2793 - Investigating the Association of Metabolic and Cardiovascular Medications with Volume Reduction after Complete Decongestive Therapy in Breast Cancer-Related Lymphedema
Presenter(s)
N. Lum1, E. Yu1, Z. A. Liu2, P. Wong1,3, and J. Y. Y. Kwan1,3; 1Radiation Medicine Program, Princess Margaret Cancer Centre, University Health Network, Toronto, ON, Canada, 2Department of Biostatistics, Princess Margaret Cancer Centre, University Health Network, Toronto, ON, Canada, 3Department of Radiation Oncology, Princess Margaret Cancer Centre, University of Toronto, Toronto, ON, Canada
Purpose/Objective(s): Breast cancer–related lymphedema (BCRL) is characterized by fibrosis in late stages, which may reduce responsiveness to complete decongestive therapy (CDT). Medications including metformin, statins, angiotensin receptor blockers (ARBs), and GLP-1 receptor agonists may exert antifibrotic effects. We hypothesized medication exposure would be associated with lower baseline excess volume and improved percent reduction of excess volume (PREV) following CDT.
Materials/Methods: We retrospectively reviewed unilateral BCRL patients treated with CDT (2018–2026) with >5% excess arm volume at baseline. Patients with bilateral/subclinical BCRL, concurrent cellulitis, or prior lymphedema surgery were excluded. Patients exposed to medications of interest at baseline were compared to controls without exposure. Outcomes included baseline excess volume (mL) and PREV following CDT, with longitudinal PREV at 2y (n=451) and 5y (n=458) where available. Univariable analyses used false discovery rate (FDR) correction per outcome. Multivariable modeling used ridge regression with 95% CIs via 1000 bootstrap resamples adjusted for confounding clinical factors: age, BMI, breast density, axillary lymph node dissection (ALND), and pathologic lymph node (LN) burden.
Results: 471 patients were included (mean age 59.7±12.0y, BMI 28.6±6.0kg/m², baseline excess volume 362.0±331.2mL, BCRL duration 5.1±5.2y, follow-up 2.2±1.7y). Drug-exposed patients (n=138) were compared with controls (n=333). Medication exposures included metformin (n=42), statins (n=84), ARBs (n=17), and semaglutide (n=9). In FDR-adjusted univariable analyses, greater baseline excess volume was associated with greater age (r=0.147; p=0.006), BMI (r=0.208; p=0.001), and pathologic LN burden (r=0.129; p=0.02). Medication exposures were not significantly associated with baseline excess volume. Greater pathologic LN burden was associated with reduced PREV following CDT (r=-0.132; p=0.041), indicating poorer response, and with lower PREV at 2y (r=-0.150; p=0.024) and 5y (r=-0.138; p=0.039). ALND was associated with reduced PREV following CDT (-53.7%; p<0.001) and at 5y (-35.7%; p=0.006). Metformin (+33.3%; p=0.041), ARBs (+42.6%; p=0.046), and semaglutide (+68.3%; p=0.029) were associated with greater PREV following CDT; however, these associations attenuated in multivariable ridge regression and bootstrap 95% CIs crossed 0, consistent with adjustment for correlated clinical predictors and small exposure subgroup sizes.
Conclusion: Greater pathologic LN burden was associated with higher baseline excess volume and poorer CDT response. Medication exposures showed significant univariable PREV improvements but not independent multivariable effects. Further prospective studies in larger cohorts are warranted.