Geodesic shape regression based deep learning segmentation for assessing longitudinal hippocampal atrophy in dementia progression.

Longitudinal hippocampal atrophy is commonly used as progressive marker assisting clinical diagnose of dementia. However, precise quantification of the atrophy is limited by longitudinal segmentation errors resulting from MRI artifacts across multiple independent scans. To accurately segment the hippocampal morphology from longitudinal 3T T1-weighted MR images, we propose a diffeomorphic geodesic guided deep learning method called the GeoLongSeg to mitigate the longitudinal variabilities that unrelated to diseases by enhancing intra-individual morphological consistency. Specifically, we integrate geodesic shape regression, an evolutional model that estimates smooth deformation process of anatomical shapes, into a two-stage segmentation network. We adopt a 3D U-Net in the first-stage network with an enhanced attention mechanism for independent segmentation. Then, a hippocampal shape evolutional trajectory is estimated by geodesic shape regression and fed into the second network to refine the independent segmentation. We verify that GeoLongSeg outperforms other four state-of-the-art segmentation pipelines in longitudinal morphological consistency evaluated by test-retest reliability, variance ratio and atrophy trajectories. When assessing hippocampal atrophy in longitudinal data from the Alzheimer's Disease Neuroimaging Initiative (ADNI), results based on GeoLongSeg exhibit spatial and temporal local atrophy in bilateral hippocampi of dementia patients. These features derived from GeoLongSeg segmentation exhibit the greatest discriminatory capability compared to the outcomes of other methods in distinguishing between patients and normal controls. Overall, GeoLongSeg provides an accurate and efficient segmentation network for extracting hippocampal morphology from longitudinal MR images, which assist precise atrophy measurement of the hippocampus in early stage of dementia.

A Chebyshev Confidence Guided Source-Free Domain Adaptation Framework for Medical Image Segmentation.

Source-free domain adaptation (SFDA) aims to adapt models trained on a labeled source domain to an unlabeled target domain without access to source data. In medical imaging scenarios, the practical significance of SFDA methods has been emphasized due to data heterogeneity and privacy concerns. Recent state-of-the-art SFDA methods primarily rely on self-training based on pseudo-labels (PLs). Unfortunately, the accuracy of PLs may deteriorate due to domain shift, thus limiting the effectiveness of the adaptation process. To address this issue, we propose a Chebyshev confidence guided SFDA framework to accurately assess the reliability of PLs and generate self-improving PLs for self-training. The Chebyshev confidence is estimated by calculating the probability lower bound of PL confidence, given the prediction and the corresponding uncertainty. Leveraging the Chebyshev confidence, we introduce two confidence-guided denoising methods: direct denoising and prototypical denoising. Additionally, we propose a novel teacher-student joint training scheme (TJTS) that incorporates a confidence weighting module to iteratively improve PLs' accuracy. The TJTS, in collaboration with the denoising methods, effectively prevents the propagation of noise and enhances the accuracy of PLs. Extensive experiments in diverse domain scenarios validate the effectiveness of our proposed framework and establish its superiority over state-of-the-art SFDA methods. Our paper contributes to the field of SFDA by providing a novel approach for precisely estimating the reliability of PLs and a framework for obtaining high-quality PLs, resulting in improved adaptation performance.

Dynamic evolution of bone marrow adipocyte in B cell acute lymphoblastic leukemia: insights from diagnosis to post-chemotherapy.

Adipocyte is a unique and versatile component of bone marrow microenvironment (BMM). However, the dynamic evolution of Bone Marrow (BM) adipocytes from the diagnosis of B cell Acute Lymphoblastic Leukemia (B-ALL) to the post-treatment state, and how they affect the progression of leukemia, remains inadequately explicated. Primary patient-derived xenograft models (PDXs) and stromal cell co-culture system are employed in this study. We show that the dynamic evolution of BM adipocytes from initial diagnosis of B-ALL to the post-chemotherapy phase, transitioning from cellular depletion in the initial leukemia niche to a fully restored state upon remission. Increased BM adipocytes retards engraftment of B-ALL cells in PDX models and inhibits cells growth of B-ALL in vitro. Mechanistically, the proliferation arrest of B-ALL cells in the context of adipocytes-enrichment niche, might attribute to the presence of adiponectin secreted by adipocytes themselves and the absence of cytokines secreted by mesenchymal stem cell (MSCs). In summary, our findings offer a novel perspective for further in-depth understanding of the dynamic balance between BMM and B-ALL.

