earliest stages, which is exciting," lead researcher A. T. Charlie Johnson said in a news release.
"If developed appropriately for the clinical setting, this could potentially be a test that's done on a standard blood draw that may be part of your annual physical," said Johnson, a professor of physics and astronomy at the University of Pennsylvania.
Johnson and his colleagues are currently working with the company VOC Health to ready the device for commercial production.
All cells give off VOCs, and previous studies have shown cancerous cells release VOCs distinct from the compounds emanating from benign tumor cells.
To build their device, researchers calibrated nanosensors to detect these VOCs.
The technology relies on artificial intelligence to distinguish between problematic and benign combinations of volatile organic compounds.
After training the machine learning algorithm, researchers tested their e-nose on actual blood plasma samples.
Scientists used the odor-test to analyze the blood of 20 patients with ovarian cancer, 20 with benign ovarian tumors and 20 cancer-free volunteers.
They also tested blood samples from 13 patients with pancreatic cancer, 10 patients with benign pancreatic disease and 10 controls.
The e-nose was able to distinguish between ovarian cancer and benign tumors with 95% accuracy, and detected pancreatic cancer cells at 90% clip.
Impressively, the electronic nose detected several instances of early stage cancer.
"Collaborating with researchers from the department of physics and astronomy, the Perelman School of Medicine and Penn Vet has allowed us to perfect and integrate our own innovations -- expediting the commercialization process," said Richard Postrel, CEO and chief innovation officer of VOC Health.
"Initial prototypes of commercial devices able to detect cancer from liquids and vapors will be ready soon and be provided to these Penn researchers to further their work," Postrel said.
Similar VOC-sniffing devices have been used to detect everything from disease-carrying dogs to COVID-19 particles. Researchers have also developed breathalyzers to detect disease-related VOC signatures in the exhaled vapors of patients.