Patients with documented exposure to a GLP-1 receptor agonist or related incretin-based agent — semaglutide, liraglutide, dulaglutide, exenatide, lixisenatide, albiglutide, or tirzepatide — formed the exposed cohort, while patients meeting the same surgical criteria without such exposure served as comparators. Because patients who take GLP-1 drugs differ systematically from those who do not — they are more likely to have obesity, diabetes, and other metabolic disease — the team used one-to-one propensity score matching to balance measured baseline characteristics between the two groups. Before matching, the eligible cohorts comprised 30,970 GLP-1 users and 376,922 non-users.

After matching, 61,630 patients remained, split evenly at 30,815 per group. The matched cohorts were closely aligned on demographics: mean age was 62.1 years in the exposed group versus 62.5 years in the comparator group, both were 99.2 percent female, and standardized mean differences for age, sex, and race fell below 0.1. Roughly 71 to 72 percent of patients in each group were White and 18 percent were Black or African American.

The primary outcome was incident coded postmastectomy lymphedema syndrome, captured by the ICD-10-CM code I97.2, assessed from one day after the index event through 650 days of follow-up. Patients who already carried the diagnosis code before the outcome window were excluded to avoid counting prevalent cases. After these exclusions, 29,259 GLP-1 receptor agonist users and 30,237 non-users entered the primary analysis.

Coded postmastectomy lymphedema occurred in 258 of the users and 779 of the non-users — incidence rates of 0.88 percent and 2.58 percent, respectively. The statistical signal was striking. In time-to-event analysis using Kaplan–Meier methods and Cox proportional hazards models, GLP-1 receptor agonist exposure was associated with a hazard ratio of 0.373 for coded postmastectomy lymphedema, with a 95 percent confidence interval of 0.324 to 0.429 and a p value below 0.001.

The absolute risk difference was minus 1.7 percent, with a confidence interval of minus 1.9 to minus 1.5 percent, which translates to an estimated number needed to treat of 59 — meaning that for every 59 patients exposed, roughly one fewer coded lymphedema case would be expected over the follow-up period. The risk ratio was 0.342 and the odds ratio 0.336. Lymphedema-free survival at 650 days was 98.87 percent in the exposed cohort versus 97.16 percent in the comparator cohort.

A secondary analysis broadened the lens to any coded lymphedema, using the ICD-10-CM code I89.0 for lymphedema not elsewhere classified. Here, 1,021 of 25,926 GLP-1 users developed the outcome compared with 2,454 of 29,092 non-users, corresponding to incidence rates of 3.9 percent and 8.4 percent. The association again favored the exposed group, with a hazard ratio of 0.507 (95 percent CI 0.472 to 0.546; p < 0.001).

The consistency of the direction of association across both the narrower postmastectomy definition and the broader lymphedema outcome strengthened the case that the finding was not an artifact of a single diagnostic code. Why would drugs designed to mimic a gut hormone and drive weight loss plausibly protect the lymphatic system? The biological rationale rests on the growing understanding that lymphedema is not simply passive fluid accumulation but a chronic inflammatory and fibroadipose disorder of the lymphatic microenvironment.

Obesity is a recognized risk factor for breast cancer-related lymphedema, and proposed mechanisms include increased lymphatic load, adipose tissue inflammation, impaired lymphatic pumping, lymphatic endothelial dysfunction, and progressive fibroadipose remodeling. GLP-1 receptor agonists produce substantial weight loss, improve glycemic control, and have demonstrated cardiovascular benefit in selected populations. Beyond weight reduction, they may exert anti-inflammatory effects, including reduced macrophage infiltration and modulation of inflammatory signaling in adipose tissue.

The new findings also align with earlier, smaller observations: a case report described improvement in breast cancer-related lymphedema after GLP-1 initiation, and a subsequent retrospective study suggested reduced cancer-related lymphedema risk after axillary lymph node dissection among users. Yet the authors are emphatic that these results are hypothesis-generating, not proof of prevention, and the list of caveats is long. First, as an observational study, it cannot establish causality.

Second, the outcome was based on diagnosis codes rather than prospective surveillance, limb-volume measurements, bioimpedance spectroscopy, or clinical adjudication — and the coded incidence observed here was substantially lower than the roughly 8 to 10 percent rates reported in prospective surveillance studies even after sentinel lymph node biopsy, suggesting that coded diagnoses likely underestimate true lymphedema incidence. If undercoding differed between users and non-users, the association could be biased; even if undercoding was nondifferential, reliance on codes limits interpretation of the absolute event rates. Further limitations compound the uncertainty.

GLP-1 exposure was treated as a binary variable, so dose, duration, adherence, persistence, discontinuation, indication, and cumulative exposure were invisible to the analysis, and the timing of drug initiation relative to surgery and lymphedema diagnosis could not be adjudicated at the individual level. Residual imbalance persisted after matching in key metabolic variables — body mass index, body weight, and hemoglobin A1c were all higher in the GLP-1 cohort — raising the possibility of confounding by indication. The analysis also could not fully account for breast cancer stage, radiation fields, the number of lymph nodes removed, reconstruction type, endocrine therapy duration, chemotherapy regimen, taxane exposure, or longitudinal weight change, and the cohort was not restricted to mastectomy patients alone, which may have diluted the observed incidence.

The 650-day follow-up window may miss lymphedema cases that emerge years after treatment, and procedure-specific, medication-specific, and obesity- or diabetes-stratified analyses were not performed. If prospective studies with objective lymphedema measurements, detailed oncologic treatment variables, and longitudinal weight data confirm the relationship — and if randomized or pragmatic intervention trials ultimately show that GLP-1 receptor agonists reduce clinically confirmed postmastectomy lymphedema — metabolic therapy could become a genuine frontier in supportive oncology. For now, the message from the data is measured: among more than 61,000 matched breast cancer patients with mastectomy or axillary procedure exposure, those on GLP-1 receptor agonists developed far fewer coded cases of lymphedema over 650 days, and understanding whether the drugs themselves deserve the credit is the urgent next question.

Subject of Research: Association between GLP-1 receptor agonist use and postmastectomy lymphedema risk in breast cancer survivors

Article Title: Glucagon-like peptide−1 receptor agonist use and risk of postmastectomy lymphedema in breast cancer survivors: a multicenter real-world cohort study

Article References: Yesho, D. H., Megiso, M. Z., Ugwu, C., Neely, A. N., Verinumbe, T., Varadi, G., & Jones, C. (2026). Glucagon-like peptide−1 receptor agonist use and risk of postmastectomy lymphedema in breast cancer survivors: a multicenter real-world cohort study. Supportive Care in Cancer, 34(11), 1159. https://doi.org/10.1007/s00520-026-11405-3

Image Credits: AI Generated

DOI: 10.1007/s00520-026-11405-3

Keywords: GLP-1 receptor agonists, breast cancer, lymphedema, mastectomy, semaglutide, obesity, survivorship, electronic health records, propensity score matching, supportive oncology, retrospective cohort study, inflammation