A retrospective analysis of metaplastic triple negative breast cancer response to sacituzumab govitecan and candidacy for targeted therapy.

Abstract

e13015 Background: Metaplastic breast cancer (MpBC) is a rare, histologically distinct form of breast cancer. Around 90% of MpBC is negative for estrogen receptor (ER), progesterone receptor (PR), and HER2. Treatment strategies for MpBC are extrapolated from triple-negative breast cancer (TNBC) data. MpBC, however, has low representation in guideline-establishing clinical trials. MpBC is associated with inferior chemotherapy response and worse prognosis compared to non-metaplastic (non-Mp) TNBC. There is scarce literature on MpBC response to antibody-drug conjugates (ADCs). We compare clinical outcomes for metastatic MpBC TNBC vs metastatic non-Mp TNBC treated with sacituzumab govitecan (SG), a TROP2-based ADC approved for metastatic TNBC, and assess for targetable genomic and pathological characteristics in MpBC. Methods: Patients (pts) with metastatic TNBC treated at the Mass General Hospital Cancer Center, between 2000 and 2025, were assessed for MpBC and receipt of SG. MpBC cases were interrogated for biomarkers qualifying for targeted therapy: HER2-low status (IHC 1+/2+), combined positive score (CPS), and targetable genomic variants based on next-generation sequencing of plasma and/or tissue ordered by the treating provider. Multivariable Cox regression analysis assessed the association of MpBC under SG treatment with progression-free survival (PFS) and overall survival (OS), adjusting for number of prior therapies and SG combined vs monotherapy. Results: Eighty-five pts with metastatic TNBC were treated with SG, 13 (15%) of whom had MpBC. Two MpBC and 23 non-Mp TNBC pts received SG with a PARP inhibitor (clinical trial), while the remaining pts received SG alone. Both cohorts received a median of 2 prior lines of therapy (range 0-8). Median age at diagnosis was 47 years in MpBC and 48 years in non-Mp TNBC. Median PFS in MpBC was 1.9 months vs 5.6 months for non-Mp TNBC (p = 0.3).Median OS in MpBC was 18.5 months vs 16.7 months in non-Mp TNBC (p = 0.9). Genomic and pathologic annotation of MpBC pts revealed that 8/13 cases were HER2-low (candidacy for trastuzumab deruxtecan), 2/13 cases had a CPS ≥ 10, 3/13 cases had a TMB ≥ 10 mut/Mb (indications for pembrolizumab), 5/13 cases had pathogenic PIK3CA variants, and 2/13 had pathogenic PTEN variants. There were no tumor-agnostic targetable genomic variants. Conclusions: Pts with MpBC treated with SG had a numerically worse PFS compared to non-Mp TNBC. While the difference is not statistically significant, the sample size of MpBC cases was small, and these findings merit evaluation in a larger cohort. Over half of MpBC cases were HER2-low and/or had pathogenic variants, which may broaden treatment options to other ADCs and/or trials. Optimizing MpBC management will require multi-institutional collaboration to assess larger numbers of pts with interrogation for biological rationale and response to targeted therapy.