Breakthrough test can screen for lung cancer using breath

Talk about a breath of fresh air.

Researchers have developed a groundbreaking device that may one day make detecting lung cancer as easy as exhaling.

“We built a screening tool that could allow physicians to catch the disease in its early phases, which improves outcomes,” Dr. Shalini Prasad, a professor at the University of Texas at Dallas, said in a press release. 

That’s no small feat. Lung cancer remains the deadliest cancer in the US by far, claiming about 350 lives every day — more than colon, breast and prostate cancers combined.

And yet, only about 30% of cases are detected early, largely because the disease often causes no symptoms until it reaches advanced, harder-to-treat stages.

Doctors currently rely on low-dose CT scans to screen for lung cancer, but those tests are generally reserved for high-risk patients, such as older adults with a history of smoking.

They also come with drawbacks. The scans are costly, expose patients to radiation and carry a high false-positive rate, which can trigger unnecessary anxiety and often lead to invasive — sometimes risky — follow-up procedures to confirm whether cancer is actually present.

Now, a future where detecting lung cancer is as simple as taking a breath may be closer than ever.

“Breath has a lot of good stuff. It can tell you how sick you are, how your metabolism is working, how stressed you are,” Prasad said. But the challenge remains that breath has a lot of complexity in it.”

While breath is easy to collect, it’s difficult to analyze and monitor, she explained.

“So we built a sensor technology that is precise, unique and highly accurate in looking for eight biomarkers associated with lung malignancy or thoracic malignancy,” Prasad said.

Here’s how it works: Patients breathe into a tube connected to a sealed plastic bag. A small electrochemical biosensor then analyzes the sample for tiny molecules called volatile organic compounds (VOCs), which can signal the presence of cancer.

Next, the data is fed into an AI program that compare’s the sample’s chemical fingerprint against known cancer biomarkers to sniff out signs of disease.

And so far, the results are promising.

The team tested the device on 67 patients, including 30 with biopsy-confirmed thoracic cancer, and it correctly identified cancer markers in 90% of those cases.

“Eventually, this technology could be deployable in your primary care provider’s office,” Prasad said. “Just as you go in for an annual physical and give an annual blood draw, you could do a breath test as well.”

From there, doctors could flag any suspicious results for follow-up scans or specialist referrals, potentially catching the disease earlier and saving lives.

“The breath profiling device and associated machine learning model have great potential for making a difference in cancer detection while improving costs,” said Dr. Ovidiu Daescu, co-author of the study and a department head of computer science.

Looking ahead, Prasad said that her team will continue working to improve the device and test it in real-world medical settings.

The stakes are high: The American Cancer Society estimates that in 2025 alone, roughly 226,650 Americans will be diagnosed with lung cancer, and 124,730 will die from the disease.