The National Cancer Institute is putting all of its lung cancer screening eggs into one very large basket. The eight-year, $200 million National Lung Screening Trial (NLST), aimed at finally determining whether screening for the disease holds promise, has drawn both praise and criticism.
The multicenter randomized controlled trial is intended to assess whether screening with low-dose spiral CT or chest x-ray will be better at reducing lung cancer deaths for people at risk. Researchers plan to recruit 50,000 current or former smokers, aged 55 to 74, over the next two years.
Opponents question the use of chest x-ray, which they say has been shown -- in other NCI-supported trials -- not to have any effect on mortality from lung cancer.
"If one were to be intellectually and scientifically rigorous, one would have to insist on a randomized controlled trial of CT scan against current medical practice, [which is] to diagnose and treat only after symptoms have developed," wrote Dr. Frederic W. Grannis Jr. in an editorial in the July issue of Chest.
Grannis, a thoracic surgeon at City of Hope National Medical Center in Duarte, CA, is a principal investigator in that institution's participation in the International Early Lung Cancer Action Program (I-ELCAP), a nonrandomized CT-only screening trail.
Dr. Claudia I. Henschke, ELCAP's principal investigator, has been a vocal opponent of lengthy randomized lung cancer screening trials. She voiced her concerns about the NLST before Congress in April.
"Ethically, I couldn't ask people at risk of lung cancer to undergo chest x-ray, because I know definitively from our study that CT picks up six times more small stage I lung cancers, those that are curable, than chest x-ray," Henschke told Diagnostic Imaging.
Proponents of the study, however, point to the seemingly conflicting findings of previous randomized controlled trials of lung cancer screening. Although these trials have demonstrated increased patient survival, at the same time they've shown no reduction in lung cancer mortality, according to NCI epidemiologist Pamela M. Marcus, Ph.D.
Randomized control, the gold standard in trial design, is the only way to definitively answer which screening test is better at lowering lung cancer death rates, said Dr. Denise R. Aberle during a recent media briefing. Aberle is chief of thoracic imaging at the University of California, Los Angeles and coprincipal investigator of the NLST.
Participants in the NLST will be randomly assigned to either the CT arm or the chest x-ray arm. They will receive an annual screening for three years and be followed for a total of eight years. The number of lung cancer deaths will be monitored year by year. Because each group has equivalent risks, the differences in lung cancer deaths can be attributed to the screening test, Aberle said.
"The single most important thing we are analyzing is the difference between the CT and the chest x-ray arms in the number of deaths from lung cancer," she said.
Other trials that have investigated lung cancer and chest x-ray and found no statistically significant decrease in death rates were designed to detect a 50% reduction in lung cancer mortality. Aberle and coprincipal investigator John K. Gohagan, Ph.D., believe the effect may be smaller than that. Consequently, NLST was designed to be large enough to determine if the difference in lung cancer mortality between the two screening modes is 20% or greater.
Henschke believes the NLST is seriously flawed because it focuses on overall mortality rather than on case-fatality during the time when the screening shows a benefit. This is the same flaw that resulted in the incorrect conclusion that there was no benefit to mammography screening for breast cancer, she said.
But Aberle stands by the randomized trial design. She insists that survival and case-fatality are appropriate outcome measures for treatment trials but do not help to determine the benefits of screening.
The best way to address this controversy is through a randomized, controlled trial, wrote Dr. Stephen J. Swenson of the Mayo Clinic in Rochester, MN, in the October issue of Radiology. Small studies at Mayo suggest that total-body screening is lifesaving and cost-effective.
"Scientifically, however, such a conclusion is absolutely unproven at this point, and the matter needs to be thoroughly studied in a responsible manner," Swenson said.
Another NCI randomized controlled trial, the Prostate Lung Colorectal and Ovarian cancer screening trial (PLCO), has finished recruiting 155,000 participants. The chest x-ray test was designed to detect a less than 20% effect in reducing mortality. If this trial finds no mortality reduction with chest x-ray, then the NLST will effectively be comparing CT with usual care, Gohagan said.
For more information from the Diagnostic Imaging archives:
