Integrated proteogenomics reveals molecular predictors of recurrence beyond HER2 IHC in HER2-positive breast cancer

Purpose: Breast cancer remains the most common malignancy among women, with about 20% driven by HER2 overexpression. Although HER2-targeted therapies have improved outcomes, acquired resistance and recurrence persist. Clinical HER2 Immunohistochemistry(IHC) often fails to reflect actual HER2 abundance, as seen in both IHC-negative/high-protein and IHC 3+/low-protein cases. Thus we conducted an integrated proteogenomic analysis to define HER2 heterogeneity and identify recurrence-associated signatures. Methods: Total of 62 patients with a median age of 41.5 years (range, 25-60) were included from National Cancer Center in Korea, and breast tumor tissues analyzed using an integrated proteogenomic workflow combining whole-exome sequencing, whole-transcriptome, global proteomics, phosphoproteomics, and PRM-based HER2 quantification. Differential genomic and proteomic features between the 46 non-recurrence and 15 recurrence patients were compared and validated in TCGA and CPTAC cohorts. Results: Description between clinical IHC and quantitative HER2 measurement were observed in 10% of low HER2 and in 25% of high HER2 cases. Recurrent patients revealed frequent TP53 mutations, MYC amplification, and ERBB2 hotspot variants (E192* and D844*) in genomic profiling. DEGs analyses integrating RNA and proteomic expression identified a shared set of recurrence-associated genes, including ATXN7L3BM GPRC5A and GPRC5A and WNT5A, which collectively reflected pathways related to extracellular matrix remodeling, metastatic, signaling, metabloic reprogramming, and transcriptional regulation. METABRIC comparison showed similar patterns of ERBB2/MYC amplification, TP53 loss, and genomic instability observed in recurrent tumors. Mutational signature analysis indicated dominant contributions from aging-related, APOBEC-driven, and homologous recombination-defective processes. Copy-number signature profiling further identified five recurrent genomic subtypes (CNV48 A-E), including chromothripsis-like and loss of heterozygosity patterns shared with TCGA and METABRIC HER2-positive tumors. Conclusions: These findings indicate that recurrence risk in HER2-positive breast cancer is determined more by initial tumor stage than by HER2 IHC status. This underscores the clinical value of quantitative proteogenomic assessment as a complementary tool to improve risk stratification beyond conventional HER2 testing. This work was supported by grant from the National Cancer Center. (Grant No. 2510692-1)