Mapping Multi-Modal Brain Connectome for Brain Disorder Diagnosis via Cross-Modal Mutual Learning.

Recently, the study of multi-modal brain connectome has recorded a tremendous increase and facilitated the diagnosis of brain disorders. In this paradigm, functional and structural networks, e.g., functional and structural connectivity derived from fMRI and DTI, are in some manner interacted but are not necessarily linearly related. Accordingly, there remains a great challenge to leverage complementary information for brain connectome analysis. Recently, Graph Convolutional Networks (GNN) have been widely applied to the fusion of multi-modal brain connectome. However, most existing GNN methods fail to couple inter-modal relationships. In this regard, we propose a Cross-modal Graph Neural Network (Cross-GNN) that captures inter-modal dependencies through dynamic graph learning and mutual learning. Specifically, the inter-modal representations are attentively coupled into a compositional space for reasoning inter-modal dependencies. Additionally, we investigate mutual learning in explicit and implicit ways: (1) Cross-modal representations are obtained by cross-embedding explicitly based on the inter-modal correspondence matrix. (2) We propose a cross-modal distillation method to implicitly regularize latent representations with cross-modal semantic contexts. We carry out statistical analysis on the attentively learned correspondence matrices to evaluate inter-modal relationships for associating disease biomarkers. Our extensive experiments on three datasets demonstrate the superiority of our proposed method for disease diagnosis with promising prediction performance and multi-modal connectome biomarker location.

Changing epidemiology, microbiology and mortality of bloodstream infections in patients with haematological malignancies before and during SARS-CoV-2 pandemic: a retrospective cohort study.

OBJECTIVE: This study was to explore the changes in bacterial bloodstream infection (BSI) in patients with haematological malignancies (HMs) before and during SARS-CoV-2 pandemic. DESIGN: Retrospective cohort study between 2018 and 2021. SETTING: The largest haematological centre in southern China. RESULTS: A total of 599 episodes of BSI occurring in 22 717 inpatients from January 2018 to December 2021 were analysed. The frequencies of the total, Gram-negative and Gram-positive BSI before and during the pandemic were 2.90% versus 2.35% (p=0.011), 2.49% versus 1.77% (p<0.001) and 0.27% versus 0.44% (p=0.027), respectively. The main isolates from Gram-negative or Gram-positive BSI and susceptibility profiles also changed. The 30-day mortality caused by BSI was lower during the pandemic (21.1% vs 14.3%, p=0.043). Multivariate analysis revealed that disease status, pulmonary infection and shock were independent predictors of 30-day mortality. CONCLUSION: Our data showed that the incidence of total and Gram-negative organisms BSI decreased, but Gram-positive BSI incidence increased in patients with HMs during the pandemic along with the changes of main isolates and susceptibility profiles. Although the 30-day mortality due to BSI was lower during the pandemic, the new infection prevention strategy should be considered for any future pandemics.

CReg-KD: Model refinement via confidence regularized knowledge distillation for brain imaging.

