These results fell in between more positive outcomes reported by the International Early Lung Cancer Action Program (I-ELCAP), and results from another model-based investigation showing that CT screening was not advantageous (Radiology 248:278-287, 2008).
“Smoking cessation should be the primary message to any continuing smokers,” lead investigator Pamela McMahon, PhD, told ONI. “Our analysis suggests that lung cancer screening may offer some additional benefit, but patients should be made aware that our study was a modeling study, not a gold-standard, randomized clinical trial.”
For this research, Dr. McMahon and her colleagues used data on 1,520 current and former smokers, screened as part of the Mayo Clinic helical CT trial, to populate the Lung Cancer Policy Model (LCPM), a comprehensive microsimulation model of lung cancer development, disease progression, lung cancer detection, treatment results, and survival.
Dr. McMahon is a senior scientist at the Institute for Technology Assessment at Massachusetts General Hospital. Her co-authors are from the departments of radiology and medicine at Harvard Medical School, Dana-Farber Cancer Institute, the University of Minnesota, and the Mayo Clinic, Rochester.
The initial data were compiled from the Mayo Clinic’s single-arm evaluation of helical CT in current and former smokers (61% of the total study population).
All patients underwent baseline screening and four annual CT exams. Nodules found on a screening exam were followed up with thin-section CT imaging.
In their study, Dr. McMahon and colleagues simulated the participants as if they had been randomized to both screened and nonscreened arms. At 6-year follow up, the screening arm had an estimated 37% relative increase in lung cancer detection, compared with those who had not been screened. Relative reduction in lung-cancer-specific mortality was 28%. The relative reduction in all-cause mortality was 4%. At 15 years, the relative reduction in lung-cancer-specific mortality was 15% and the reduction in all-cause mortality was only 2%.
Dr. McMahon’s group discussed why the study led to a middle-of-the-road conclusion on the benefits of screening.
I-ELCAP predicted 80% of people whose lung cancer was caught early with CT screening could expect to live at least another decade (Radiology 243:239-249, 2007; N Engl J Med 355:1763-1771, 2006).
A longitudinal modeling analysis of the Mayo data by Peter Bach, MD, and colleagues found that CT lung screening increased the rate of lung cancer diagnosis and treatment, but did not meaningfully reduce the risk of advanced lung cancer or death from lung cancer (JAMA 297:953-961, 2007).
“These three studies did use entirely different methods, but I would argue that the differences in results are due more to differences in reporting of endpoints, Dr. McMahon said.
The I-ELCAP trial reported survival—the time between diagnosis and death, often expressed as the proportion of patients alive at a particular time point after diagnosis—whereas the current study and the Bach analysis reported the change in mortality rates—deaths in a fixed length of time from the beginning of the study—between the screening and (simulated) control arms, she said.
Commentary
Results from modeling studies are exciting, but the differing outcomes in these studies need to be emphasized to the public, said Rob van Klaveren, MD, PhD, an associate professor at Erasmus Medical College, Rotterdam, The Netherlands.
Dr. van Klaveren is leading the Dutch-Belgian NELSON randomized CT lung screening trial that began in 2004.
At the 2008 ASCO annual meeting, David Christiani, MD, offered his thoughts on NELSON as well as the Mayo Clinic CT screening modeling study. Dr. Christiani is a professor of occupational medicine and epidemiology, Harvard School of Public Health, and professor of medicine, Harvard Medical School.
‘Prudent to wait’
“Lung cancer screening might well be shown to be beneficial, and we anticipate interesting results from NELSON and other [randomized, controlled trials],” Dr. Christiani said. “It would be prudent to await the final results before recommending widespread adoption.”
