Risk Factors and Racial and Ethnic Disparities in Patients With Breast Cancer–Related Lymphedema

Abstract

Risk factors for breast cancer-related lymphedema (BCRL) after axillary lymph node dissection (ALND) are poorly understood. To evaluate rates of and risk factors associated with BCRL in a prospective cohort of women treated with ALND. This prospective BCRL screening study performed at a tertiary cancer center enrolled women with breast cancer 18 years and older undergoing breast surgery and unilateral ALND in the primary setting or after sentinel lymph node biopsy. Risk of BCRL during the first 2 years after ALND and radiotherapy. Patients were prospectively evaluated with arm volume (perometer) measurements, and BCRL was defined as a relative volume change of 10% or greater from baseline. Cumulative incidence of BCRL was assessed using competing risk analysis. Risk factors for BCRL were assessed on univariate and multivariable analyses. From November 2016 to March 2020, 304 patients were enrolled; 276 had at least 1 longitudinal measurement. Median (IQR) age was 48 (40-57) years; median (IQR) body mass index, calculated as weight in kilograms divided by height in meters squared, was 26.4 (22.5-31.2). Of the 276 patients included in the analysis, 29 (11%) self-identified as Asian, 55 (20%) as Black, 16 (6%) as Hispanic, 166 (60%) as White, and 10 (3%) as unknown race and ethnicity; 70% received neoadjuvant chemotherapy (NAC); 93% received nodal irradiation. The 24-month BCRL rate was 23.8% (95% CI, 17.9%-29.8%), with significant variation by race and ethnicity (24-month rate: 37.2% [Black], 27.7% [Hispanic], 22.5% [Asian], and 19.8% [White]; P = .004). The BCRL rates were also higher among patients receiving NAC vs up-front surgery (24-month rate: 29.3% vs 11.1%; P = .01). On multivariable analysis, Black race and Hispanic ethnicity (compared with White race) (odds ratio [OR], 3.88; 95% CI, 2.14-7.08 and OR, 3.01; 95% CI, 1.10-7.62, respectively; P < .001 for each), receipt of NAC (compared with up-front surgery) (OR, 2.10; 95% CI, 1.16-3.95; P = .01), older age (OR, 1.04; 95% CI, 1.02-1.07 per 1-year increase; P = .001), and a longer follow-up interval (OR, 1.57; 95% CI, 1.30-1.90 per 6-month increase; P < .001) were independently associated with an increased risk of BCRL, while ERBB2-positive subtype was associated with a decreased risk of BCRL (compared with hormone receptor positive/ERBB2 negative): OR, 0.50; 95% CI, 0.23-0.99; P = .04). In this cohort study, Black race, Hispanic ethnicity, NAC receipt, older age, and longer follow-up were independently associated with risk of BCRL. Studies are warranted to evaluate the biologic mechanisms behind racial and ethnic disparities in BCRL development and alternatives to NAC to avoid ALND in tumor subtypes unlikely to achieve nodal pathologic complete response.