Chest X-Ray Screening Does Not Reduce Lung Cancer Mortality

From Medscape Medical News Laird Harrison October 26, 2011 (Honolulu, Hawaii) — The largest study yet to examine the issue shows that screening with chest radiographs does not reduce mortality from lung cancer, researchers reported here at CHEST 2011: American College of Chest Physicians Annual Meeting. The results, which were also published online October 26 in JAMA, confirmed earlier research on the issue but still came as a disappointment. "We were hopeful the chest X-rays we did would make a difference," coauthor Paul A. Kvale, MD, told Medscape Medical News. Putting these findings together with a those of study published in August in the New England Journal of Medicine, health policy groups are likely to recommend screening with low-dose computed tomography (CT), but not chest X-ray, and only for patients at high risk for lung cancer, said Dr. Kvale, a pulmonologist at Henry Ford Hospital in Detroit, Michigan. For the current study, part of the Prostate, Lung, Colorectal and Ovarian Cancer Cancer Screening Trial, researchers enrolled 154,901 participants aged 55 through 74 years. "This was the biggest study of its kind ever done," said Dr. Kvale. The investigators randomly assigned 77,445 of the patients to receive annual screenings with chest X-rays, and 77,456 to receive usual care, at 1 of 10 centers across the United States between November 1993 and July 2001. They offered participants in the screening group annual posterio-anterior view chest radiographs for 4 years. The participants in the usual care group were not offered screenings as part of the study, although 11% undertook chest X-rays independent of the study. Between 79% and 87% of the participants in the screening group got the X-rays offered each year. Researchers followed-up the patients for a maximum of 13 years until December 31, 2009. They tallied a lung cancer incidence of 20.1 per 10,000 person-years in the screening group and 19.2 per 10,000 person-years in the usual care group, for a rate ratio (RR) of 1.05 (95% confidence interval, 0.98 - 1.12). They counted 1213 deaths from lung cancer in the screening group compared with 1230 in the usual care group, for an RR of 0.99 (95% confidence interval, 0.87 - 1.22). The study avoided problems of previous studies, such as a large number of screenings in the control group, but still came up with the same bottom line, said Dr. Kvale. Another important aspect of the study was its analysis of a subgroup of patients who were at high risk for lung cancer because of their age, smoking, and other factors. Just as in the study as a whole, researchers randomly assigned 15,183 members of this subgroup to screening, and the same number to usual care. After 6 years of follow-up, 518 members of the high-risk screening group got lung cancer, and 316 died of the disease. In the high-risk usual care group, 520 got lung cancer and 334 died of it, for an RR of 0.94 (95% confidence interval, 0.81 - 1.10). The high-risk group was intentionally selected to match a high-risk group in the National Lung Screening Trial, published in the New England Journal of Medicine. In that study, researchers compared high-risk participants screened with chest X-rays with high-risk patients screened with low-dose CT. They concluded that the low-dose CT screening reduced the mortality rate of these patients by 20%. If low-dose CT screening is 20% better than X-ray screening, and X-ray screening is the same as usual care, then it would be logical to assume that CT screening is 20% better than usual care. However, more statistical analysis should be done before reaching that conclusion, writes Harold C. Sox, MD, from Dartmouth Medical School in West Lebanon, New Hampshire, in an editorial published along with the study in JAMA. New studies should directly compare usual care to low-dose CT-screening, Dr. Sox writes. Still, the findings are already being taken into consideration by a coalition of groups working on new guidelines for lung cancer screening, said Frank C. Deterrbeck, MD, chief of thoracic surgery at Yale University, New Haven, Connecticut, and cochair of the coalition. The coalition, made up of the American College of Chest Physicians, the American Cancer Society, the American Society of Clinical Oncology, the National Comprehensive Cancer Network, and to a lesser extent, the American Thoracic Society, will publish new guidelines within a few months, Dr. Deterrbeck told Medscape Medical News. However, he would not confirm that the guidelines will call for low-dose CT screening for everyone who is at high risk for lung cancer. "I think there is a potential benefit, as well as a potential harm," he said. "Selection of the appropriate population is something we have to pay careful attention to." Although low-dose CT poses a low risk from radiation, it often leads to other diagnostic procedures, some of which may not be necessary. "Low-dose CT picks up a lot of stuff that's nothing," he said. On the basis of the low-dose scans, patients may be referred for regular CT, with its higher doses of radiation, and for biopsies, which can cause complications. The cost questions are complicated, too, Dr. Deterrbeck said, as the expense of the screening must be weighed against the costs that are saved in treatment costs if cancer is caught earlier. However, the short-term implications of the study are clear, he said. "We have not employed X-rays as a screening tool for lung cancer, and I guess we won't." Dr. Kvale and Dr. Detterrbeck have disclosed no relevant financial relationships. Dr. Sox has disclosed that he is an unpaid member of advisory boards for the Southwest Oncology Group and the Fred Hutchinson University of Washington Cancer Consortium. JAMA. Published online October 26, 2011. Full text, Editorial CHEST 2011: American College of Chest Physicians Annual Meeting: Session 7225. Presented October 26, 2011.