For Smokers, a Tough Call
Despite New Study, Experts Are Split on Whether the Benefits Of Early CT Screening for Lung Cancer Outweigh the Potential Risks

By Sandra G. Boodman
Washington Post Staff Writer
Tuesday, November 14, 2006

To screen or not to screen?

That is the question patients -- most of them current or former smokers -- are asking doctors following the publication of a large international study that found that spiral CT scans can detect lung cancer at its earliest and most curable stage.

The results, published last month in the New England Journal of Medicine, have fueled hope that the technology can lead to early treatment of the most common cause of cancer death, as mammography has done for breast cancer.

But the study involving more than 31,000 current and former smokers headed by researchers at Weill Medical College at Cornell University failed to answer a fundamental question: Does screening extend lives or merely find cancer earlier? Because the study did not use a control group, there is no way to tell whether the risks of screening, which include repeated radiation exposure and the possibility of aggressive follow-up treatment, outweigh the benefits of early detection, experts from the American Cancer Society and other groups say.

Therein lies the rub -- and the reason doctors are sharply divided about testing people who don't have symptoms, even for a cancer that kills 85 percent of the 174,000 Americans found to have the disease annually within five years of diagnosis.

Interest in lung cancer screening is growing, experts agree. Some hospitals and radiology clinics in the Washington area offer tests that range from $175 to $800 and are rarely covered by insurance. About 60 percent of hospitals and clinics own CT scanners, which were developed in the 1990s and can detect tumors as small as a grain of rice.

"I think the answer is pretty clear: no," said Ned Patz, a professor of radiology, cancer biology and pharmacology at Duke University Medical Center, when asked about screening. "I'm not saying it won't work, just that there's no solid proof yet that it does."

That's what Patz told his 74-year-old father, a former smoker who has no symptoms but worries about the consequences of his 20-year pack-a-day habit. A more definitive answer, Patz said, must await the outcome of a National Cancer Institute study involving 53,000 current and former smokers. That study, launched in the mid-1990s, is comparing the death rates of patients who received spiral CT scans with those screened with chest X-rays. Results are expected by about 2010.

Radiologist Alex Kladakis says he thinks there's no need to wait. One of 22 physicians at Washington Radiology Associates, among the area's largest radiology practices, Kladakis said that the recent study shows "the proof is really there" and that people over 40 who have smoked the equivalent of a half-pack a day for 15 years should consider being screened. Washington Radiology charges $686 for the non-invasive test, which takes about 30 seconds; a physician's order is required.

To Kladakis, the most persuasive finding in the study, dubbed I-ELCAP (International Early Lung Cancer Action Project), is the survival data. Of the 412 people in the study found to have Stage 1 cancer who underwent surgery, 85 percent were alive five years later; the eight patients who declined treatment were dead. The average five-year survival rate for Stage 1 lung cancer is about 70 percent, according to federal statistics.

"It seems to me those eight are kind of a mini-control group," Kladakis said, adding that requests for lung screenings have increased since the study was published. "Lung cancer is a very, very bad disease," he said. And by the time a patient has symptoms, such as a persistent cough, it usually means the disease is advanced, Kladakis said, citing the example of ABC anchorman Peter Jennings, who died of lung cancer last year.

Wide Interest

Claudia I. Henschke, the researcher who headed the I-ELCAP study, said the study shows that early diagnosis means that a cure is possible. She has said she does not think that a randomized controlled trial, the gold standard in medicine, is necessary to prove the effectiveness of lung CT scans, as do officials at the National Cancer Institute and the American Cancer Society.

Some doctors apparently agree with her. Last week, Nebraska officials announced the nation's first free statewide screening for patients at high risk: those who have smoked for 35 years or two packs a day for 15 years.

"If our effort succeeds, we could help shape lung cancer screening programs across America," said Kenneth H. Cowan, director of the Eppley Institute for Research in Cancer and Allied Diseases at the University of Nebraska Medical Center.

The potential cost of mass screening is enormous. A 2003 study by researchers at Johns Hopkins University estimated that 50 million Americans ages 45 to 75 are current or former smokers. Screening half of them with periodic lung CT scans would cost about $115 billion, the researchers said.

Consumers currently don't have enough information to decide whether screening is a wise choice, said Arthur Levin, director of the New York-based Center for Medical Consumers. "One study is not sufficient. The problem is that in this country we let the marketplace decide this. And because lung cancer is such a major killer, there's going to be a lot of interest in moving this down the road" and many people who say they want the test but may not understand the pitfalls.

CT scans, which are routinely used to stage lung cancer once it has been diagnosed, often pick up nodules, clumps of potentially cancerous tissue, during screening. Their size matters: Smaller nodules are less likely to be malignant than larger ones, experts say. Thoracic experts estimate that as many as 60 percent of smokers and former smokers have lung nodules, Patz said, but people who never smoked also can have them.

Finding a nodule usually triggers a cascade of interventions, starting with a biopsy, which can cause infection, scarring and a collapsed lung. A cancer diagnosis typically results in surgery, often followed by chemotherapy or radiation or both. Doctors have no accurate way of determining which tumors are fast-growing and aggressive and which grow so slowly that patients are likely to die with and not of them, as with some types of prostate cancer.

One Man's Story

Dennis Barry, 51, said he was stunned to learn last year that the scan he underwent as part of the I-ELCAP study at Holy Cross Hospital in Silver Spring turned up a suspicious nodule. Barry, an executive vice president of Donohoe Construction Co., said he enrolled in the study at the suggestion of a relative who works at the hospital. He said he smoked a pack a day from age 15 to age 40, when he quit.

Barry, whose father died of lung cancer at 65 after years of smoking, is one of three people out of approximately 500 screened during the Holy Cross study who were found to have Stage 1 lung cancer, according to oncologist Kashif Firozvi.

Last summer, weeks after the scan, Barry had part of his lung removed, followed by chemotherapy.

"I'm a lucky guy," Barry said. "I think this test saved my life." Two weeks ago, he said, tests showed he was cancer-free.

Laurie Fenton, president of the Lung Cancer Alliance, a Washington-based advocacy group, said that good news like Barry's has been all too rare when it comes to lung cancer.

She said that the findings of the study, which she called "a watershed moment," are so persuasive that her group is advocating screening scans for patients at high risk similar to those in the I-ELCAP study: people older than 50 who are current smokers or who smoked the equivalent of a pack a day for 10 years, and those with significant exposure to secondhand smoke because they grew up in a home with smokers or who were exposed to high levels of radon gas or other known lung carcinogens.

"We have to educate doctors" about the benefits of screening, Fenton said.

Lung cancer screening, she said, is being held to a higher standard than mammography or colonoscopy, which were widely used and covered by insurance while studies were underway to determine their effectiveness. "For us, it's so frustrating," Fenton said.

Firozvi agreed. "There has never been a randomized controlled trial of Pap smears," he said. "I think this test is comparable to mammography."

Fenton said she thinks stigma has colored attitudes toward lung cancer, which in about 85 percent of cases results from smoking. "The attitude is it's your problem. You deal with it."

But to Duke's Patz, the overarching issue is the strength of scientific evidence. Earlier studies of technology initially regarded as promising, such as chest X-rays, found the technology was ineffective in reducing the lung cancer death rate.

"Just because we haven't done the right thing in the past doesn't mean we shouldn't do it now," Patz said. "If we'd done the right studies to begin with, we'd have the answer now."

View all comments that have been posted about this article.

© 2006 The Washington Post Company