乌鸦传媒

Annual screening with low-dose computed tomography significantly reduced the risk of lung cancer mortality in current and former smokers as compared with chest x-ray, updated results from a large randomized trial confirmed.

Participants screened with low-dose helical CT had a 20% lower lung cancer mortality than the group screened by conventional chest radiographs. Screening with CT also was associated with a small but statistically significant reduction in overall mortality, as reported online in the New England Journal of Medicine.

Despite the positive outcome, the investigators offered a cautious assessment of the implications, noting unresolved issues related to patient selection and cost. They also cited a high rate of false-positive results and the potential for overdiagnosis and overtreatment.

"Before public policy recommendations are crafted, the cost-effectiveness of low-dose CT screening must be rigorously analyzed," Christine Berg, MD, of the National Cancer Institute, and co-authors wrote in their discussion of the results.

"The reduction in lung-cancer mortality must be weighed against the harms from positive screening results and overdiagnosis, as well as the costs."

The published results confirmed those reported in late 2010 when the National Lung Screening Trial (NLST) ended prematurely because of the significant benefit revealed by a planned interim analysis of the data.

The plans for and design of the NLST emerged from two distinct but related lines of evidence. Observational studies showed that low-dose helical CT detected more lung nodules and lung cancers, including early-stage cancers, as compared with chest x-ray, the authors noted.

Conventional radiography was chosen for the control arm of the trial because NCI-sponsored Prostate, Lung, Colon, and Ovary (PLCO) trial was evaluating screening by chest x-ray versus community care when NLST was designed. If PLCO showed a benefit for chest x-ray, a trial comparing CT and community care would have been less informative than a comparison of CT and chest x-ray, which would have become the screening standard, the authors noted.

The NLST involved 53,454 high-risk men and women ages 55 to 74 enrolled at 33 U.S. medical centers. All participants had at least a 30 pack-year history of smoking, regardless of whether they were current or former smokers.

Participants were randomized to low-dose CT or to single-view posteroanterior chest x-ray. They were invited to undergo three screening tests at one-year intervals.

Investigators enrolled participants from August 2002 through April 2004, and ascertainment of lung-cancer mortality (the primary endpoint) and death from any cause continued through December 2009.

The trial ended after a median follow-up of 6.5 years and a maximum of 7.4 years in each group. Adherence to all three rounds of screening was 95% in the CT arm and 93% in the x-ray arm.

Across the three rounds of screening, 39.1% of the CT arm and 16% of the x-ray arm had at least one positive result. The proportion of screening tests that identified a clinically significant abnormality other than lung cancer was 7.5% with CT and 2.1% with chest x-ray.

The authors reported, however, that 96.4% of positive results in the CT arm and 94.5% in the x-ray group proved to be false-positives. The false-positive rate did not differ significantly across the three rounds of screening.

Overall, 1,060 lung cancers were diagnosed in the CT group -- 649 cancers diagnosed after a positive screening test, 44 after a negative test, and 367 among participants who missed a screening test or who received the diagnosis after screening had ended.

In the radiography group, 941 lung cancers were diagnosed: 279 after a positive screen, 137 after a negative screen, and 525 among participants who missed a screen or learned of the cancer after screening had ended.

The CT arm had more cancers diagnosed at stage IA (40.0% versus 21.1%) and fewer stage IV tumors (21.7% versus 36.1%).

Analysis of outcome data showed that 356 lung cancer deaths occurred in the CT arm and 443 in the x-ray arm, translating into lung cancer mortality rates of 247 and 309 per 100,000 person-years, respectively. The difference represented a 20% reduction in mortality risk in the CT arm (P=0.004).

With respect to overall mortality, 1,877 participants in the CT arm died during the study period, as did 2,000 in the radiograph arm, representing a 6.7% reduction in the risk of death from any cause with CT (P=0.02).

Lung cancer accounted for 24.1% of all deaths in the trial but 60.3% of excess deaths in the radiography group. When lung cancer deaths were excluded, the reduction in mortality in the CT arm declined to 3.2%, which was no longer significant (P=0.28).

Comment