NCI Begins Trial of Spiral CT vs X-rays as a Screening Test

November 1, 2002

BETHESDA, Maryland-Researchers have begun accruing 50,000 patients for the National Lung Screening Trial (NLST), the much-awaited comparison of the efficacy of spiral CT scans and chest x-rays in reducing lung cancer mortality. Investigators in the 8-year, National Cancer Institute-supported study expect to complete enrollment within 2 years. Enrollment will be aided by a $5 million educational campaign funded and organized by the American Cancer Society (ACS), and aimed at encouraging current and former cigarette smokers to participate in the trial.

BETHESDA, Maryland—Researchers have begun accruing 50,000 patients for the National Lung Screening Trial (NLST), the much-awaited comparison of the efficacy of spiral CT scans and chest x-rays in reducing lung cancer mortality. Investigators in the 8-year, National Cancer Institute-supported study expect to complete enrollment within 2 years. Enrollment will be aided by a $5 million educational campaign funded and organized by the American Cancer Society (ACS), and aimed at encouraging current and former cigarette smokers to participate in the trial.

"We estimate that about 169,000 men and women will be diagnosed with lung cancer in 2002, and about 155,000 deaths will occur," said co-principal investigator John K. Gohagan, PhD, chief of the NCI Early Detection Research Group. He emphasized that the tests being studied are the only accepted and proven screening tests for lung cancer that could reduce mortality.

Thirty medical centers throughout the United States will take part in the $200 million randomized, controlled study. Results from earlier spiral CT studies, conducted by a collaboration of New York City researchers and a group in Japan, have indicated that spiral CT, also known as helical CT, can detect tumors much smaller than 1 cm. Chest x-rays can reveal tumors as small as 1 to 2 cm in size. However, there is no scientific evidence showing that either chest x-rays or spiral CT screening has a positive effect on lung cancer mortality.

Despite this lack of mortality evidence, the potential for finding lung cancer at a size that theoretically should increase a patient’s chances of surviving the disease has led many present and past smokers to have spiral CT scans. A number of specialized centers have opened to meet the demand, which has disturbed oncologists and public health officials.

"In many instances, those individuals, in good faith, are having screening tests done on the assumption that they are helping themselves from a medical perspective," said Denise Aberle, MD, professor and chief of thoracic imaging, UCLA, and the NLST’s other co-principal investigator.

"The reality is we don’t know that the screening test is actually of benefit. We don’t know whether that test may actually expose them to downstream complications of additional diagnostics, which have their own morbidity and potential mortality that may do them more harm than the screening test itself," she said.

NCI has sponsored three previous studies, which began in the 1970s, examining the use of chest x-rays and sputum cytology as potential screening tests for lung cancer.

"Those trials were too small to determine the kinds of effect we think we might see with chest x-rays," Dr. Gohagan said. "All were designed to look for a 50% effect, and we think the effect is much smaller than that, based on postanalyses of those data." The NLST is powered to detect a 20% effect by chest x-rays.

Health Policy Implications

Whatever the NLST results, they will have significant health policy implications, "particularly if it turns out that spiral CT, as many suspect, is more effective at reducing deaths from lung cancer," Dr. Gohagan said. John R. Seffrin, PhD, ACS chief executive officer, added that if spiral CT does prove a life-saving screening test, "we are talking about something that could virtually change the cancer curves all over the world."

Two research groups will conduct the screening trial. One of them is currently carrying out the Prostate, Lung, Co-lorectal and Ovarian (PLCO) Cancer Screening Trial, in which some patients are receiving chest x-rays as a screening test for lung cancer.

The second is the American College of Radiology Imaging Network (ACRIN), which conducts imaging trials.

"The single most important thing we are analyzing in the NLST is the difference between spiral CT and chest x-rays in the number of deaths from lung cancers," Dr. Aberle said. "But there are a number of other important endpoints that some of the sites will be studying."

Those endpoints include the impact of screening on the smoking behavior of current smokers who join the trial; the emotional effect of screening on participants and the emotional impact of a positive or negative result, particularly on individuals with a false-positive or false-negative test; and the kinds of medical tests that persons with a positive test undergo as a result of their screening.

Another important element of the trial is the collection of blood, sputum, and urine specimens from some participants, which will be placed in a specimen bank for later research use. "As we learn more about the molecular events and the genetics of lung cancer, these specimens will help us identify biomarkers of lung cancer that can help predict who is at highest risk of lung cancer or who may have a precancerous condition," Dr. Aberle said.

In the randomized NLST, one arm will receive three spiral CT scans, one in each of the first 3 years of their participation. Similarly, the other group will receive three annual chest x-ray examinations. Screening tests are free, but participants or their health plans must pay for any diagnostic follow-up and treatment. The number of lung cancers detected and lung cancer deaths will be monitored yearly until 2009.

Previous studies have shown that up to 60% of spiral CT scans of current and former smokers show abnormalities. In addition, these studies indicate that spiral CT scans may detect tumors so small that they would never become life threatening, leading to unnecessary diagnostic tests and treatments.

"The vast majority of things that the spiral CT scan detects will not be lung cancer," Dr. Aberle said. "So when you look at cost, you can’t simply look at the cost of the screening test but at the cost of the downstream implications of the screening test."

Trial investigators are seeking current and former smokers ages 55 to 74 who are in generally good health and at relatively high risk of lung cancer. In addition, participants may never have had lung cancer or had any cancer within the last 5 years other than some skin cancers or in situ cancers. Finally, participants cannot be enrolled in any other cancer screening or cancer prevention trial, and they cannot have had a CT scan of the lungs within the 18 months prior to enrolling in the NLST.

The ACS campaign will focus on the areas around the 30 NLST sites. Promotional and outreach efforts will be designed in cooperation with researchers at the centers to best tailor materials to the population living near each site.

Additional information about the trial is available by calling 1-800-422-6237 or online at www.cancer.gov/NLST