One of the core challenges of deep learning in medical image analysis is data insufficiency, especially for 3D brain imaging, which may lead to model over-fitting and poor generalization. Regularization strategies such as knowledge distillation are powerful tools to mitigate the issue by penalizing predictive distributions and introducing additional knowledge to reinforce the training process. In this paper, we revisit knowledge distillation as a regularization paradigm by penalizing attentive output distributions and intermediate representations. In particular, we propose a Confidence Regularized Knowledge Distillation (CReg-KD) framework, which adaptively transfers knowledge for distillation in light of knowledge confidence. Two strategies are advocated to regularize the global and local dependencies between teacher and student knowledge. In detail, a gated distillation mechanism is proposed to soften the transferred knowledge globally by utilizing the teacher loss as a confidence score. Moreover, the intermediate representations are attentively and locally refined with key semantic context to mimic meaningful features. To demonstrate the superiority of our proposed framework, we evaluated the framework on two brain imaging analysis tasks (i.e. Alzheimer's Disease classification and brain age estimation based on T1-weighted MRI) on the Alzheimer's Disease Neuroimaging Initiative dataset including 902 subjects and a cohort of 3655 subjects from 4 public datasets. Extensive experimental results show that CReg-KD achieves consistent improvements over the baseline teacher model and outperforms other state-of-the-art knowledge distillation approaches, manifesting that CReg-KD as a powerful medical image analysis tool in terms of both promising prediction performance and generalizability.

Prediction of microseismic events in rock burst mines based on MEA-BP neural network.

Microseismic monitoring is an important tool for predicting and preventing rock burst incidents in mines, as it provides precursor information on rock burst. To improve the prediction accuracy of microseismic events in rock burst mines, the working face of the Hegang Junde coal mine is selected as the research object, and the research data will consist of the microseismic monitoring data from this working face over the past 4 years, adopts expert system and temporal energy data mining method to fuse and analyze the mine pressure manifestation regularity and microseismic data, and the "noise reduction" data model is established. By comparing the MEA-BP and traditional BP neural network models, the results of the study show that the prediction accuracy of the MEA-BP neural network model was higher than that of the BP neural network. The absolute and relative errors of the MEA-BP neural network were reduced by 247.24 J and 46.6%, respectively. Combined with the online monitoring data of the KJ550 rock burst, the MEA-BP neural network proved to be more effective in microseismic energy prediction and improved the accuracy of microseismic event prediction in rock burst mines.

Green synthesis of mesoporous and biodegradable iron silicide nanoparticles for photothermal cancer therapy.

Photothermal nanomaterials have shown great potential for photothermal therapy. In this study, we developed a simple green method of magnesiothermic co-reduction for the synthesis of mesoporous, magnetic and biodegradable iron silicide nanoparticles (FeSi NPs) as applied to photothermal therapy (PTT). Starting from biogenic tabasheer extracted from bamboo and Fe2O3, the resultant FeSi NPs with a much lower band gap exhibited excellent optical absorption with a photothermal conversion efficiency of 76.2%, indicating a good photothermal performance. The weight extinction coefficient was measured to be 13.3 L g-1 cm-1 at 1064 nm (second near-infrared window, NIR-II), which surpassed the performance of other competitive Si-based and Fe-based photothermal agents. Results of the cell viability assay showed that cells could be killed by NIR-II laser irradiation with the synthesized FeSi NPs. In vivo results on mice showed clearly an efficient suppression of tumour growth by photothermal treatment with FeSi NPs. FeSi NPs were found to be biodegradable in simulated body fluids. The results from our work indicate that FeSi NPs are a new class of promising photothermal agents (PTAs) for application in cancer therapy.

Ensemble learning via supervision augmentation for white matter hyperintensity segmentation.

Since the ambiguous boundary of the lesion and inter-observer variability, white matter hyperintensity segmentation annotations are inherently noisy and uncertain. On the other hand, the high capacity of deep neural networks (DNN) enables them to overfit labels with noise and uncertainty, which may lead to biased models with weak generalization ability. This challenge has been addressed by leveraging multiple annotations per image. However, multiple annotations are often not available in a real-world scenario. To mitigate the issue, this paper proposes a supervision augmentation method (SA) and combines it with ensemble learning (SA-EN) to improve the generalization ability of the model. SA can obtain diverse supervision information by estimating the uncertainty of annotation in a real-world scenario that per image have only one ambiguous annotation. Then different base learners in EN are trained with diverse supervision information. The experimental results on two white matter hyperintensity segmentation datasets demonstrate that SA-EN gets the optimal accuracy compared with other state-of-the-art ensemble methods. SA-EN is more effective on small datasets, which is more suitable for medical image segmentation with few annotations. A quantitative study is presented to show the effect of ensemble size and the effectiveness of the ensemble model. Furthermore, SA-EN can capture two types of uncertainty, aleatoric uncertainty modeled in SA and epistemic uncertainty modeled in EN.

