Lung cancer screening test called exciting - Boston University team shows it is possible to detect precancerous changes in normal tissue

Lung cancer screening test called exciting - Boston University team shows it is possible to detect precancerous changes in normal tissue
By Judy Peres
Copyright © 2007, Chicago Tribune
Published March 7, 2007

Scientists may be close to being able to predict who will develop lung cancer, a development that could prevent tens of thousands of deaths a year in the U.S.

Using a molecular test called a microarray, a research team from Boston University has come up with an 80-gene "signature" that can identify lung cancer in smokers at a very early stage. More important, the team showed that it is possible to detect precancerous changes in normal tissue.

"It's like a molecular Pap smear," said Dr. Avrum Spira, who led the investigation.

The Pap test, which can detect abnormal cells in cervical tissue, is credited with greatly reducing deaths from cervical cancer. But instead of examining whole cells, as the Pap does, microarrays examine every gene encoded in the cell's DNA.

Although lung cancer kills more people than any other malignancy, research into the disease has lagged behind others. For example, genetic tests already can predict which breast cancer patients don't need chemotherapy and which are likely to respond to targeted drugs.

The new study suggests personalized medicine may now be coming to lung cancer.

Most cancers are diagnosed based on what cells look like under a microscope and are treated with the same handful of drugs. In personalized medicine, high-tech tests can spot individual genes, or patterns of genes, that make the cells become cancerous. New drugs can target cancer cells, avoiding normal cells and sparing the patient unnecessary side-effects.

"A molecular signature is far more informative than what a pathologist can see under the microscope," Spira said. "Two cancers might look the same, but one [patient] will be dead in two years and one will be fine.

"I think you'll see an explosion of molecular tools in the next five to 10 years for all forms of cancer and potentially other diseases," he said.

His preliminary study appears in this month's issue of the journal Nature Medicine.

The next step, Spira said, is a large clinical trial "to get FDA approval for an early detection tool."

That trial should get under way later this year, and the test could be available in 2009. Within four or five years, he said, it may be possible to identify who is at risk of developing lung cancer so doctors can take steps to prevent it.

Spira's work is based on a new concept in cancer research known as "field of injury."

In the case of smoking and lung cancer, the field of injury is the entire airway, from the nose to the lungs. "We hypothesize that all the cells that line your airway are affected at a genetic level by the toxins in the cigarette smoke," he said.

Rather than examining lung tissue, which would involve an invasive procedure, his research team sampled cells from high up in the bronchial tubes that were easy to reach. "We can use these cells to tell us how the smoker is responding to the exposure, whether he has lung cancer now, and whether he may be at risk for developing it," Spira said.

Dr. Ravi Salgia of the University of Chicago called the research exciting. "We're behind the curve" in early detection of lung cancer, he said. "With techniques like this, we should be able to catch up relatively quickly."

Dr. Kathy Albain of Loyola University Chicago said: "I believe they have identified a molecular profile in normal-appearing airway cells that coexists with cancer. If longitudinal follow-up validates the association between this field damage and subsequent development of cancer, we might have a new screening test."

About 90 percent of lung cancer in this country occurs in current or former smokers. However, fewer than 20 percent of smokers get the disease, and doctors can't predict which ones are at risk.

More than 200,000 Americans will be diagnosed with lung cancer this year, and more than 160,000 will die of the disease. Lung cancer is a major killer, Spira said, largely because it is rarely detected early enough to be treated successfully.

Spira said the gene signature could be used as an adjunct to existing tests, such as bronchoscopy or CT. In the Nature Medicine paper, the authors noted that bronchoscopy identified only 53 percent of smokers with lung cancer. When the authors added their gene tests, they were able to identify 95 percent of the cancers.

----------

jperes@tribune.com

Scans don't save lives, study says
By Judy Peres
Copyright © 2007, Chicago Tribune
Published March 7, 2007

Screening current or former smokers with annual CT scans does not appear to prevent deaths from lung cancer, a new study concludes.

The study, in Wednesday's edition of the Journal of the American Medical Association, reported that CT screening found three times more cancers than expected and resulted in nearly 10 times as many cancer surgeries. But there was no reduction in deaths from lung cancer.

The JAMA study, by Dr. Peter Bach of Memorial Sloan-Kettering Cancer Center, seems to contradict a study published last year in the New England Journal of Medicine.

That one, by Dr. Claudia Henschke of New York-Presbyterian Hospital, reported that CT scans produced a 10-year survival rate of 88 percent for patients with the earliest stage of disease. She argued that CT screening of high-risk people could prevent 80 percent of lung-cancer deaths.

In an editorial accompanying the JAMA study, Drs. William Black and John Baron of Dartmouth said the discrepancy between the two studies might be explained because they examined different things--one looked at how long patients lived from detection and the other counted deaths. Early detection does not automatically translate into delayed deaths, they said.

Screening tests (one used in people with no symptoms) are considered useful only if the test reduces death or disability from a disease. Otherwise, the tests could lead to unnecessary treatment that can have its own risks.

Two large, ongoing trials are expected to determine in the next year or two whether people who get regular CT scans are less likely to die of lung cancer than those who don't.