Tool Calculates Competing Risks of Death in Older Women With Early Breast Cancer

By Marilynn Larkin

April 16, 2019

NEW YORK (Reuters Health) - A risk calculator incorporating age, comorbidities and estrogen receptor status can estimate competing risks of death for older women with stage 1 breast cancer, potentially aiding therapeutic decision making, researchers say.

Dr. Nabil Wasif of Mayo Clinic Arizona told Reuters Health by email, "Our ability to cure early-stage breast cancer is approaching 100% with modern surgery, radiation and chemotherapy. However, while such treatments are effective, there may be collateral damage that the patient has to live with for the rest of her life, as a constant reminder of the devil's bargain made to treat the cancer."

"Further complicating the picture is that the majority of breast cancer is diagnosed in women who are 65 years or older," he said. "These patients often have other health problems that may present a bigger threat to long-term survival than an early breast cancer picked up on a screening mammogram."

Although many physicians will consider these comorbidities, "there is not a lot of science to guide decision making," he noted. "This study is an attempt to put survival from breast cancer into context by considering all the medical conditions a patient may have - a holistic approach as opposed to a breast cancer-centric one," he said.

Dr. Wasif and colleagues studied 7,220 women (mean age, 74; 87% white) with stage I adenocarcinoma of the breast treated from 2004 to 2012. Most tumors (84%) were estrogen receptor-positive. Three competing mortality outcomes were analyzed: dead of disease (DOD), dead of other cancers (DOC) and non-cancer death (NCD).

As reported online March 28 in the Journal of the American College of Surgeons, at a median follow-up of 50 months, most patients were alive (86%); 3% had died of breast cancer, 2% of other cancers and 10% from other causes.

The overall cumulative incidence of mortality was 4.9% for DOD, 3.7% for DOC and 21.3% for NCD. The five- and eight-year probability of DOD was 3% and 4.7%; of DOC, 1.9% and 3.5%; and of NCD, 9.8% and 18.9%. As expected, the probability of DOC and NCD rose with increasing age.

However, an increase in the probability of DOD with age was also seen in patients older than 81, who had a 7.4% chance of DOD at eight years compared with 3.1% for patients 65-70 years old.

Blacks had a higher probability of DOD compared to whites or other ethnicities. The presence of any major comorbidity (e.g., cardiovascular or neurological disorders) significantly increased the probability of an NCD.

Estrogen receptor status was the strongest predictor of DOD for cancer-related variables: five-year mortality, 2.4% for ER+ versus 6.8% for ER-; eight-year mortality, 4% for ER+ versus 9.2% for ER-.

"Given patient age, comorbidity and estrogen receptor status, an estimate of competing risks of death from DOD, DOC and NCD can be calculated," the authors say.

"For example," they write, "a 70-year-old woman with no comorbidity, 'Other' race designation and an ER positive Stage I breast cancer has an 8-year probability of 2.0% for DOD, 2.4% for DOC and 7.5% of NCD. For a 70-year-old woman with cardiac and neurologic comorbidity, white race and an ER negative tumor these probabilities are 9.3% for DOD, 4.2% for DOC and 15.3% for NCD."

They added, "Calibration plots showed that the risk models were fairly accurate . . . (and) the predicted scores are generally reflective of the actual risk."

"Future research will build on this initial work to look at ways to improve on the performance of our predictive model," Dr. Wasif said.

Dr. Tracey O'Connor, an associate professor of medicine at Roswell Park Comprehensive Cancer Center in Buffalo, New York, told Reuters Health that the study, "builds on knowledge we already have that older patients need to have therapy decisions made in the context of their function and co-morbidity."

"In my mind, this gives additional support to the routine use of geriatric assessment tools in cancer patients to help with decision making," she said by email. "It also supports research exploring less invasive treatment options in selected individuals with low-risk breast cancer."


J Am Coll Surg 2019.