Logo image
Back
Peripheral immune profiling during neoadjuvant HER2-targeted therapy in HER2-positive (HER2+) early breast cancer (EBC): Mass cytometry (CyTOF) analysis from the MARGOT trial
Conference poster   Peer reviewed

Peripheral immune profiling during neoadjuvant HER2-targeted therapy in HER2-positive (HER2+) early breast cancer (EBC): Mass cytometry (CyTOF) analysis from the MARGOT trial

E. Segui, H. Heiling, J. Baginska, A. N. Nau, I. Gomez Diaz, J. Huezo, P. R. Pohlmann, M. F. Rimawi, N. F. Sinclair, R. L. Yung, …
Clinical cancer research, v 32(4_Supplement), PS2-08-10
17 Feb 2026
DOI: https://doi.org/10.1158/1557-3265.SABCS25-PS2-08-10

Abstract

Background: Anti-tumor immune activity impacts outcomes in HER2+ EBC, and anti-HER2 antibodies act in part through immune mechanisms such as antibody-dependent cell-mediated cytotoxicity (ADCC). Margetuximab is an anti-HER2 antibody engineered to enhance ADCC through increased affinity for FcγRIIIa. Early changes in circulating immune populations during anti-HER2 therapy remain poorly characterized. We used CyTOF to profile peripheral blood mononuclear cells (PBMCs) at baseline and early on-treatment in patients receiving neoadjuvant chemotherapy plus HER2-directed therapy. Methods: The investigator-initiated MARGOT trial (NCT04425018/TBCRC052) randomized 171 patients with stage II-III HER2+ EBC 1:2 to receive neoadjuvant THP (trastuzumab/pertuzumab/paclitaxel) or TMP (margetuximab/pertuzumab/paclitaxel). The primary endpoint, pathologic complete response (pCR), was previously reported; pCR was numerically higher with TMP but not statistically significant. This PBMC sub-study included all patients with paired PBMCs at baseline and cycle 2 day 1 (C2D1; 3 weeks on-treatment) from sites meeting quality criteria before the data cut-off. A total of 91 patients (THP n = 25, TMP n = 66) underwent CyTOF profiling at both timepoints using 22 surface and 13 functional markers, identifying 26 immune cell populations. Descriptive statistics were used to summarize baseline and on-treatment cell type changes (Δ). Associations with pCR were evaluated using logistic regression; on-treatment changes by treatment arm were analyzed using linear regression. Multiple testing was addressed using false discovery rate (q-values); raw p-values <0.05 were considered of interest. We hypothesized that ADCC effector populations (e.g., NK cells, monocytes) would change more with margetuximab, and that evidence of ADCC would correlate with response. Results: For the 91 patients (56 HR+, 35 HR-), pCR rates were similar to the overall trial: 52% TMP, 48% THP. Baseline cell types associated with pCR overall and by treatment arm are shown in Table 1. After 3 weeks of therapy, classical monocytes showed the largest change across all cell populations in both treatment arms (mean Δ -5.1% (p<0.001, q<0.001) in TMP and -4.6% (p = 0.013, q = 0.110) in THP). A decrease in one activated NK cell subset (NK CD56dimCD57+CD8-) was observed only with TMP (TMP: mean Δ -1.83% vs THP: mean Δ -0.47%; mean difference TMP vs THP = -1.36; p = 0.002, q = 0.052). Functional marker analyses will be presented at the meeting.

Metrics

1 Record Views

Details

Logo image