ven as scientists continue to search for a better understanding of cancer, some existing knowledge may be pushed further. A new Columbia Health Sciences statistics-based study finds that elderly patients with advanced, but non-metastatic, colon cancer live longer if they get chemotherapy after surgery. The problem is that 50 percent of such patients do not get the drugs.
In the study, the investigators analyzed data on elderly patients who had colon cancer that had spread to the lymph nodes (stage III colon cancer) and were treated with surgery alone and with the surgery plus the drug 5-fluorouracil (5-FU). This year approximately 43,000 Americans will get node-positive colon cancer, two-thirds of whom are older than 65. The researchers found that node-positive patients with both treatments had a 30 percent reduction in mortality at five years after diagnosis compared with individuals who received surgery alone. The increased survival rate for those elderly who had both treatments was similar to that for patients under 65 who received both therapies.
The researchers used data from the National Cancer Institute for 4,768 patients who were at least 65 years old and who were diagnosed with node-positive colon cancer between 1992 and 1996. Approximately half received surgery and the other half surgery and 5-FU. After performing statistical analyses, they found the mortality rate for those who received surgery alone was approximately 50 percent to 60 percent at five years, compared with approximately 30 percent to 40 percent for those who received both treatments.
The findings were published in the March 5 issue of the Annals of Internal Medicine.
These and other comparable studies should allow doctors and patients to make better decisions about treatment for colon cancer, the researchers say. Age should not be a deterrent for doctors and patients making the decision to treat someone with both surgery and chemotherapy, says Dr. Alfred I. Neugut, principal investigator for the study and professor of medicine at P&S and professor of epidemiology at the Mailman School of Public Health. Although the elderly often do not get treated with both surgery and chemotherapy, our study shows that they benefit from both.
Historically, doctors may not have treated the elderly with both surgery and chemotherapy because clinical trials from the 1980s showing the effectiveness of both therapies had been performed on younger people, the researchers surmise. Doctors rely on the medical literature to make treatment decisions.
Further, few clinical studies have been performed that compare surgery with surgery and chemotherapy in the elderly population. Future clinical trials should not have upper age restrictions, Dr. Neugut says.
Physicians also may have been fearful of offering chemotherapy in addition to surgery to elderly patients because older people tend to have other health problems. But the Annals study shows that in a sample of patients representative of the population, both treatments benefit. Doctors still have to assess, though, whether elderly patients might have other conditions in which chemotherapy might be contraindicated, Dr. Neugut says.
A problem with the study, Dr. Neugut says, is that the patients were not randomized to either the surgery or the surgery and drug group because the analysis was based on already existing data. To try to approximate a randomized study, the researchers used propensity scores, a way to eliminate the potential bias that would occur if patients who received chemotherapy and surgery were different, i.e., healthier, than patients who received only surgery, says Dr. Daniel F. Heitjan, professor of biostatistics at Mailman and one of the investigators in the study.
The propensity scores take into account race, age, gender, and other characteristics of the population studied. The statisticians divided the patients into five groups in order of who was most likely to get chemotherapy. Each group had patients from both the surgery only and the surgery and 5-FU groups so the patients inside each group could be compared with each other in terms of survival. By comparing the patients most similar to each other, most of the bias was removed, Dr. Heitjan says.
Still, patients who received the 5-FU may have been generally healthier, in ways that the researchers could not determine, than patients who did not get 5-FU.
Understanding the data also was important in doing the study. Dr. Victor Grann, associate clinical professor of medicine at P&S and in public health at Mailman, traveled to the NCI to become proficient in the colon cancer database. The NCI database is powerful because it gives insight into how many patients were treated, what that treatment was, and what the actual benefit was, Dr. Grann says. The database shows the effectiveness of clinical trials and cancer care guidelines.
Other Columbia Health Sciences researchers who participated in the study were Dr. Vijaya Sundararajan, formerly an associate research scientist at Mailman, and Nandita Mitra, formerly a graduate fellow in biostatistics at Mailman.
As the population ages, the number of new cases of cancer will increase, putting greater emphasis on geriatric patient care. Maybe as people live longer there will be a greater willingness to treat the elderly more aggressively, Dr. Neugut says.
Matthew Dougherty contributed to this report.