WEDNESDAY, 12/1/99 ~ MORNING EDITION

CT screening prevents lung cancer death

By Charles Bankhead

Lung cancer screening with CT saves lives, investigators in an ongoing clinical trial have concluded from a comparison of baseline and repeat screening studies.

Repeat screening revealed new, growing pulmonary nodules in 3% of patients, and 23% of the nodules proved to be malignant. The vast majority of the malignancies were stage I and highly resectable, however.

"I think we can change the cure rate [for lung cancer]," said Dr. Claudia Henschke, a radiologist at Columbia University in New York. "These results show that CT screening saves lives. We have to follow the patients to cure to see how many lives we can save. We should concentrate on early detection of lymph node metastases and the development of treatment for stage I disease."

The findings and conclusions come from an analysis of the first annual repeat screening exams conducted in high-risk patients recruited into the Early Lung Cancer Action Project (ELCAP). The project is a multidisciplinary investigation involving radiologists, pulmonologists, oncologists, surgeons, and epidemiologists. The aim of the project is to determine whether annual CT screening exams lead to earlier detection of lung cancer and a reduction in lung cancer morbidity and mortality.

Investigators in the multicenter effort enrolled 1000 cancer-free patients age 60 and older who had at least a 10-pack-a-year history of cigarette smoking. Each of the patients had a baseline screening CT exam, which was repeated a year later unless the patient had a lung cancer diagnosis before the annual repeat exam.

Henschke reported that 814 patients had a first annual repeat, and 597 have had subsequent annual repeats. The first repeat exams identified a total of 74 noncalcified nodules, 23 of which were identified in retrospect on prior exams. Nine additional nodules were excluded for other reasons, leaving 42 patients with newly detected and growing nodules.

Of the 42 nodules, 16 resolved with antibiotics, indicating the presence of unresolved pneumonia, Henschke said. Nine of the remaining 26 unresolved nodules proved to be malignant. One of the cancers was 25 mm, a size associated with increased risk. Seven of the remaining eight malignancies were stage I, and one small lesion (2 to 5 mm) was stage III.

"We know from major screening studies that if stage I lung cancer is not resected, those individuals die, but if the cancer is resected, the individuals live for a long time," Henschke said.

The patients with nonmalignant nodules have been followed for up to 1.5 years without evidence of malignant transformation.

In a related report at the RSNA Scientific Session, Henschke and colleagues reported that only female gender, lesion size, and number of smoking cessation years predicted whether a nodule identified at baseline was malignant. No recognized lung cancer risk factors predicted the presence of a nodule on the initial CT screening exam.

A third ELCAP report at RSNA showed that noncalcified lesions missed on low-dose helical CT exams are generally small (5 mm or less) and in a peripheral location. None of the noncalcified lesions missed on baseline CT scans (36 patients) turned out to be malignant. All the lesions were identified by follow-up CT evaluation in patients who had positive CT scans at baseline.