The DestinyBreast-(DB)-04 and DB06-trials have shown clinical activity of Trastuzumab-Deruxtecan (T-DXd) in HER2-low and HER2-ultralow metastatic breast cancer (BC). The identification of HER2-low and -ultralow BC is therefore essential for personalized therapy with T-DXd. We evaluated 723 residual tumors from the PENELOPE-B trial (NCT01864746) and correlated different levels of HER2 protein expression with prognosis and mRNA profiles, including HER2 transcripts. In PENELOPE-B, 57.68% (n=417) of 723 residual tumors were HER2-low. The HER2-ultralow category was assigned to 109 tumors (15.08%), and 197 tumors (27.25%) were completely HER2 negative (HER2-zero). In Kaplan-Meier analysis, there were no survival differences among these three subgroups. There was no significant difference in HER2 mRNA expression between HER2-zero and HER2-ultralow tumors (p=0.08). In contrast, there was a highly significant difference in HER2 mRNA expression between HER2-ultralow and HER2-low tumors (p<0.0001) and between HER2-low and HER2-positive tumors (p<0.0001). The extracellular protease Cathepsin-L, which has been suggested as a biomarker for extracellular cleavage of T-DXd, was detectable in all HER2-related subgroups, and was a negative prognostic factor for iDFS and OS (p=0.0001) in pre-neoadjuvant core biopsies. In our study, we were able to characterize HER2-low as a clinically relevant and molecular defined tumor group with significantly increased HER2-expression. In contrast, for HER2-ultralow we did not observe a defined molecular phenotype, despite the clinically relevant regulatory approval of T-DXd also in the ultra-low subgroup. Additional investigations are needed to identify biomarkers beyond HER2 for T-DXd response as a basis for refined criteria for treatment eligibility.
Keywords
