SCREENING FOR LUNG CANCER

People at the highest risk of developing lung cancer appear to derive the most benefit from screening.

According to new data, screening with low-dose CT prevents the greatest number of lung-cancer-related in those at highest risk. In contrast, screening prevented very few deaths in those at lowest risk.

Overall, 88% of the screening-prevented lung-cancer deaths and 64% of the false-positive results occurred in people at the highest risk for lung cancer mortality.

The study appears in the July 18 issue of the New England Journal of Medicine.

The number of CT-prevented lung-cancer deaths was strongly correlated with the prescreening risk for death from lung cancer. Thus, the number of people requiring screening to prevent 1 death from lung cancer was 5276 in the 20% of study participants at the lowest risk and 161 in the 20% at the highest risk.

In addition, the ratio of the number of people with a false-positive result on screening to CT-prevented lung cancer deaths was also correlated with risk; the ratio was 1648 in the 20% of study participants at the lowest risk and 65 in the 20% at the highest risk.

"Among the heavy smokers who were eligible for the NLST [National Lung Screening Trial], those who had the highest lung cancer death risk benefited more from lung screening than those with the lowest lung cancer death risk," said study author Hormuzd A. Katki, PhD, from the divisions of cancer epidemiology and genetics at the National Cancer Institute in Bethesda, Maryland.

"Our findings are a proof of principle that calculating a smoker's risk of lung cancer death can better predict each smoker's benefit from undergoing low-dose CT lung screening," he told Medscape Medical News. "We are validating our risk calculations in additional study populations before making this tool available to physicians and the public."

Next Steps

The pivotal NLST showed that screening with low-dose CT could reduce lung cancer mortality in people at a high risk of developing the disease, compared with radiography. In NLST, participants were 55 to 74 years of age and had a history of heavy smoking. Participants underwent screening once a year for 3 years, and were then followed for another 3.5 years with no screening.

There were 247 deaths from lung cancer per 100,000 person-years in the low-dose CT group and 309 deaths per 100,000 person-years in the radiography group, which is a relative reduction in mortality from lung cancer with low-dose CT screening of 20% (95% confidence interval [CI], 6.8 to 26.7).

Several organizations, including the National Comprehensive Cancer Network, issued guidelines for lung cancer screening on the basis of the NLST results. A recent analysis showed that if all screening-eligible current and former smokers underwent low-dose CT screening, 12,000 deaths from lung cancer could be prevented each year in the United States.

It is clear that CT screening detects lesions that are very small, and that they will be missed with chest x-ray," said Zab Mosenifar, MD, codirector of the Women's Guild Lung Institute at Cedars-Sinai in Los Angeles. "But the challenge is how to reconcile false positives."

Although chest radiography can miss small malignant lesions, CT is very sensitive and can lead to unnecessary tests and invasive procedures, Dr. Mosenifar told Medscape Medical News. He was not involved in the NLST.

"This study is a step in the right direction, trying to define who is at the highest risk and who would benefit the most from screening," said Dr. Mosenifar. A critical next step will be to determine cost-effectiveness. "We need to narrow screening to a very special population — one at very high risk — but even then it's a huge number."

Defining the Screening Population

Screening limited to high-risk people for whom the potential benefits of low-dose CT screening outweigh the potential harms has been endorsed by professional societies. However, "there is uncertainty as to how a high-risk target population should be defined," Dr. Katki and colleagues note.

In their current study, the researchers assessed participants from the NLST intention-to-screen sample — 26,604 in the CT group and 26,554 in the radiography group. The primary end point was the rate of death from lung cancer from August 2002 to January 2009.

Participants were stratified into quintiles according to their predicted 5-year risk for death from lung cancer. The researchers compared the efficacy, the number of false-positive results, and the number of lung-cancer deaths prevented in those who underwent low-dose CT screening and those who underwent chest radiography.

The 5-year risk for lung cancer death ranged from 0.15% to 0.55% in the lowest-risk quintile to more than 2.00% in the highest-risk quintile.

The selected risk factors for the hazard-ratio model for lung cancer death included age, body mass index, family history of lung cancer, pack-years of smoking, years since smoking cessation, and a diagnosis of emphysema. For their hazard-ratio model for competing causes of death, the researchers included sex and race and excluded a family history of lung cancer.

During a median follow-up of 5.5 years, there were 354 lung cancer deaths in the CT group and 442 in the radiography group. The rate of lung-cancer-related mortality was lower in the CT group than in the radiography group (24.6 vs 30.9 per 10,000 person-years). This can be extrapolated to a relative reduction of 20.4% in the CT group (rate ratio, 0.80; P = .001) and to 6.3 fewer lung cancer deaths per 10,000 person-years (P = .001).

Lung-cancer mortality ratios with low-dose CT, compared with radiography, did not differ significantly from the lowest-risk to the highest-risk quintiles (0.97 in quintile 1, 0.78 in quintile 2, 0.75 in quintile 3, 0.70 in quintile 4, and 0.84 in quintile 5; P =.80 for trend).

However, because there was an increase in the risk for lung cancer death across quintiles, the "overall 20% reduction in the rate of lung cancer death in the CT group meant that the number of lung cancer deaths per 10,000 person-years that were prevented by low-dose CT screening increased significantly across risk quintiles (0.2 in quintile 1, 3.5 in quintile 2, 5.1 in quintile 3, 11.0 in quintile 4, and 12.0 in quintile 5; P = .01 for trend)," the researchers report.

The trends in lung cancer mortality ratios were similar, and differences in mortality were seen whether risk quintiles were based on the predicted risk for lung cancer or on the risk for lung cancer death (P = .90 for heterogeneity for both efficacy comparisons).

In addition, there was a significant decreasing trend in the number of participants with false-positive results per screening-prevented lung-cancer death across the quintiles (1648 in quintile 1, 181 in quintile 2, 147 in quintile 3, 64 in quintile 4, and 65 in quintile 5).

"Our study shows the value of risk calculations for smokers meeting NLST entry criteria, providing evidence for external organizations who are interested in developing risk-based lung screening guidelines in the future," said Dr. Katki. "However, more research needs to be conducted to understand whether risk calculations could also be used for other smokers."

The study was supported by the Intramural Research Program of the Division of Cancer Epidemiology and Genetics, National Cancer Institute.

N Engl J Med. 2013;369:245-254. Abstract