Concordance of HER2 status between core needle biopsy and surgical resection specimens of breast cancer: an analysis focusing on the HER2-low status.

BACKGROUND: Human epidermal growth factor receptor 2 (HER2)-low status has recently gained attention because of the potential therapeutic benefits of antibody-drug conjugates (ADCs) in breast cancer patients. We aimed to investigate the concordance of HER2 status between core needle biopsy (CNB) and subsequent surgical resection specimens focusing on the HER2-low status. METHODS: This retrospective study was conducted in 1,387 patients with invasive breast cancer whose HER2 status was evaluated in both CNB and surgical resection specimens. The discordance rates between CNB and surgical resection specimens and the clinicopathological features associated with HER2 status discordance were analyzed. RESULTS: The overall concordance rates of HER2 status between CNB and surgical resection specimens were 99.0% (κ = 0.925) for two-group classification (negative vs. positive) and 78.5% (κ = 0.587) for three-group classification (zero vs. low vs. positive). The largest discordance occurred in CNB-HER2-zero cases with 42.8% of them reclassified as HER2-low in surgical resection. HER2 discordance was associated with lower histologic grade, tumor multiplicity, and luminal A subtype. In multivariate analysis, tumor multiplicity and estrogen receptor (ER) positivity were independent predictive factors for HER2-zero to low conversion. CONCLUSIONS: Incorporation of HER2-low category in HER2 status interpretation reduces the concordance rate between CNB and surgical resection specimens. Tumor multiplicity and ER positivity are predictive factors for conversion from HER2-zero to HER2-low status. Therefore, HER2 status should be re-evaluated in resection specimens when considering ADCs in tumors exhibiting multiplicity and ER positivity.

Prognostic Impact of the Immunoscore Based on Whole-Slide Image Analysis of CD3+ Tumor-Infiltrating Lymphocytes in Diffuse Large B-Cell Lymphoma.

An Immunoscore based on tumor-infiltrating T-cell density was validated as a prognostic factor in patients with solid tumors. However, the potential utility of the Immunoscore in predicting the prognosis of patients with diffuse large B-cell lymphoma (DLBCL) is unclear. Here, the prognostic value of an Immunoscore based on tumor-infiltrating CD3+ T-cell density was evaluated in 104 patients with DLBCL who underwent R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisolone) therapy. Digitally scanned whole-slide images were analyzed using Aperio ImageScope software. CD3+ cell densities in the whole tumor area were quantitated using 3 different methods, including number of CD3+ cells/area (mm2), ratio of CD3+ cells to total cells, and ratio of CD3+ cells to CD20+ cells. There was a high concordance among the 3 methods. Patients with low CD3+ cell density had an elevated serum lactate dehydrogenase level and a high Ki-67 proliferation index (all, P < .05). Patients with low CD3+ cell density, according to all 3 methods, had worse overall survival (OS) and worse progression-free survival (P < .05, all). They also had poor OS, independent of MYC/BCL2 double expression (DE) status, Eastern Cooperative Oncology Group performance status, or Ann Arbor stage (all, P < .05). These results were validated using 2 publicly available data sets. In both validation cohorts, patients with low CD3E mRNA expression had an elevated serum lactate dehydrogenase level, extranodal site involvement, and DE status (P < .05). They also had worse progression-free survival (P = .067 and P = .002, respectively) and OS (both P < .05). A low CD3E mRNA level was predictive of poor OS, independent of DE status. An Immunoscore based on whole-slide image analysis of CD3+ T-cell infiltration was sufficient to predict survival in patients with DLBCL. Low CD3+ cell density was a poor prognostic factor, independent of other prognostic parameters and DE status.