During the early years of his fellowship, Dr. Charles Powell learned first hand how conventional treatment seemed to be able to do little for lung cancer patients. "It struck me that 12 months after discussing a patient's case, I'd frequently find that the patient had relapsed or died," says Dr. Powell, assistant professor of medicine at P&S. "Basically, lung cancer survival rates are dismal."
Unlike more common breast and prostate cancers, lung cancer survival rates have remained essentially unchanged for decades. Between 1960 and 1990, five-year survival rates for breast cancer went from 60 percent to 82 percent, while rates for lung cancer barely rose from 8 percent to 14 percent. The high mortality rate makes lung cancer America's No. 1 cancer killer, taking the lives of more people each year than breast, prostate and colon cancers combined.
Late detection may be part of the reason lung cancer remains such a potent killer. About half of all lung cancers have already spread by the time the cancer is found, and the tumors can't be surgically removed.
Finding cancers before they spread could help, yet a study of X-ray screening in the 1970s showed that finding cancer early didn't change mortality. Spiral CT screening may prove to do a better job, and a study of CT screening is currently under way.
But it's also possible, Dr. Powell says, that some lung cancers may be deadlier than others, even if detected early. If lung cancer is really many different diseases, as oncologists suspect, should all patients get the same treatment, as they do now?
Research by Dr. Powell and others on genetic differences among lung tumors indicates the answer is probably no, and nowhere is that more apparent than in the differences Dr. Powell has found between the cancers of smokers and nonsmokers.
Smokers account for most lung cancer cases, but about 15 percent occur in people who have never even tried a cigarette. Working with Dr. Alain Borczuk, associate professor of clinical pathology, and researchers at Boston University, Dr. Powell has found that the tumors in smokers are different from those in nonsmokers.
"We found that smokers have wide damage throughout the lung. Their cells are already halfway toward malignancy," Dr. Powell says. "In contrast, most lung tissue in nonsmokers has fewer genetic changes and lung cancer in nonsmokers appears to be a local event."
In addition to developing in a different background, the smoker's tumors also use different genes to grow. "It shows that cancer in smokers is really a different disease than cancer in selected nonsmokers," Dr. Powell says. "And it suggests that in some cases, secondary smoke may not be the culprit in the lung cancers of nonsmokers."
Now the researchers are focusing on what changes in a smoker's lungs are responsible for turning an already damaged cell into a malignant one. "All smokers probably have these pre-malignant changes, but only 10 percent of heavy smokers will ever develop lung cancer," Dr. Powell says.
When Dr. Powell and his postdoc Dr. Han Kim looked over their gene array results, one difference stood out: Smokers with cancer tend to lose activity of a gene, Glypican 3, a protein involved in regulating cell growth. The same gene also has been linked to other types of cancer.
In lungs, the researchers found, Glypican 3 is turned off in smokers' cells even before the cells become malignant. Once a cancer develops, the gene's level of activity is associated with tumor size. The less active the gene is, the larger the tumor.
Just why Glypican 3 inactivity promotes tumor growth in smokers isn't entirely clear, but Dr. Powell's latest work suggests that turning the gene off prevents damaged cells from committing suicide. Cells with irreparable DNA damage usually kill themselves through apoptosis, but Dr. Powell finds that human lung cancer cells with inactive Glypican 3 can't do that and, instead, continue to grow and accumulate more damage that leads to malignancy.
One of the potential benefits of learning more about Glypican 3 may be a more effective cancer-screening program for smokers if an inactive Glypican 3 gene turns out to be a good warning signal of the disease. Dr. Powell has already found that Glypican 3 is inactivated when the gene's promoter is covered in methyl groups and is now trying to find out if this hypermethylation puts smokers at risk of lung cancer.
Even though some lung cancers may be more aggressive than others, "we think we have a better chance if we detect them early," Dr. Powell says. "Finding hypermethylation in these patients would make them good candidates for screening and cancer prevention trials."
There are many implications for lung cancer patients in the future, Dr. Powell says. "With our findings there are opportunities to intervene. We're trying to take this back to the patient as soon as we can."