Negative volume CT screening reduced need for further tests in high-risk patients

High-risk patients who were screened for lung cancer using CT and had negative first-round results were unlikely to be diagnosed with lung cancer one to two years after the screen.

A team of researchers in the Netherlands evaluated 7,557 participants who underwent CT screening in years one, two and four of the NELSON trial, a randomized study of lung cancer screening. Using a commercially-available 16-detector CT scanner, researchers evaluated a detected noncalcified nodule according to its volume or volume-doubling time. Growth was defined as an increase in volume of at least 25% between two scans.

In round one, the lung cancer detection rate was 0.9%. Sensitivity was 94.6% (95% CI; 86.5-98.0) and specificity was 98.3% (95% CI; 98.0-98.6). In a subject with a positive CT screening test, the researchers said the probability that the lesion would be malignant was 36%. With a negative screening test, the probability that a participant would not have lung cancer was 99.9%.

Of the 7,361 patients with a negative CT scan after round one, 20 lung cancers were detected during the two years of follow-up.

In round two, screening sensitivity was 96.4% (95% CI; 86.8-99.1) and specificity was 99% (95% CI, 98.7-99.2). The positive predictive value was 42.2% (95% CI; 33.9-50.9), and the negative predictive value was 99.9% (95% CI; 99.9-100.0). The results were indeterminate in 480 (6.6%) of patients and 276 had a follow-up scan roughly 90 days later. Another 231 were rescreened approximately 12 months later. Thirty-eight patients had positive results upon follow-up.

The recall rates for participants with indeterminate results was 19.0% after the first round of screening and 3.8% following round two.

In an accompanying editorial, James L. Mulshine, MD, Rush University Medical Center, Chicago, and David M. Jablons, MD, University of California, San Francisco, said these results are important because they show “the diagnostic efficiency of a defined clinical management approach within the context of a randomized trial.”

“The results suggest that the efficiency of the diagnostic workup for lung cancer can be improved by integrating the measurement of volume growth of lung nodules as an indicator of clinically significant lung cancer while limiting the need for additional costly or potentially harmful diagnostic procedures,” they wrote. “This quantitative imaging application may represent an important advantage over the usual qualitative application of imaging tools in a cancer-screening context.”

Van Klaveren RJ. New Engl J Med. 2009; 361:2221-9.