nCounter NanoString Assay Shows Variable Concordance With Immunohistochemistry-based Algorithms in Classifying Cases of Diffuse Large B-Cell Lymphoma According to the Cell-of-Origin

Classifying diffuse large B-cell lymphoma (DLBCL) according to the cell-of-origin (COO) was first proposed using gene expression profiling; accordingly, DLBCL is classified into germinal-center B-cell type and activated B-cell type. Immunohistochemistry (IHC)-based classification using different algorithms is used widely due to the ability to use formalin-fixed paraffin-embedded tissue. Recently, newer techniques using RNA expression from formalin-fixed paraffin-embedded were introduced including the nCounter NanoString platform assay. In this brief report, we study the degree of concordance between the NanoString assay and 6 commonly utilized IHC-based algorithms to classify DLBCL cases by COO. Stains for CD10, BCL2, BCL6, FOXP-1, MUM-1, and LOM2 were used to classify a cohort of DLBCL by COO according to the respective IHC-algorithms. Then, RNA was extracted from the same cases for NanoString assay classification. The degree of concordance was calculated between the NanoString classification and each IHC-algorithm as well as among the different IHC-algorithm themselves. The concordance in COO classification of DLBCL between NanonoString assay and IHC-based algorithms is variable depending on the used IHC-algorithm; the highest concordance is seen with the Visco algorithm (kappa=0.69; P=0.001). Therefore, discrepancies between the recently introduced NanoString assay and the commonly utilized IHC-algorithms are expected to some extent and should be taken into consideration when interpreting conflicting results.