The genetics of Richter syndrome reveals disease heterogeneity and predicts survival post-transformation.

Richter syndrome (RS) represents the development of diffuse large B-cell lymphoma (DLBCL) in the context of chronic lymphocytic leukemia (CLL). The scarcity of biologic information about RS has hampered the identification of molecular predictors of RS outcome. We addressed this issue by performing a comprehensive molecular characterization of 86 pathologically-proven RS. TP53 disruption (47.1%) and c-MYC abnormalities (26.2%) were the most frequent alterations, whereas common genetic lesions of de novo DLBCL were rare or absent. By multivariate analysis, lack of TP53 disruption (HR:0.43; p=.003) translated into significant survival advantage with 57% reduction in risk of death. An algorithm based on TP53 disruption, response to RS treatment and ECOG performance status had 80.9% probability of correctly discriminating RS survival (c-index=.809). RS that were clonally unrelated to the paired CLL phase were clinically and biologically different from clonally related RS because of significantly longer survival (median: 62.5 months vs 14.2 months; p=.017), and lower prevalence of TP53 disruption (23.1% vs 60.0%; p=.018) and B-cell receptor stereotypy (7.6% vs 50.0%; p=.009). The molecular dissection of RS into biologically distinct categories highlights the genetic heterogeneity of this disorder and provides clinically relevant information for refining the prognostic stratification of patients.