Novel ERBB2 Variant Potentially Associated with Resistance Against Anti-HER2 Monoclonal Antibody-Based Therapy in ERBB2-Amplified Metastatic Colorectal Cancer.

PURPOSE:HER2-targeted therapies in ERBB2-amplified metastatic colorectal cancer (mCRC) are effective; however, a notable portion of patients do not respond to treatment, and secondary resistance occurs in most patients receiving these treatments. The purpose of this study was to investigate determinants of treatment efficacy and resistance in patients with ERBB2-amplified mCRC who received HER2-targeted therapy by analyzing multiomics data. EXPERIMENTAL DESIGN:We investigated genomic data from a nationwide large cancer genomic screening project, the SCRUM-Japan project. We analyzed paired genome and transcriptome data of tissue and genomic data of ctDNA collected pre- and postprogression in patients enrolled in the related trial, TRIUMPH, in ERBB2-amplified mCRC. RESULTS:In 155 patients with ERBB2-amplified solid tumors who received HER2-targeted therapy based on the SCRUM-Japan project, the objective response rate was 50%, 51%, and 35% in ERBB2 wild-type, variant of unknown significance, and pathogenic variant groups, respectively. In the paired genome and transcriptome data analyses in TRIUMPH, we identified the novel splicing-associated variant c.644-66_-2del in one of the 11 patients with paired whole-exome sequencing and whole-transcriptome sequencing data sets, which lacks the binding domain of pertuzumab, in progressed metastatic tumor as a variant with potential pathogenicity. The time-course ctDNA analysis detected c.644-66_-2del as an acquired variant. CONCLUSIONS:This study highlighted the importance of ERBB2 genomic status when evaluating the efficacy of HER2-targeted therapies in ERBB2-amplified mCRC. The identification of a novel splicing-associated variant may provide insights into potential mechanisms of treatment resistance. Furthermore, we demonstrated the utility of ctDNA to follow the acquired genomic status of mCRC tumors.