CT Screening Finds Lung Cancers Early

85% of Cancers Found in Study Were More Survivable Stage 1 Type

Medically Reviewed by Louise Chang, MD on October 25, 2006
From the WebMD Archives

Oct. 25, 2006 - More than 90% of lung cancer patients end up dying of their disease; now a potentially landmark study suggests early screening could save most of them.

In the international study, routine screening using spiral computed tomography (CT) imaging resulted in the detection of lungcancercancer at its earliest and most treatable stage in 85% of cancers found.

The estimated 10-year survival rate for patients with these stage I cancers was 88%.

That compares with a typical five-year survival rate of around 5% for patients with advanced, stage IV lung cancer.

"Lung cancer is highly curable if we find it early enough," lead researcher Claudia Henschke, MD, PhD, tells WebMD. "We believe this study provides compelling evidence that CT screening for lung cancer offers new hope for millions of people at risk for this disease and could dramatically reverse lung cancer death rates."

More than 31,500 people at high risk for lung cancer took part in the International Early Lung Cancer Action Project (I-ELCAP), carried out at 38 centers in seven countries.

A total of 484 lung cancers were found in this high-risk group, which included people with a history of cigarette smoking, exposure to secondhand smoke, or exposure to occupational hazards such as asbestos or radon.

The study is published in the Oct. 26 issue of The New England Journal of Medicine.

Cautious Optimism

Experts who spoke to WebMD agreed the new findings are encouraging. But they also said important questions remain about the value of CT screening as a routine test for lung cancer.

In a 2004 report, a government task force concluded there is not enough evidence to either recommend or discourage routine screening of asymptomatic people for lung cancercancer.

Henschke and colleagues from New York Presbyterian Hospital-Weill Cornell Medical Center developed the screening model used in the international study. She acknowledges that adapting the model for widespread use will not be easy.

"A multidisciplinary approach and very careful follow-up is needed to make sure you aren't missing something," Henschke says.

One of the biggest concerns about the screening procedure is its high percentage of false-positive results.

According to the National Cancer Institute (NCI), studies indicate that between a quarter to more than half of CT scans among smokers and former smokers show abnormalities, and in most cases these do not turn out to be lung cancer.

The problem is that malignant tumors and benign lesions often look the same on a CT scan, and decisions about how to proceed once a lesion is found are not clear-cut.

Potentially risky invasive procedures such as needle biopsy or even surgery are usually needed to confirm a diagnosis of lung cancer.

In the I-ELCAP study, patients with small but suspicious lesions were either tested again using a different screening method or had a repeat CT scan three months later.

When a lung infection was suspected of causing a lesion, a two-week course of antibiotics was prescribed and a repeat CT was performed a month later.

If the second image showed that the lesion or lesions had grown, a needle biopsy was usually performed.


Another concern about CT scanning is that it could result in the diagnosis and treatment of tumors that would never become life-threatening.

Overdiagnosis is impossible to document in living patients, but autopsy studies have shown undetected lung cancers in many people who died of other causes.

The I-ELCAP results appear to address this concern. Eight of the study participants whose stage I cancers were first identified through CT scanning were not treated, and all eight died within five years of diagnosis.

The NCI is conducting its own study to compare CT scanning to traditional chest X-ray for the early detection of lung cancerlung cancer.

The aim of the study, which includes 52,000 current smokers or former smokers, is to determine if regular screening with either test reduces deaths from the disease.

Results from the trial are expected by 2009, but Gary Kelloff, MD, of the NCI, tells WebMD they could come sooner if clear trends are seen before then.

"We still don't know if we are reducing lung cancercancer mortality in a screened population," he says. "The I-ELCAP trial was well done and it certainly adds to what we know. But we still have a lot to learn."

Kelloff is special advisor to the NCI's Cancer Imaging Program, in the Division of Cancer Treatment and Diagnosis.

'Exciting Findings'

American Cancer Society Director of Screening Robert Smith, PhD, says the I-ELCAP study proves that a specific CT scanning model can be duplicated in different settings and that false-positive findings can be minimized with strict adherence to a rigorous screening protocol.

"These are very exciting findings that show real promise for reducing this country's top cause of cancer death," he says. "But health policy isn't made on the basis of one study, or by one organization."

Smith adds that the ongoing NCI lung cancer trial should answer many questions about the benefits vs. risks of lung cancer screening.

In the meantime, he says, people at risk for lung cancer who are considering CT screening should discuss the matter with their doctor.

Smith also recommends choosing a testing site with experience in lung scanning.

WebMD Health News


SOURCES: Henschke, C. The New England Journal of Medicine, Oct. 26, 2006; vol 355: pp 1763-1771. Claudia I. Henschke, MD, PhD, chief of chest imaging, New York-Presbyterian/Weill Cornell Medical College; professor of radiology and cardiothoracic surgery, Weill Cornell Medical College, New York, N.Y. Robert Smith, PhD, director of screening, American Cancer Society. Gary Kelloff, MD, special advisor, National Cancer Institute's Cancer Imaging Program, Division of Cancer Treatment and Diagnosis. NCI web site: "National Lung Screening Trial" report. NCI: "NCI Lung Cancer Screening." Agency for Healthcare Research and Quality web site. Unger, M. The New England Journal of Medicine, Oct. 26, 2006; vol 355: pp 1822-1824.
© 2006 WebMD, Inc. All rights reserved.