Using low-dose CT scans to screen high-risk patients for lung tumors is far more effective at preventing lung cancer deaths than scanning those at low risk, according to a new analysis of over 53,000 volunteers.
The study, published in the New England Journal of Medicine, comes at a time when doctors are struggling to reduce the death rate among lung cancer patients, who account for more than one quarter of all cancer deaths.
An earlier analysis from the National Lung Screening Trial showed that compared to conventional X-rays, screening with low-dose CT scans cut the overall death rate by 20 percent.
The new study, which used the same data, was designed to see whether the risks and benefits of screening varied by a person's likelihood of developing lung cancer in the first place.
Researchers divided volunteers into five groups based on their risk of being diagnosed in the next five years - anywhere from about a 0.5 percent risk or less to a more than 2 percent risk - based on a list of risk factors and individual traits.
Everyone in the study was between age 55 and 74, had smoked within the last 15 years and had a history of smoking the equivalent of at least one pack per day for 30 years.
Among participants with the lowest risk, CT screening prevented only one additional lung cancer death per 50,000 people each year, compared to screening with conventional X-rays.
But in the highest risk group, CT screening saved about one additional life per 800 people.
Coauthor of the report Hormuzd Katki said his group is working to release the risk formula they developed for the study so doctors can determine who falls into the high-risk category and is most likely to benefit from such screening. People's family history of lung cancer and how much they've smoked both go into the equation, for example.
"You can predict the benefit from screening and the harms you might get from screening," Katki, from the National Cancer Institute, told Reuters Health.
The researchers estimated that 161 people in the highest-risk group would need to be screened to prevent one lung cancer death. But there would also be 65 false positives per life saved, and those people would require needless follow-up testing.
In the low-risk group, 5,276 people would have to be screened with low-dose CT to prevent one cancer death, and 1,648 would be told there was something suspicious on their test when there was, in fact, no tumor.
Dr. Otis Brawley, chief medical officer of the American Cancer Society, who was not involved in the research, said the new study "tells you that among people who are at low risk or no risk for lung cancer, the chance that you will benefit from this screening test is pretty small."
He told Reuters Health that it's important to tease out who will benefit because some hospitals - he cited one in Atlanta - have already advertised such CT scans at a deep discount.
"According to them, people at risk for lung cancer include non-smoking 40-year-old women who have lived in an urban area for 10 years," said Brawley.
Even though such women would have very little risk, "they're advocating for this stuff so the hospital can make money" off of the follow-up testing.
The good news is that most follow-ups in the study did not require potentially-dangerous invasive testing, the researchers found.
"On the first screen, you'll see something suspicious. They'll say come back in a year, and if it's still there or has grown, only then will you do something invasive," said Katki. "In most of those false positives, you come back the next time and it's gone."
The original National Lung Screening Trial showed that for every five to six lives saved by screening, one person died as a result of post-screening procedures, Brawley said. That included people who had a needle biopsy that collapsed their lung, for example.
The study did not assess the costs of routine CT screening. Each test typically costs a few hundred dollars.
In addition, the study did not assess the potential impact of false positive results, the likelihood of radiation-induced cancers or the complications that can come from further testing to check on a suspicious result.