To Have and Have Not
Cancer patients are either haves or have-nots: Some develop tumors that have an early-screening test; others get ones that don't. The haves boast a good track record. Just look at the many cures that have come with the early detection of prostate and breast cancer. But lung cancers are have-nots. Only 15 percent of those getting this diagnosis will survive past five years. That's why the new report of the International Early Lung Cancer Action Program that appeared in the New England Journal of Medicine is such a shining light. It points to a way at last that might cure or substantially lengthen the lives of most of the 163,000 patients lost each year to this killer.
The study, conducted at 38 centers around the world for the past 13 years, screened 31,567 symptom-free people at high risk for lung cancer, mostly because of their smoking. Using a rapid CT scan designed to deliver low-dose radiation equivalent to that of a mammogram, the researchers detected many cancers-most in the early stage of development. Regardless of stage or treatment, the group overall had an estimated 80 percent 10-year survival rate. Patients with early tumors promptly removed by surgery had over 90 percent.
Imagine what this could mean. Lung is our nation's biggest cancer killer, swamping the number of patients who die from cancers of the breast, prostate, colon, and cervix combined. And as it advances, lung cancer robs patients of their breath and spreads into places like the heart and blood vessels, and to distant parts of the body like bones and the brain. What frustrates doctors is that the gold standard for diagnosis of lung disease, the chest X-ray, misses 85 percent of stage 1 cancers. CT scans, however, detect most lung cancers when they are early, small, and curable. That's because they now can find spots or nodules smaller than a tenth of an inch, with a resolution that lets trained eyes assess nodule growth, millimeter by millimeter.
That this approach uses nodule growth seems missed by many who fear early screening will lead to unnecessary lung biopsies. In the I-ELCAP protocol, growth is the major trigger, along with size and shape, for getting a biopsy. Thus, only 535 of the 5,646 patients with positive scans had them. When the biopsies were done, all but 43 showed cancer. I can't believe any doctor would not want to biopsy a growing mass so likely to be cancerous.
Others worry that small lung cancers, even those growing before a doctor's eyes, may not be dangerous since there was no control group. That is, patients with a similar number of early, undetected (and untreated) cancers might have done as well. This ignores what we know: The smaller the size of the cancer, the better the patient survival. It also supposes that informed patients would join an early-detection study that offers only a 50 percent chance of being screened. Or that it's A-OK to bill a study as a cancer action program when the game plan for half its subjects is "don't look, don't find."
Adopters. The controlled study that needs to be done, says Claudia Henschke, radiologist at Cornell's Weill Medical College and head of I-ELCAP, is one that compares a lumpectomy for these smaller cancers with the standard lobectomy. As doctors find more small tumors, it might be possible to remove less lung tissue without affecting survival rates. The more lung you leave behind, the better the quality of life.
We need guidelines, says Henschke, for doctors and patients already seeking this approach, as well as identification of centers with the technology and expertise. The interest is likely to grow as the Lung Cancer Alliance, a patient advocacy group, also supports early screening. Kenneth Cowan, director of the University of Nebraska's Eppley Cancer Center, recently launched what may be the first statewide screening program for heavy smokers using the I-ELCAP model. This will allow a comparison with unscreened patients in Nebraska's tumor registry-all states keep tabs on cancers in their populations-providing a kind of control group.
Long ago as an intern at Johns Hopkins Hospital, I admitted a patient for a blood pressure problem and noticed he had a cough at night. I rechecked his chest X-ray; it was perfect. Then I sent his sputum to the lab; it turned up cancer. My residents and I thought we'd saved his life, but it was more advanced than we realized, and he died within the year. Decades later we may finally have an early-detection option that would have helped him and the many millions who have been lost to lung cancer since then. Let's get on with it.
This story appears in the November 13, 2006 print edition of U.S. News & World Report.