Assessment of 2022 European LeukemiaNet risk classification system in real-world cohort from China.

BACKGROUND: The European LeukemiaNet (ELN) risk classification system for acute myeloid leukemia (AML) patients has been used worldwide. In 2022, the ELN risk classification system modified risk genes including CEBPA mutation status, myelodysplasia-related (MR) gene mutations and internal tandem duplications of FLT3 (FLT3-ITD). METHODS: We include newly diagnosed de novo AML patients at our center from January 2017 to December 2021, regardless of the further treatment received. Clinical data and date of survival were included. Survival analysis were performed using the Kaplan-Meier method, and the log-rank test was used to compare survival between different risk groups. RESULTS: We include 363 newly diagnosed de novo AML patients from 2017 to 2021 to assess the accuracy of the ELN risk classification system. Their survival results show that the ELN-2022 risk classification system is not superior to the ELN-2017 version; for patients with FLT3-ITD mutations but without FLT3 inhibitor treatment, their survival is similar to the ELN-2022 adverse risk group. The ELN-2022 risk classification system cannot accurately clarify ECOG performance status (PS) 2-4 patients, especially in the ELN-2022 favorable risk group. CONCLUSION: The ELN-2022 risk stratification system may not be appropriate for patients unable to receive intensive therapy or FLT3 inhibitor; more real-world data is needed to straify patients with worse ECOG PS and inferior intensive therapy.

Dexamethasone enhances venetoclax-induced apoptosis in acute myeloid leukemia cells.

Acute myeloid leukemia (AML) therapies have been significantly improved by the development of medicines that can target BCL-2. On the other hand, non-recurrent alterations in oncogenic pathways and gene expression patterns have already been linked to therapeutic resistance to venetoclax therapy. Bone marrow mesenchymal stromal cells (BM-MSCs) support leukemic cells in preventing chemotherapy-induced apoptosis by mitochondrial transfer in leukemic microenvironment. In this study, we investigated the enhancement of the antitumor effect of BCL-2 inhibitor venetoclax by dexamethasone. In particular, dexamethasone had no significant effect on the viability of AML cells, but dexamethasone combined with venetoclax could significantly increase the apoptosis of AML cells induced by venetoclax. When AML cells were co-cultured with BM-MSCs, dexamethasone combined with venetoclax showed additional anti-tumor effect compared to venetoclax alone. Venetoclax increased reactive oxygen species level in co-cultured AML cells, contributed to transfer more mitochondria from BM-MSCs to AML cells and protect AML cells from apoptosis. Dexamethasone combined with venetoclax induced more apoptosis, but dexamethasone reduced the venetoclax-induced reactive oxygen species level in AML cells and reduced the transfer of mitochondria from BM-MSCs to AML cells. This may lead to a diminished protective effect of BM-MSCs on AML cells. Together, our findings indicated that venetoclax in combination with dexamethasone could be a promising therapy in AML.

Visualizing Conformational Space of Functional Biomolecular Complexes by Deep Manifold Learning.

The cellular functions are executed by biological macromolecular complexes in nonequilibrium dynamic processes, which exhibit a vast diversity of conformational states. Solving the conformational continuum of important biomolecular complexes at the atomic level is essential to understanding their functional mechanisms and guiding structure-based drug discovery. Here, we introduce a deep manifold learning framework, named AlphaCryo4D, which enables atomic-level cryogenic electron microscopy (cryo-EM) reconstructions that approximately visualize the conformational space of biomolecular complexes of interest. AlphaCryo4D integrates 3D deep residual learning with manifold embedding of pseudo-energy landscapes, which simultaneously improves 3D classification accuracy and reconstruction resolution via an energy-based particle-voting algorithm. In blind assessments using simulated heterogeneous datasets, AlphaCryo4D achieved 3D classification accuracy three times those of alternative methods and reconstructed continuous conformational changes of a 130-kDa protein at sub-3 Å resolution. By applying this approach to analyze several experimental datasets of the proteasome, ribosome and spliceosome, we demonstrate its potential generality in exploring hidden conformational space or transient states of macromolecular complexes that remain hitherto invisible. Integration of this approach with time-resolved cryo-EM further allows visualization of conformational continuum in a nonequilibrium regime at the atomic level, thus potentially enabling therapeutic discovery against highly dynamic biomolecular targets.