MRF-Net: A multi-branch residual fusion network for fast and accurate whole-brain MRI segmentation.

Whole-brain segmentation from T1-weighted magnetic resonance imaging (MRI) is an essential prerequisite for brain structural analysis, e.g., locating morphometric changes for brain aging analysis. Traditional neuroimaging analysis pipelines are implemented based on registration methods, which involve time-consuming optimization steps. Recent related deep learning methods speed up the segmentation pipeline but are limited to distinguishing fuzzy boundaries, especially encountering the multi-grained whole-brain segmentation task, where there exists high variability in size and shape among various anatomical regions. In this article, we propose a deep learning-based network, termed Multi-branch Residual Fusion Network, for the whole brain segmentation, which is capable of segmenting the whole brain into 136 parcels in seconds, outperforming the existing state-of-the-art networks. To tackle the multi-grained regions, the multi-branch cross-attention module (MCAM) is proposed to relate and aggregate the dependencies among multi-grained contextual information. Moreover, we propose a residual error fusion module (REFM) to improve the network's representations fuzzy boundaries. Evaluations of two datasets demonstrate the reliability and generalization ability of our method for the whole brain segmentation, indicating that our method represents a rapid and efficient segmentation tool for neuroimage analysis.

CD43 is an adverse prognostic factor in newly diagnosed multiple myeloma.

Currently, the expression pattern and prognostic value of CD43 expression in multiple myeloma (MM) remain unknown. 109 newly diagnosed MM patients were recruited and CD43 expression was determined by multiparameter flow cytometry, of which 77 (70.6%) were CD43 positive. Patients with positive CD43 expression were more likely to present with, hemoglobin < 85 g/L (p = 0.008), International Staging System (ISS) stage III (p = 0.044), 13q14 deletion (p = 0.034) and more monoclonal plasma cells (p = 0.003). Patients with CD43 positive had significantly poor treatment response (p = 0.021), progression-free survival (PFS) (p = 0.012), and overall survival (OS) (p = 0.023) than those without CD43. The poorer prognosis of CD43-positive patients was retained in multivariate analysis (p = 0.005 for PFS; p = 0.013 for OS). Our study indicated that CD43 was an independent adverse prognostic factor in multiple myeloma.

An anatomical knowledge-based MRI deep learning pipeline for white matter hyperintensity quantification associated with cognitive impairment.

Recent studies have confirmed that white matter hyperintensities (WMHs) accumulated in strategic brain regions can predict cognitive impairments associated with Alzheimer's disease (AD). The knowledge of white matter anatomy facilitates lesion-symptom mapping associated with cognition, and provides important spatial information for lesion segmentation algorithms. However, deep learning-based methods in the white matter hyperintensity (WMH) segmentation realm do not take full advantage of anatomical knowledge in decision-making and lesion localization processes. In this paper, we proposed an anatomical knowledge-based MRI deep learning pipeline (AU-Net), handcrafted anatomical-based spatial features developed from brain atlas were integrated with a well-designed U-Net configuration to simultaneously segment and locate WMHs. Manually annotated data from WMH segmentation challenge were used for the evaluation. We then applied this pipeline to investigate the association between WMH burden and cognition within another publicly available database. The results showed that AU-Net significantly improved segmentation performance compared with methods that did not incorporate anatomical knowledge (p < 0.05), and achieved a 14-17% increase based on area under the curve (AUC) in the cohort with mild WMH burden. Different configurations for incorporating anatomical knowledge were evaluated, the proposed stage-wise AU-Net-two-step method achieved the best performance (Dice: 0.86, modified Hausdorff distance: 3.06 mm), which was comparable with the state-of-the-art method (Dice: 0.87, modified Hausdorff distance: 3.62 mm). We observed different WMH accumulation patterns associated with normal aging and cognitive impairments. Furthermore, the characteristics of individual WMH lesions located in strategic regions (frontal and parietal subcortical white matter, as well as corpus callosum) were significantly correlated with cognition after adjusting for total lesion volumes. Our findings suggest that AU-Net is a reliable method to segment and quantify brain WMHs in elderly cohorts, and is valuable in individual cognition evaluation.

