Abstract
Blastoid B-cell neoplasms mainly include B-lymphoblastic leukemia/lymphoma (B-ALL), blastoid mantle cell lymphoma, and high-grade B-cell lymphoma with blastoid morphologic features (blastoid HGBL). Distinguishing blastoid HGBL from B-ALL can be challenging and we previously developed six-point flow cytometry-focused and three-point immunohistochemistry-focused scoring systems to aid in differential diagnosis. However, the six-point scoring system was derived from bone marrow cases and occasional cases may have a misleading score using either system. In this study, we assessed 121 cases of blastoid-HGBL (37 BM and 84 extramedullary) to validate the six-point scoring system in all tissue types and to further compare the two scoring systems. Compared with 47 B-ALL cases enriched for CD34-negative neoplasm, the 121 blastoid-HGBL cases showed distinctive pathologic features. The six-point scoring system showed a sensitivity of 100%. A comparison of the two scoring systems in blastoid HGBL (n = 64) and B-ALL (n = 37) showed a concordance score rate of 88%. Thirteen cases showed misleading scores, including five HGBL and eight B-ALL, and the diagnosis was further validated by gene transcriptome profiling. Twelve of thirteen cases had discordant scores between the two scoring systems. Simultaneous employment of both scoring systems improved the accuracy of classification of blastoid B-cell neoplasms to 99%. In conclusion, the previously defined six-point scoring system showed an excellent performance regardless of the tissue origin. Using both scoring systems together improves the accuracy of classification of blastoid B-cell neoplasms. Cases with discordant scores between the two scoring systems were extremely challenging neoplasms and classification required correlation with all available clinical and genetic features.