Lessons Learned From Large Lung Cancer Screening Program

NEW YORK—A program to screen for lung cancer that was remarkably successful in recruiting participants has produced some surprises and taught the investigators some lessons about what to expect from large-scale lung cancer screening.

James R. Jett, MD, consultant in pulmonary and critical care medicine and medical oncology, Mayo Clinic, reported the findings at the 3^rd International Conference on Screening for Lung Cancer.

The Mayo Clinic study, launched in January 1999, enrolled 1,520 individuals within 12 months, when recruitment stopped. "We have a waiting list of 3,000 individuals who did not get into the trial," he said.

Eligibility requirements included being age 50 or older, a history of at least 20 pack-years of smoking, and a 5-year life expectancy. The mean pack-years for the enrolled group was 45, with 61% of the participants being active smokers and 39% having stopped smoking. Men made up 52% of the group, women 48%.

Screening includes sputum cytology and spiral CT scanning with low-dose radiation. The GE scanner used, Dr. Jett said, has 5 mm collimation and 3.5 mm reconstruction intervals. Because scanning from the top of the lungs to the iliac crest can be done in 12 seconds, patients are in and out of the scanner in less than 5 minutes, he said.

Scans are repeated annually for 3 years, and most patients have now had their first repeat scan.

Initial Screening

The first surprise the researchers encountered was that the initial scan produced many more positive scans than expected. "We detected a nodule in 51% of our individuals with the thin-section CT scans, a total of 1,300 nodules," Dr. Jett said.

About 90% of the nodules were 7 mm or less in diameter. "So far, we’ve found no cancers in nodules 7 mm in size or less," he said. "It will happen, but we haven’t found them yet." Nodules were 8 mm or larger in 11% of participants.

To date, Dr. Jett said, with second-year scans completed in about 75% of participants, "we have detected 19 prevalence cancers [ie, initial scan positive] and 2 incidence cancers [ie, year-2 scan positive]."

Of the 18 non-small-cell lung cancers detected, 88% were stage I and II. Seventeen were detected on initial screening and one, a stage IIA squamous-cell carcinoma, was found on second-year screening. Of the three small-cell lung cancers detected, two were found in the prevalence scan and one in the incidence scan.

"With aggressive, fast-growing cancers, it’s questionable how good screening is going to be, Dr. Jett commented. "We are not going to cure small-cell lung cancers found with a screening scan," he said.

Repeat Scans

Analysis of the repeat scans of 1,026 persons has been completed, Dr. Jett said. With just a year between scans, he noted, "20% of the time, in retrospect, we saw a nodule that we had missed the year before." These nodules ranged from 1 mm to 20 mm in size. "Most of them were small," he said, "but some of them are larger. This is a bit scary."

New nodules were detected in 12% of the individuals who have had repeat scans. "You’d better have a database where you can analyze all of this," Dr. Jett cautioned, "and you’d better have good comparisons with the previous CT scans. Next year, it’s going to be even tougher, because we’re going to have to compare the CT scans with two prior CT scans."

In addition to finding lung cancers, the scans at the Mayo Clinic have also detected three kidney cancers, 50 abdominal aortic aneurysms, an atrial myxoma, a bronchial carcinoid tumor, and a breast cancer. "This is what you’re going to find," Dr. Jett said, "and you’d better be prepared to be sending letters out to these people as to what they need to do in their follow-up."