Older women with early-stage, invasive breast cancer had better survival rates than what was estimated by a popular online tool for predicting survival, according to researchers at the Duke Cancer Institute.
The finding provides a stronger rationale for women over the age of 70 -- even those who have additional minor health concerns -- to undergo aggressive treatments such as chemotherapy to prevent their cancer from returning.
“When making decisions about whether or not to use potentially toxic preventive chemotherapy for breast cancer in older women, patients and doctors debate what they should do,” said Gretchen Kimmick, M.D., M.S., an associate professor of medicine at Duke who is presenting the study findings at the San Antonio Breast Cancer Symposium. “This predictive model can help us show patients that they are going to survive long enough to see the benefit of treatment.”
Kimmick and colleagues used data from two breast cancer studies in the Alliance for Clinical Trials in Oncology (formerly called Cancer and Leukemia Group B) initiative funded by the National Cancer Institute. One of the trials enrolled women ages 65 and older to compare two chemotherapy regimens designed to prevent cancer recurrence; the other study enrolled patients to assess treatment and quality-of-life issues in older women who did not participate in the treatment trial.
The researchers tracked two-year survival rates among the study participants. They also plugged in the participants’ data -- including age, gender and ability to perform daily functions such as bathing and shopping for groceries -- to a popular online program called ePrognosis. Doctors often use the online tool to gauge whether treatments might be warranted in light of an individual patient’s potential lifespan.
For the breast cancer patients with good overall function, the ePrognosis tool predicted a 95 percent two-year survival rate, which matched exactly the observed overall survival of participants in the Alliance studies.
Among the women who had a moderate score, indicating poorer function, the ePrognosis tool underestimated survival, predicting an 88 percent chance of survival at two years vs. the 94 percent survival rate among the Alliance study patients.
Though the numbers were small in the group of older patients with the highest scores indicating poor functionality, the survival discrepancy was largest and the difference was statistically significant. The ePrognosis tool calculated survival for this group at 64 percent at two years, whereas the Alliance patients had an 81 percent survival rate at two years.
“Preventing cancer recurrence in older women is hugely important because if their cancer comes back, their survival is shortened by the cancer and quality of life is worse,” Kimmick said.
“Doctors and patients both need to understand the risks of cancer recurrence vs. chemotherapy and its side effects,” she added. “Having an accurate estimation of the chances of survival in general is critical information for older patients when weighing risks and benefits of preventive therapies. From what we see in this study, the survival for breast cancer patients is actually better than what is predicted by this tool in ePrognosis.”
Kimmick said additional research would assess the validity of the tools in predicting five- and 10-year mortality risks.
In addition to Kimmick, study authors include Brandelyn Pitcher, Jeanne Mandelblatt, Jonathan Clapp, Karla Ballman, Myra Barginear, Rachel Freedman, Andrew Artz, Heidi Klepin, Jacqueline Lafky, Judith Hopkins, Eric Winer, Clifford Hudis, Hyman Muss, Harvey Cohen, Aminah Jatoi and Arti Hurria.