U-net combined with CRF and anatomical based spatial features to segment white matter hyperintensities.

White matter hyperintensities (WMH) are important biomarkers for cerebral small vessel disease and closely associated with other neurodegenerative process. In this paper, we proposed a fully automatic WMH segmentation method based on U-net architecture. CRF were combined with U-net to refine segmentation results. We used a new anatomical based spatial feature produced by brain tissue segmentation based on T1 image, along with intensities of T1 and T2-FLAIR images to train our neural network. We compared 8 forms of automated WMH segmentation methods, range from traditional statistical learnng methods to deep learning based methods, with different architecture and used different features. Results showed our proposed method achieved best performance in terms of most metrics, and the inclusion of anatomical based spatial features strongly increase the segmentation performance.

Automatic Brain Tumor Segmentation Method Based on Modified Convolutional Neural Network.

In the domain of brain diseases, it is difficult for image registration after some brain structures are severely deformed because of diseases. Fortunately, convolutional neural network have gained many promising results in semantic segmentation challenging tasks in recent years. To enhance the performance of automatic brain tumor segmentation, this paper presents a robust segmentation algorithm based on convolutional neural network, which achieved improvement of 3.84% in segmenting the enhancing tumor. Our network architecture is developed from the prevalent U-Net. We combined it with ResNet and modified it to maximize its performance in our brain tumor segmentation task. In this work, BraTS 2017 dataset was employed to train and test the proposed network. Data imbalance was dealt with using a weighted cross entropy loss function. The problem of overfitting was handled through data augmentation. The proposed method achieved averaged dice scores of 0.883, 0.781 and 0.748 for whole tumor, tumor core and enhancing tumor respectively in the validation set and 0.877, 0.774, 0.757 respectively in the testing set.

Apatinib exerts anti-tumor activity to non-Hodgkin lymphoma by inhibition of the Ras pathway.

Apatinib is a tyrosine kinase inhibitor that selectively targets vascular endothelial growth factor receptor-2 (VEGFR-2). Although apatinib has shown promising anti-tumor activity against several types of tumor, its role and underlying mechanism against non-Hodgkin lymphoma (NHL) remain to be explored. Here, we report that apatinib dramatically inhibited in vitro the proliferation of various human NHL cell lines, including Burkitt lymphoma (BL), mantle cell lymphoma (MCL), and diffuse large B-cell lymphoma (DLBCL), in a dose-dependent manner. Moreover, administration of apatinib markedly delayed tumor growth in vivo in a xenograft mouse model derived from human DLBCL OCI-ly3 cells, in association with significantly prolonged survival of tumor-bearing mice. Mechanistically, apatinib suppressed activation of VEGFR2 (manifested by reduced VEGFR2 phosphorylation), accompanied by inhibition of the Ras pathway (reflected by down-regulation Ras, Raf, pMEK1/2, pERK1/2) in OCI-ly1 (GCB subtype of DLBCL) and SU-DHL2 (ABC subtype of DLBCL) cells. Of note, apatinib sharply impaired angiogenesis in vivo in tumor tissues. Together, these results indicate that apatinib displays a marked cytotoxic activity against various types of NHL cells (including BL, MCL, and GCB- or ABC-DLBCL) both in vitro and in vivo. They also suggest that anti-NHL activity of apatinib might be associated with inhibition of tumor cell growth and induction of apoptosis as well as anti-angiogenesis by targeting VEGFR2 and its downstream Ras/Raf/MEK/ERK pathway.

