RESUMO
T-prolymphocytic leukemia (T-PLL) is a mature T-cell neoplasm associated with marked chemotherapy resistance and continued poor clinical outcomes. Current treatments, i.e. the CD52-antibody alemtuzumab, offer transient responses, with relapses being almost inevitable without consolidating allogeneic transplantation. Recent more detailed concepts of T-PLL's pathobiology fostered the identification of actionable vulnerabilities: (i) altered epigenetics, (ii) defective DNA damage responses, (iii) aberrant cell-cycle regulation, and (iv) deregulated pro-survival pathways, including TCR and JAK/STAT signaling. To further develop related pre-clinical therapeutic concepts, we studied inhibitors of (H)DACs, BCL2, CDK, MDM2, and clas-sical cytostatics, utilizing (a) single-agent and combinatorial compound testing in 20 well-characterized and molecularly-profiled primary T-PLL (validated by additional 42 cases), and (b) 2 independent murine models (syngeneic transplants and patient-derived xenografts). Overall, the most efficient/selective single-agents and combinations (in vitro and in mice) in-cluded Cladribine, Romidepsin ((H)DAC), Venetoclax (BCL2), and/or Idasanutlin (MDM2). Cladribine sensitivity correlated with expression of its target RRM2. T-PLL cells revealed low overall apoptotic priming with heterogeneous dependencies on BCL2 proteins. In additional 38 T-cell leukemia/lymphoma lines, TP53 mutations were associated with resistance towards MDM2 inhibitors. P53 of T-PLL cells, predominantly in wild-type configuration, was amenable to MDM2 inhibition, which increased its MDM2-unbound fraction. This facilitated P53 activa-tion and down-stream signals (including enhanced accessibility of target-gene chromatin re-gions), in particular synergy with insults by Cladribine. Our data emphasize the therapeutic potential of pharmacologic strategies to reinstate P53-mediated apoptotic responses. The identified efficacies and their synergies provide an informative background on compound and patient selection for trial designs in T-PLL.
RESUMO
High-throughput sequencing plays a pivotal role in hematological malignancy diagnostics, but interpreting missense mutations remains challenging. In this study, we used the newly available AlphaMissense database to assess the efficacy of machine learning to predict missense mutation effects and its impact to improve our ability to interpret them. Based on the analysis of 2073 variants from 686 patients analyzed for clinical purpose, we confirmed the very high accuracy of AlphaMissense predictions in a large real-life data set of missense mutations (AUC of ROC curve 0.95), and provided a comprehensive analysis of the discrepancies between AlphaMissense predictions and state of the art clinical interpretation.