Incidence of Breast Lymphedema and Predictors of its Development in Patients Receiving Whole Breast Radiation Therapy Alone after Breast-Conservation Surgery

Abstract

Breast-conserving surgery followed by whole breast radiation therapy (WBRT) is the standard of care in the management of early stage breast cancer. A common adverse effect of WBRT is lymphedema. While arm lymphedema has been well-characterized, there has been less investigation into breast lymphedema. Patients receiving breast surgery at our institution are seen preoperatively and then followed in a lymphedema prevention clinic, where development of arm or breast lymphedema is documented. We sought to characterize rates of breast lymphedema in patients receiving WBRT and identify potential predictors in its development. A total of 243 consecutive patients treated with lumpectomy and WBRT from January 2016 – June 2017 were included. Patients with supraclavicular radiation were excluded. All patients were seen in our lymphedema clinic at baseline and for at least one follow-up measurement. Breast lymphedema grades were assigned by trained lymphedema nurses. Patient demographic and treatment data was abstracted from the electronic medical record. Univariate logistic regression models were generated to identify factors associated with lymphedema; two-tailed p-values were reported and a p-value of <0.05 considered significant. Statistical analysis was performed with SAS 9.3 software (SAS Institute, Cary, NC). Median patient age was 62 (range 31-90). Median follow-up from surgery was 15.3 months. 44 patients were diagnosed with lymphedema of the breast (18.1%). Rates of grade 1, 2 and 3 breast lymphedema were 93%, 7%, and 0%, respectively. Median time to diagnosis of breast lymphedema after surgery was 7.9 months. On univariate analysis, her-2 positive patients had a 35.0% rate of breast lymphedema, significantly higher than hormone positive (17.9%) and triple negative (6.9%) patients (p=0.02). The association between her-2 positive subtype and development of breast lymphedema remained significant on multivariate analysis (p=0.04). There was no association of breast lymphedema with age, T stage, use of hypofractionated radiation, radiation boost, number of lymph nodes removed, oncoplastic reduction, or arm lymphedema. In patients without a sentinel lymph node procedure, there were no reports of arm lymphedema, but breast lymphedema rate was 13%. Several her-2 negative patients received trastuzumab on clinical trial. Use of trastuzumab was not associated with breast lymphedema, nor was use of chemotherapy in general or any specific drug or drug class, such as taxanes. Breast lymphedema is a distinct entity from arm lymphedema, and a common finding in women treated with breast conservation. While arm lymphedema is often attributed to radiation or axillary surgery, our findings suggest that the etiology of breast lymphedema is related to tumor factors, as her-2 positive subtype was the only significant predictor in our population. More research is needed to investigate the biologic mechanisms responsible for this finding.