A retrospective comparison of Escherichia coli and polyethylene glycol-conjugated asparaginase for the treatment of adolescents and adults with newly diagnosed acute lymphoblastic leukemia.

Data from clinical trials suggest that polyethylene glycol-conjugated asparaginase (PEG asparaginase) should be recommended as a replacement for Escherichia coli (E. coli) asparaginase in the treatment of pediatric acute lymphoblastic leukemia (ALL) due to its prolonged effect, similar safety profile and convenience. The present study investigated the efficacy and safety of PEG asparaginase in adolescents and adults with newly diagnosed ALL. The clinical data of 122 patients, ≥14 years old with de novo ALL, who received either PEG asparaginase or E. coli asparaginase as part of an induction regimen, were retrospectively analyzed. The results revealed that PEG asparaginase had a comparable complete remission rate (95.65 vs. 90.79%), median overall survival time (14.07 vs. 16.29 months) and median relapse-free survival time (10.00 vs. 8.57 months) with E. coli asparaginase. In addition, patients <35 years old receiving PEG asparaginase obtained a higher median RFS time compared with those receiving E. coli asparaginase (10.93 vs. 8.97 months; P=0.037). Patients treated with E. coli asparaginase exhibited a significantly higher incidence of central nervous system leukemia (CNSL) compared with those treated with PEG asparaginase (27.63 vs. 10.87%; P=0.028) during the consolidation phase. Toxic events, including allergy, grade III-IV liver dysfunction, renal function damage and pancreatic lesions were similar between the two groups. A longer duration of coagulation dysfunction (9.80±5.51 vs. 6.80±4.21 days; P=0.002) and agranulocytosis (18.89±8.79 vs. 12.03±8.34 days; P<0.01), and a higher incidence of grade IV-V infections (22.73 vs. 7.25%; P=0.018) were observed in the PEG asparaginase group. However, these did not increase bleeding events or infection-associated mortalities. When taking the convenience and superior efficacy in preventing CNSL into consideration, PEG asparaginase is a candidate for first-line treatment of adolescent and adult ALL. A larger prospective clinical trial is required to further confirm this point of view.

Low-dose decitabine enhances chidamide-induced apoptosis in adult acute lymphoblast leukemia, especially for p16-deleted patients through DNA damage.

AIM: To investigate the combined action of decitabine (DAC) with chidamide (CS055) on acute lymphoblastic leukemia (ALL) cells. MATERIALS & METHODS: ALL cell lines as well as primary cells from 17 ALL patients were subjected to different treatments and thereafter cell counting Kit-8 (CCK-8) assay, flow cytometry and western blot were employed to determine IC50, apoptosis and checkpoint kinase 1 and γH2A.X expression. RESULTS: Low-dose DAC combined with CS055 could effectively kill ALL cells by the reduction of cell viability and induction of apoptosis. This was also observed in primary cells from 17 ALL patients, especially for those with p16 gene deletion. Suppression of checkpoint kinase 1 phosphorylation and upregulation of γH2A.X expression was demonstrated to participate in DAC plus CS055-induced apoptosis. CONCLUSION: Low-dose DAC could enhance chidamide-induced apoptosis in adult ALL, especially for patients with p16 gene deletion through DNA damage.

CXCR5+CD8+ T cells present elevated capacity in mediating cytotoxicity toward autologous tumor cells through interleukin 10 in diffuse large B-cell lymphoma.

