Scanning for Lung Cancer
Promising early-detection test may not be as effective as first thought
Carolyn Jones / SF Chronicle 6nov00

A year ago, millions of smokers and ex-smokers were excited over what looked like the biggest breakthrough in lung cancer survival rates since the invention of chemotherapy.

But now, scientists are cautioning smokers to think twice before signing up for a simple, 25-second CT scan that can detect cancerous nodules years before they show up on a standard X-ray.

``I don't want to rain on this parade -- this could turn out to be the greatest thing since sliced bread -- but we just don't know yet if this test actually works,'' said Dr. Virginia Ernster, vice chair of epidemiology and biostatistics at the University of California at San Francisco and associate director of the UCSF Comprehensive Cancer Center.

Ordinarily, patients wouldn't know they had lung cancer until they went to the hospital for another ailment, such as pneumonia, and a routine lung X-ray revealed lesions an inch in diameter or bigger. By that time, the cancer would likely have metastasized and doctors could only hope to extend a patient's life, not save it. Life expectancy for most patients is less than a year.

But by catching the lesions early, when they're only 6 to 8 millimeters in diameter, doctors can remove them through surgery. According to one study, the test could improve the five-year survival rate from 14 to 80 percent.

``Among patients, there's a lot of excitement about it,'' said Dr. Paul Brunetta, assistant clinical professor of pulmonary medicine at UCSF and part of the thoracic oncology unit at the school's cancer center.

Feared Disease

``Lung cancer is one of the diseases people fear the most, and patients are tired of doing nothing about it. Researchers are taking a wait-and-see attitude, but if I was a patient I wouldn't want to wait and see. I'd want the test,'' he said.

There's good reason for excitement over the test's possible benefits. Lung cancer kills more people than any other cancer, and the survival rates haven't changed in decades. More than 156,000 people in the United States died of lung cancer last year, and 164,000 new cases were diagnosed. About 90 percent of lung cancer patients are smokers or ex-smokers.

The new CT technology, which allows the computerized spiral imaging machines to run faster and more accurately, was introduced about a decade ago. But no one thought to use it for lung cancer detection until a New York doctor, Claudia Henschke at Weill Cornell Medical Center, released a study in the British medical journal The Lancet in July 1999 that said the machines could pick up tiny cancerous nodules in the lungs and potentially save thousands of lives.

The reaction was immediate. Smokers and former smokers across the country flocked to their doctors' offices clamoring for the test. Some clinics, including ones in Oregon, Arizona and New York, received hundreds of phone calls a day. Others started advertising the tests and even offering special discounts.

The response was tamer in the Bay Area, possibly because there are fewer smokers here. But six months ago, a clinic called HeartScan in South San Francisco started advertising in a local newspaper that it offers lung cancer scans. The clinic now averages three or four lung scans a day, at a cost of $495 each (or $350 if the patient gets a heart scan as well).

Numerous Problems With Test

But now, doctors are seeing numerous problems with the test. The main catch is that 85 percent of the nodules picked up on the CT scan are benign, and the only way to find out is by doing repeated biopsies or even surgery.

``Say you screen 1,000 patients and find 200 nodules, and 10 percent of those are cancer. That's a lot of unnecessary tests and surgeries. I'd feel a lot more comfortable with this if we had better numbers,'' said Dr. Richard Whyte, head of thoracic surgery at Stanford Medical Center. ``Otherwise I'd be a very busy man.''

In some cases, the repeated CT scans, biopsies and surgeries could be more harmful than anything found in the lungs. The CT scan itself exposes patients to radiation, patients occasionally die on operating tables, and there's a fairly high likelihood that a biopsy needle could miss a 6-millimeter lesion and puncture the lung instead.

Another problem is that so far, a CT scan for lung cancer is not covered by insurance because it's still considered experimental. Follow-up treatments would be covered, but patients would have to pay between $250 to $500 themselves for the initial test.

Another issue is the limited number of CT scan machines. Most major medical centers have one or two machines that are already in continuous use for other tests, such as brain or abdominal scans.

Dr. John Rego, chief radiologist at Kaiser Permanente in San Francisco, figures that his hospital would have to spend $1.3 million to buy a new scanner just to screen all the local Kaiser members who are smokers.

Soaring Costs

The costs could skyrocket with thousands of follow-up scans, biopsies and surgeries -- all for lesions that are most likely benign.

``It's clear that there will always be smokers in this country, and they'll always be consuming a big piece of the overall health dollar,'' Rego said. ``But maybe the money for CT scans would be better spent on smoking cessation instead.''

Some doctors also expressed concern that the test could represent an incentive for smokers to keep smoking. If a smoker pays $250 for a test and gets a clean result, he might feel free to keep smoking. In reality, smoking could cause a variety of other cancers and health problems, such as emphysema, heart disease and stroke.

But perhaps the most significant problem is that there have been no studies showing that patients who've had a CT scan live longer than patients who haven't had one. The test is not likely to become standard procedure until wide-ranging studies can show definitively that it reduces the mortality rate, like mammography has cut down on breast cancer deaths, researchers said.

Professor Brunetta is organizing a study at UCSF of 150 people over age 60 who smoked an average of one pack a day for 20 years. He plans to give them free CT scans, evaluations and phlegm tests over a two-year period, in hopes of showing that the patients fare better than lung cancer patients at large.

``This test isn't for everyone,'' Brunetta said. ``It's extremely controversial, and patients should know all the viewpoints before they pursue it. But I also think the test is likely to be very beneficial, if we can prove it works.''

Dr. Michael Rizzo, radiologist at Alta Bates Summit Medical Center in Berkeley and Oakland, said he hopes that all the smokers and ex-smokers who are willing to pay for the test themselves decide to participate in a study.

``Otherwise, five years will go by and millions of dollars will be spent, and we won't have anything to show for it,'' he said. ``And that would be the real travesty.''

For information about the UCSF study, call (415) 885-7387