CAD machine, courtesy of the Cleveland Clinic
X-Ray, courtesy of Cleveland Clinic
Lung shown on a CT scan (courtesy of Cleveland Clinic)
Lung cancer is the leading cause of mortality from cancer in the United States, accounting for an estimated 160,000 deaths in 2012.
When caught early, the five-year survival rate is 52 percent, but because there are not good screening tests to catch lung cancer early enough, the average five-year survival rate in the U.S. is only 17 percent.
X-rays are not a great screening tool for those with an increased risk of lung cancer because nodules are too small to detect in the earliest stages. This is because they are too small or because they are obstructed by ribs. One large study, the National Lung Screening Trial, found that screening with low-dose CT scans reduced lung cancer deaths by 20 percent.
But CT scanning exposes patients to high doses of radiation. A low-dose chest CT scan exposes patients to 15 times more radiation than a chest X-ray.
CT scans also result in a high number of false positives, leading to unnecessary follow up x-rays, more CT scans and possibly biopsies, not to mention the anxiety that comes with hearing you have a “suspicious” finding. And they’re expensive, costing an average of $1,800, compared to a chest X-ray, which costs an average of $370.
One group of researchers from the Cleveland Clinic are exploring how to improve the sensitivity of X-rays so it can be a viable screening tool without all the drawbacks of CT scanning. The group recently published a study in the online journal PLoS ONE, examining the use of X-rays enhanced with computer aided detection (CAD). CAD can enhance images to better highlight the findings, and can digitally remove ribs, allowing pathologists to get a better view of the lungs.
“The hope is that this specialized kind of X-ray might improve the ability over a standard X-ray to find cancers and minimize some of the drawbacks of CT scans,” said the study’s lead author, Dr Peter Mazzone, director of the Cleveland Clinic Lung Cancer Center.
In the study, half of the patients who were at high-risk for lung cancer got chest X-rays and half did not. A radiologist would read the chest X-rays first, then a radiologist would read it with the help of the CAD system.
“They were able to identify more lung nodules when using the CAD system,” Mazzone said. The CAD system found 14 nodules that weren’t caught by the first radiologist. There were two cancers found and both were also spotted by the radiologist without using the system, so it’s difficult to conclude that the CAD system caught any cancers that were missed by x rays alone. Though CAD found nodules that were not cancerous, the number of these “false positives” was much lower than the number typically found with CT scans.
CAD systems are improving, and the hope is they will become sensitive enough to detect even smaller nodules, but more research is needed.
Mazzone has launched a new study comparing CAD X-rays to CT scans.
Laurie Tarkan is an award-winning health journalist whose work appears in the New York Times, among other national magazines and websites. She blogs about the Affordable Care Act for the WellBeeFile. Follow her on Twitter and Facebook.