Diffuse large B-cell lymphoma (DLBCL) is a common and aggressive subtype of non-Hodgkin's lymphomas, with limited treatment options in refractory and relapsed patients. Growing evidence supports the notion that CD8+ T cell immunity could be utilized to eliminate B cell lymphomas. CXCR5+CD8+ T cell is a novel cell subtype and share CXCR5 expression with CD19+ tumor cells. In this study, we investigated the frequency and function of existing CXCR5+CD8+ T cells in DLBCL patients. We found that DLBCL patients as a group demonstrated significantly higher level of CXCR5+CD8+ T cells than healthy individuals, with huge variability in each patient. Using anti-CD3/CD28-stimulated CD8+ T cells as effector (E) cells and autologous CD19+ tumor cells as target (T) cells, at high E:T ratio, no difference between the intensities of CXCR5+CD8+ T cell- and CXCR5-CD8+ T cell-mediated cytotoxicity were observed. However, at intermediate and low E:T ratios, the CXCR5+CD8+ T cells presented stronger cytotoxicity than CXCR5-CD8+ T cells. The expressions of granzyme A, granzyme B, and perforin were significantly higher in CXCR5+CD8+ T cells than in CXCR5-CD8+ T cells, with no significant difference in the level of degranulation. Tumor cells in DLBCL were known to secrete high level of interleukin 10 (IL-10). We therefore blocked the IL-10/IL-10R pathway, and found that the expressions of granzyme A, granzyme B, and perforin by CXCR5+CD8+ T cells were significantly elevated. Together, these results suggest that CXCR5+CD8+ T cells are potential candidates of CD8+ T cell-based immunotherapies, could mediate elimination of autologous tumor cells in DLBCL patients, but are also susceptible to IL-10-mediated suppression.

High BCL11A Expression in Adult Acute Myeloid Leukemia Patients Predicts a Worse Clinical Outcome.

BACKGROUND: Recent reports showed BCL11A may be causatively involved in myeloid leukemia. This study investigated the relationship between BCL11A expression levels and adult acute myeloid leukemia patient characteristics as well as clinical outcomes. METHODS: RT-PCR was employed to detect BCL11A gene expression levels in 80 patients with acute myeloid leukemia. RESULTS: Median BCL11A expression levels of 80 AML bone marrow samples were found to be higher than the control group (0.039 vs. 0.014, p < 0.005). Patients with low BCL11A expression levels had a significantly higher CR (complete remission) rate compared with patients with high BCL11A expression levels (90% vs. 53%, p < 0.005). Moreover, the median OS (overall survival) in patients with low BCL11A expression (268 d) was also longer than that in patients with high BCL11A expression (101.5 d) (p < 0.05). No significant difference was observed between the high and low BCL11A groups with respect to white blood cells, haemoglobin, platelet count, French-American-Britain (FAB) subtypes, percentage of blasts in bone marrow, peripheral blood, cytogenetic risk groups, and CD34 expression. CONCLUSIONS: Adult acute myeloid leukemia had a higher BCL11A expression level. High BCL11A expression level was correlated with lower CR rate and shorter OS, suggesting that BCL11A expression could potentially be used as a prognosis indicator.

Defined, serum/feeder-free conditions for expansion and drug screening of primary B-acute lymphoblastic leukemia.

Functional screening for compounds represents a major hurdle in the development of rational therapeutics for B-acute lymphoblastic leukemia (B-ALL). In addition, using cell lines as valid models for evaluating responses to novel drug therapies raises serious concerns, as cell lines are prone to genotypic/phenotypic drift and loss of heterogeneity in vitro. Here, we reported that OP9 cells, not OP9-derived adipocytes (OP9TA), support the growth of primary B-ALL cells in vitro. To identify the factors from OP9 cells that support the growth of primary B-ALL cells, we performed RNA-Seq to analyze the gene expression profiles of OP9 and OP9TA cells. We thus developed a defined, serum/feeder-free condition (FI76V) that can support the expansion of a range of clinically distinct primary B-ALL cells that still maintain their leukemia-initiating ability. We demonstrated the suitability of high-throughput drug screening based on our B-ALL cultured conditions. Upon screening 378 kinase inhibitors, we identified a cluster of 17 kinase inhibitors that can efficiently kill B-ALL cells in vitro. Importantly, we demonstrated the synergistic cytotoxicity of dinaciclib/BTG226 to B-ALL cells. Taken together, we developed a defined condition for the ex vivo expansion of primary B-ALL cells that is suitable for high-throughput screening of novel compounds.