- Home>
- Specialties>
- General Medicine>
- Feature
Screening for Lung Cancer by Spiral CT: The Controversy Continues
The jury's still out on whether we should screen for lung cancer with spiral CT.
Use of low-dose spiral computed tomography to screen for lung cancer has sparked considerable debate. In 2 new reports, investigators examine the yield of this intervention.
In a Japanese study, 1611 asymptomatic smokers underwent spiral CT scanning, chest x-ray, and sputum cytology. CT scans were abnormal in 186 patients (12%). Many patients also underwent additional screenings at 6-month intervals, and abnormalities were detected on 721 (9%) of 7891 repeat CTs. Overall, 36 patients were diagnosed with lung cancer (32 by CT, 4 by cytology only). Mean tumor diameter was about 17 mm, and most cancers were stage IA. Estimated 5-year survival for the 36 patients with lung cancer was 71%.
In another study, Mayo Clinic researchers enrolled 1520 smokers who underwent spiral CT scanning and sputum cytology at baseline and again 1 year later. Initial CT scans showed uncalcified lung nodules in 51% of enrollees, and 1-year scans revealed new nodules in an additional 13% of patients. Twenty-five lung cancers were diagnosed (23 by CT), but 7 patients underwent lung resections for benign lesions. Twelve of 21 non-small-cell cancers were stage IA (mean size, 17 mm).
Comment: In both studies, screening led to lung cancer diagnoses in about 2% of patients. Differences in patient characteristics, CT scanning technique, and scan interpretation presumably account for the marked difference in nodule prevalence in the 2 studies. The 5-year survival rate of 71% in the Japanese study is deceptive, because lead-time and length biases inflate survival rates in nonrandomized, uncontrolled screening studies like these.
Given the lack of evidence that lung cancer screening reduces mortality, the growing use of CT for screening remains controversial. Editorialists Heffner and Silvestri are not convinced that the small size of CT-detected nodules translates into good prognoses, and they worry about the consequences of numerous false-positive findings. In another essay, Lee and Brennan express reservations about trends in which spiral CT and other expensive and unproven screening tests are marketed directly to consumers.
The National Cancer Institute currently is conducting the Lung Screening Study, in which patients are being randomized to chest x-ray or low-dose CT; this study is designed to look only at the yield and consequences of screening during 1 year. Planning is underway for a larger and longer-term study to examine the effect of screening on mortality.
— Allan S. Brett, MD
Published in Journal Watch General Medicine March 12, 2002
Citation(s):
Sobue T et al. Screening for lung cancer with low-dose helical computed tomography: Anti-Lung Cancer Association project. J Clin Oncol 2002 Feb 15; 20:911-20.
- Original article (Subscription may be required)
- Medline abstract (Free)
Swensen SJ et al. Screening for lung cancer with low-dose spiral computed tomography. Am J Respir Crit Care Med 2002 Feb 15; 165:508-13.
- Original article (Subscription may be required)
- Medline abstract (Free)
Heffner JE and Silvestri G. CT screening for lung cancer: Is smaller better? Am J Respir Crit Care Med 2002 Feb 15; 165:433-4.
- Original article (Subscription may be required)
- Medline abstract (Free)
Lee TH and Brennan TA. Direct-to-consumer marketing of high-technology screening tests. N Engl J Med 2002 Feb 14; 346:529-31.
- Original article (Subscription may be required)
- Medline abstract (Free)