A study appearing in this week's issue of The Lancet computes 20-year survival estimates that are 1 percent to 11 percent higher for a range of cancers when calculated with a new method that uses up-to-date computer programs.
"This is really good news and optimistic. It shows that in recent years more cancer patients are living longer," says Dr. Ruth Oratz, an associate professor of medicine at New York University School of Medicine in New York City. "We are making progress in the war against cancer."
Cancer patients also should be buoyed by the findings.
Dr. Allan Novetsky, director of medical oncology at Maimonides Medical Center in Brooklyn, N.Y., says, "As a practicing oncologist, I am often faced with trying to convey to patients what the current prognosis is for their disease. And the more accurately I can tell a patient what the probability of them living five, 10, 15 or 20 years is, the more I am able to help them accept treatment, understand what the impact is and make plans for their short-term and long-term needs."
In the new study, Dr. Hermann Brenner, scientific director of the German Centre for Research on Aging in Heidelberg, Germany, compared survival estimates using the relatively new "period analysis" method and the traditional "cohort method." The cohort method involves looking at longevity in patients who were diagnosed with cancer many years ago. Period analysis uses more recent data, thereby, theoretically, reflecting advances in detection and treatment.
Brenner compared survival estimates obtained with period analysis for the year 1998, and cohort analysis for patients diagnosed between 1978 and 1993. He used data from the U.S. National Cancer Institute's Surveillance, Epidemiology and End Results program.
With period analysis, estimates of five-year, 10-year, 15-year and 20-year survival rates for all types of cancer were 63 percent, 57 percent, 53 percent and 51 percent, respectively. This represented increases of 1 percent, 7 percent, 11 percent and 11 percent, respectively, over the cohort-based analysis.
Period analysis also showed 20-year survival rates that were nearly 90 percent for thyroid and testicular cancer, greater than 80 percent for melanomas and prostate cancer, about 80 percent for endometrial cancer and almost 70 percent for bladder cancer and Hodgkin's disease. The 20-year survival rate for breast cancer was estimated to be 65 percent, for cervical cancer 60 percent and for colorectal, ovarian and kidney cancer, 50 percent.
Is this just an exercise in statistical sophistry?
It depends on whom you ask. Some experts versed in the art and science of numbers are skeptical of the study's value.
"I think it's similar to people in Congress wanting to change the way they define poverty," says Andre Rogatko, chairman of biostatistics at Fox Chase Cancer Center in Philadelphia. "You're not solving the problem. You're just changing your perception."
Oncologists have a different perception.
Novetsky points out the study could not only influence the ability to obtain funding for research, it also helps people focus on where progress has occurred (for example, prostate cancer) and where more progress still needs to take place (lung cancer, for instance).
Then there are the patients. "Being able to encourage a patient is a tremendous blessing," Novetsky says. "One of the most common questions we get asked is, 'What are my chances?' 'What are my chances of being alive to see my son's wedding or my daughter's graduation?' "
"The more hope you can offer a patient, the better they deal with their disease, the more likely they are to accept the side effects of treatment because they know that there is a much stronger probability of being successful," he says.
What To Do