New tool incorporates race to better predict risk for breast cancer-related lymphedema
Click Here to Manage Email Alerts
Researchers developed and evaluated streamlined models for predicting and preventing breast cancer-related lymphedema.
The new models are designed to address racial disparities in lymphedema risk that are not accounted for in current risk prediction models.
“Lymphedema is one of the most pressing breast cancer survivorship issues. About 17% of patients develop lymphedema after surgery for breast cancer,” study lead author Danielle H. Rochlin, MD, a plastic and reconstructive surgeon at Memorial Sloan Kettering Cancer Center, said in an interview with Healio. “Considering that female breast cancer is the most frequently diagnosed cancer worldwide, with approximately 2.3 million new cases annually, this is a pervasive problem. For patients, lymphedema is very burdensome; it has been associated with diminished health-related quality of life, particularly in the later stages, when infection and physical impairment become more pronounced.”
Healio spoke with Rochlin about the impetus for developing the new models, their accuracy in predicting breast cancer-associated lymphedema, and the potential implications of these models for breast cancer treatment and survivorship.
Healio: What inspired you to develop a more effective model for predicting which patients are at highest risk for breast cancer surgery-associated lymphedema?
Rochlin: We know that early diagnosis and management of lymphedema can prevent the progression of disease, so identification of high-risk patients becomes a very important task. Existing models to predict patients who are at risk for developing lymphedema omit key factors, such as race and sentinel lymph node biopsy, and/or use obscure datapoints that are not readily available in a patient-facing setting, such as radiation field and number of cycles of chemotherapy in the affected arm. Our goal was to create simple and accurate prediction models that could be easily deployed to identify at-risk patients.
Healio: How did you develop this new model?
Rochlin: We analyzed a cohort of nearly 2,000 women with breast cancer from both Memorial Sloan Kettering and Mayo Clinic. We developed two models to predict the risk of lymphedema — one to use before a patient has surgery, and a second to use after a patient has surgery. We only considered variables that are easily available to patients and physicians, such as age, height, weight or history of radiation. We then conducted statistical analyses to determine the combination of variables in each setting that yielded the most accurate model.
Healio: How do current risk prediction models fail to represent racial disparities in lymphedema risk among patients treated for breast cancer?
Rochlin: Current models do not include race as a variable. We know that there are differences in the risk for lymphedema among racial groups, so omitting race as a variable in prediction modeling effectively fails to represent racial disparities in lymphedema risk.
Healio: How does your model correct this?
Rochlin: We include race as a variable in our preoperative prediction model. Though we know that Black race and Hispanic ethnicity are independent risk factors for breast cancer-related lymphedema, we still do not know why these groups have an increased risk. Additional research is needed to identify the factors underlying this association.
Healio: How has the model performed?
Rochlin: Very well. Our preoperative model identifies patients who have at least an 18% risk of developing lymphedema with 73% accuracy. Our postoperative model identifies patients who have at least a 10% risk with 81% accuracy. Though we chose these probabilities as the optimal cutoffs based on our analysis, users of our predictive models can adjust the probability threshold to suit their specific purpose in terms of sensitivity and specificity.
Healio: How might the new risk prediction model also benefit patients at lower risk for lymphedema?
Rochlin: Typically, [patients with breast cancer] are followed closely after surgery with circumferential arm measurements since, as mentioned before, early diagnosis and management of lymphedema can prevent the progression to late-stage disease that has a lot of associated morbidity. This takes up a lot of patient and physician time, and health care resources. With our postoperative model, patients can be quickly screened remotely at frequent intervals since the only patient input required is an answer to a question about the presence of arm swelling. Patients who screen positive or high risk can be called into clinic for arm measurements, while those who screen negative or low risk can stay home and avoid a clinic visit.
For more information:
Danielle H. Rochlin, MD, can be reached at Memorial Sloan Kettering Cancer Center, 1275 York Ave., New York, NY 10065; email: rochlind@mskcc.org.
Collapse
Click Here to Manage Email Alerts