Limitations of screening for lung cancer with low-dose spiral computed tomography

Abstract

Past lung cancer screening trials in the United States with chest X-ray and sputum cytology were not able to show any decrease in lung cancer mortality; however, these trials are over 20 years old. Recent follow-up of the Mayo Lung Project showed a better survival from lung cancer in the screened arm, but no difference in overall mortality, suggesting an overdiagnosis of nonfatal cancers. Recent reports of low radiation dose spiral computed tomography (CT) chest screening for lung cancer have shown that CT screening detects cancers at a smaller size than chest X-rays. To date, there have been no randomized trials of CT versus observation or chest radiographs for screening purposes. All data available thus far on CT screening are from phase II proof-of-principle trials. The major limitations of CT screening, discussed here, include (a) a high rate of nodule detection: over 50% of participants will have at least one noncalcified nodule; (b) resulting follow-up CT scans, associated with increased costs; (c) cost and morbidity of biopsy or resection of benign noncalcified nodule (20-25% of such procedures in several trials); and (d) a small, but difficult to quantify, risk of cancer associated with multiple follow-up CT scans.