NLST update: CT detects 2X as many early-stage lung cancers as x-ray

Low-dose screening CT detected 132 cases of stage IA lung cancer, while chest x-ray detected 46 stage IA cancers in the first-screen results of the National Lung Screening Trial (NLST), according to research published May 23 in the New England Journal of Medicine.

The first-screen results reaffirm results reported in other large screening trials, including Early Lung Cancer Action Project (ELCAP), International Early Lung Cancer Action Project (I-ELCAP) and NELSON, and could help to build the case for lung cancer screening.

The NLST enrolled 53,439 current and former smokers with a pack history of at least 30 years, ages 55 to 74 years, and randomized them to annual low-dose CT or chest x-ray from August 2002 to April 2004 at 33 sites in the U.S. The trial was called to an early halt in November 2010 as researchers observed a 20 percent mortality reduction in the low-dose CT arm.

The current article reviews the screening, diagnosis and limited treatment results in the first screening round.

In the CT arm, all noncalcified nodules with long-axis diameters of 4 mm or greater in the axial plane were considered positive for potential lung cancer, according to Timothy R. Church, PhD, from the University of Minnesota School of Public Health in Minneapolis, and colleagues. All noncalcified nodules and masses detected on chest x-ray screens were considered potentially positive for lung cancer.

“As expected, more positive screening results, more diagnostic procedures, more biopsies and other invasive procedures, and more lung cancers were seen in the low-dose CT group than in the radiography group during the first screening round.” More early-stage cancers were diagnosed in the CT arm, but both groups had a comparable number of late-stage cancers.

A total of 7,191 of the 26,309 participants in the CT arm had a positive screening result compared to 2,387 of 26,035 in the x-ray arm.

In the low-dose CT group, 292 patients were diagnosed with lung cancer. In the chest-x-ray group, 190 were diagnosed. Stage IA cancers accounted for nearly all of the difference.

Sensitivity and specificity of the screening results were 93.8 percent and 73.4 percent, respectively, for low-dose CT and 73.5 percent and 91.3 percent, respectively, for chest x-ray.

The positive predictive value for any positive finding that led to biopsy was 52.9 percent in the low-dose CT group. However, it dropped to 3.8 percent for positive screening results overall, according to Church et al. The corresponding values for chest x-ray were 70.2 percent and 5.7 percent, respectively.

With respect to diagnostic procedures, 90.4 percent of participants in the low-dose CT group and 92.7 percent in the radiography group underwent at least one diagnostic follow-up procedure.

The most common imaging follow-up exams in the CT group and x-ray group were chest CT at 73.1 percent and 65.8 percent, respectively, and FDG-PET in 10.3 percent and 7.6 percent, respectively. A total of 2.2 percent of participants in the low-dose CT group and 3.5 percent in the chest x-ray group underwent at least one biopsy.

“In the low-dose CT group, a total of 10,313 imaging procedures were performed, including 7,288 chest CT examinations, as compared with 3,657 imaging procedures in the radiography group, including 2,158 chest CT examinations.”

The high compliance rate of 98.5 percent in the NLST and low number of nondiagnostic CT studies (four) suggest CT screening was well-implemented, according to Church and colleagues.

Although CT screening led to many diagnostic exams, the number of follow-up chest CT exams per positive screening result was modest at 7,288 CT scans per 7,049 participants with a positive result, according to the researchers. In addition, updated protocols that recommend less frequent follow-up CT exams could reduce the number of CT studies and bolster cost-effectiveness.

Denise R. Aberle, MD, the national principal investigator for NLST American College of Radiology Imaging Network (ACRIN), emphasized in a release that the first-screen result strongly suggests that CT lung cancer screening programs with radiologists who possess similar expertise and interpret similar numbers of CT cases that are obtained on scanners of the same caliber or better as those required for the NLST are likely to have results similar to those reported in the paper.

Church et al urged caution in generalizing these initial screening round results to the general population of smokers in the U.S., as NLST participants included a higher proportion of former smokers and a higher education level.

“What we’ve learned from the analysis of the first-screen results provides clinicians additional facts to discuss with patients who share similar characteristics as the NLST participants (current or former heavy smokers over the age of 55),” said Church in a release. “The results also caution against making blanket lung cancer screening recommendations, because each person’s trade-off between the risk of having an unnecessary procedure and the fear of dying of lung cancer is uniquely individual.”