Background: Lung cancer is the leading cause of cancer-related death in the United States. However, persons with early lung cancer have lower lung cancer–related mortality than those with extensive disease, suggesting early detection and treatment of lung cancer might be beneficial. Low-dose computed tomography (LDCT) and chest x-ray (CXR) have been studied for early screening, with several new studies reporting results since the last review.
Purpose: To update the 2004 review of screening for lung cancer for the U.S. Preventive Services Task Force.
Data Sources: MEDLINE (2000 to 2012), Cochrane Central Register of Controlled Trials and Cochrane Database of Systematic Reviews (through fourth quarter 2012), Scopus, and reference lists.
Study Selection: English-language randomized, controlled trials or cohort studies that evaluated screening or treatment interventions for lung cancer and reported health outcomes.
Data Extraction: Details about participants, study design, analysis, followup, and results were abstracted; study quality was rated using established criteria, where applicable.
Data Synthesis (Results): Four trials reported the effectiveness of screening with LDCT for lung cancer in patients with personal smoking exposure: one large good-quality trial reported screening was associated with reduced lung cancer and all-cause mortality reductions of 20 percent (95% CI, 6.8 to 27.6) and 6.7 percent (95% CI, 1.2 to 13.6), respectively. Three small European trials (two fair- and one poor-quality) showed no benefit of screening. When the three good- or fair-quality trials were combined in random effects meta-analysis, the relative risk of lung cancer mortality was 0.81 (95% CI, 0.72 to 0.91). One trial evaluated CXR screening in over 150,000 participants from the general population and reported no benefit of screening in this group or in a subset with personal tobacco smoke exposure. The reported sensitivity of LDCT for detecting lung cancer ranged from 80 to 100 percent and specificity from 28 to 100 percent in six studies; each study varied in its reporting method. The harms associated with LDCT screening included radiation exposure ranging from 0.61 to 1.5 mSv per scan, some degree of overdiagnosis of lung cancer that varied by study, and a high rate of false-positive examinations, which were typically resolved with further imaging. Most patients with positive results who underwent an invasive procedure were diagnosed with lung cancer. Smoking cessation was not significantly impacted by screening, although individuals with positive or indeterminate screens showed a trend toward reduced smoking or sustained abstinence. Patients with positive or indeterminate scans had some evidence of short-term increases in anxiety and distress but not long-term in the five studies evaluating this; patients with negative scans had a reduction in distress. Finally, no trials comparing treatment of stage I non-small cell lung cancer (NSCLC) with no treatment have been conducted. However, survival associated with surgical resection was evaluated in 11 studies of mostly symptomatic and unselected patients that have shown 5-year survival rates in the 71 to 90 percent range for stage IA NSCLC and 42 to 75 percent for stage IB NSCLC and that surgical resection is the U.S. standard of care. Harms of treatment of stage I NSCLC were poorly reported and ranged among the studies that reported them.
Limitations: Three trials were underpowered and of too short of duration to reach conclusions on effectiveness of screening. Overdiagnosis is an important harm of screening but its magnitude is uncertain. No studies of LDCT have reported results in women or minority populations.
Conclusions: Good evidence shows LDCT can significantly reduce mortality from lung cancer. However, there are significant harms associated with screening that must be balanced with the benefits. More efforts to reduce false-positive examinations are of paramount importance and smoking cessation remains the most important approach to reducing lung cancer mortality.