Small or slow-growing nodules discovered on a lung scan are unlikely to develop into tumors over the next two years, researchers reported on Wednesday.
The findings, reported in the New England Journal of Medicine, could help doctors decide when to do more aggressive testing for lung cancer. They could also help patients avoid unnecessarily aggressive and potentially harmful testing when lesions are found.
Lung cancer, the biggest cancer killer in the United States and globally, is often not diagnosed until it has spread. It kills 159,000 people a year in the United States alone.
The work is part of a larger effort to develop guidelines to help doctors decide what to do when such growths, often discovered by accident, appear in a scan.
High-tech X-rays called CT scans can detect tumors -- but they see all sorts of other blobs that are not tumors, and often the only way to tell the difference is to take a biopsy, a dangerous procedure.
Tested guidelines for dealing with the nodules do not exist, said Dr. James Mulshine of Rush University Medical Center in Chicago and David Jablons of the University of California San Francisco Cancer Center, in a Journal editorial.
Good guidelines could help make lung cancer screening practical, Dr. Rob van Klaveren of the Erasmus Medical Center in Rotterdam, the Netherlands, who led the new study, said in a telephone interview.
At the moment, routine lung cancer screening is considered impractical because of its high cost and because too many healthy people are called back for further testing.
All these recall CT scans give rise to a lot of anxiety, said van Klaveren.
The team looked at 7,557 people at high risk for lung cancer because they were current and former smokers. All received multidetector CT scans that measured the size of any suspicious-looking nodules.
Volunteers who had nodules over 9.7 millimeters in width, or had growths of 4.6 millimeters that grew fast enough to more than double in volume every 400 days, were sent for further testing. Of the 196 people who fell into that category, 70 were found to have lung cancer; 10 additional cases were found years later.
But of the 7,361 who tested negative during screening, only 20 lung cancer cases later developed.
In a second round of screening, done one year after the first, 1.8 percent were sent to the doctor because they had a nodule that was large or fast-growing. More than half turned out to have lung cancer.
The result means that if the screening test says you don't have lung cancer, you probably don't, the researchers said. The chances of finding lung cancer one and two years after a negative first-round test were 1 in 1,000 and 3 in 1,000 respectively, they concluded.
The study is part of a larger project, known as NELSON, designed to see if a screening program can, over the long term, cut lung cancer death rates by 25 percent. Final results are expected in 2015.