Background/Objectives: Spontaneous remission (SR) of acute myeloid leukemia (AML) offers unique clinical insights into host anti-tumor immunity. However, the comprehensive clinical landscape and molecular dynamics of blast clearance and subsequent relapse remain unclear. This study aimed to elucidate these dynamics. Methods: We conducted a two-phase observational study: a systematic pooled analysis of 66 adult AML SR cases (1990-2024) to define clinical triggers and outcomes and longitudinal molecular tracking of two institutional cases to map clonal shifts (with immune profiling for Patient 1 and genomic tracking for both). Results: In the pooled analysis, infection was the predominant trigger, accounting for 78.6% (95% CI: 65.6-88.4%) of SR events. The dataset showed male predominance and monocytic leukemia enrichment (57.6% [95% CI: 44.1-70.4%]), suggesting lineage-specific susceptibility. SR duration and relapse risk were independent of the infection trigger, AML subtype, or age. When integrated with these clinical patterns, institutional tracking was consistent with a biphasic evolutionary model: an acute IL-8 surge alongside NKT and CD4+ T cell activation coincided with blast clearance, as observed primarily in Patient 1. Subsequently, the emergence of TP53 or NRAS mutations within persistent DNMT3A-mutated clones during relapse raised the hypothesis that unresolved chronic inflammation could potentially exert selective pressure favoring resistant subclones. Such interpretations remain correlational and require prospective validation. Conclusions: Our findings outline a clinical-evolutionary framework for AML SR. Remission durability likely relies on balancing acute immune activation with underlying clonal stability. These observational insights highlight complex immune-genomic crosstalk, generating hypotheses for future prospective investigations.

Lymphoma-associated haemophagocytic lymphohistiocytosis (LA-HLH) is associated with a high mortality rate, making early diagnosis and appropriate treatment critical for improving patient outcomes. In this study, we enrolled 126 patients diagnosed with LA-HLH and 254 with non-LA-HLH. A machine learning-based predictive model was developed and validated to enable timely differentiation of LA-HLH from other HLH subtypes. The model incorporated 11 predictive variables for early LA-HLH prediction, among which the top five most influential were age, ferritin, monocyte percentage, haemoglobin and platelet count. Among seven machine learning algorithms evaluated, the random forest model demonstrated the best performance, achieving an area under the curve (AUC) of 0.946 on the training set and 0.794 on the validation set. We subsequently evaluated combined models that incorporated disease-specific indicators such as soluble interleukin-2 receptor (sCD25) and PET-CT SUVmax, which resulted in a non-significant increase in AUC on the validation set. Finally, the optimal model was deployed as a web-based tool to support early aetiological differentiation. This may facilitate prompt initiation of targeted examination and appropriate treatment for patients with LA-HLH.

Temporal modeling is crucial for gesture recognition. Conventional three e dimensional convolutional neural networks (3D CNNs) can efficiently model spatial appearance and temporal evolution but consume substantial computational resources. Two dimensional convolutional neural networks (2D CNNs) enjoy the advantages of lightness and fast inference but cannot capture temporal relationships. Based on this situation, we propose a novel plug-and-play temporalchannel attention module, namely TCA. Our TCA can facilitate information interaction in the temporal dimension, which makes up for the lack of temporal modeling in 2D CNNs. Meanwhile, TCA is able to adaptively refine channel-wise features and enhance the attention of networks to relevant features, which considerably improves the competitiveness of 2D CNNs in gesture recognition. By inserting TCA into 2D CNNs, such as ResNet and MobileNet-V2, we propose a simple yet effective TCA-Net. We validate the generalization performance of TCA-Net for various types of datasets, namely gesture recognition and action recognition datasets (e.g., Jester, EgoGesture, and Kinetics-400). Further, we conduct extensive experiments on the gesture recognition datasets to evaluate the efficiency of TCA-Net. The experimental results demonstrate that TCA-Net achieves 97.0 % and 94.5 % top-1 accuracy on Jester and EgoGesture, respectively, which not only consistently outperforms the initial 2D CNNs but also obtains superior or comparable results to the state-of-the-art methods. For Kinetics- 400, the top- 1 accuracy of 75.1 % is comparable to the state-of-the-art methods when the input is $\mathbf{8}$ frames. Thus, we demonstrate the effectiveness of the proposed method

[This corrects the article DOI: 10.3389/fimmu.2025.1725218.].

Visual object-goal navigation requires an agent to make decisions to search for specified target objects within an unknown environment. While learning-based approaches have achieved progress, they still face two limitations: (1) visual representation: the lack of a low-dimensional representation that balances object priors with current observations, and the absence of effective modeling of scene spatial structures, which restricts the agent’s ability to perceive its surroundings. (2) policy learning: existing reward functions often fail to consider the visual essence of object goal-driven tasks, leading to inefficient learning and suboptimal performance. To address these issues, this paper introduces a semantic-spatial fusion representation framework that incorporates both object semantics and the intrinsic spatial structure of the scene. Specifically, a new object context matrix captures semantic relationships between objects while providing distinct low-dimensional representations for different observations, and an episode memory graph is also constructed and dynamically updated based on observation similarity and geodesic distance to represent the spatial structure of the agent’s environment in real-time. Then, a transformer-based adaptive multimodal feature fusion module is proposed to integrate these dual representations. Moreover, a sparse reward function is designed based on the target’s bounding box to guide the agent to learn correctly. The proposed method is the first to be evaluated on both the simulation platform AI2-THOR and the real-world dataset AVD, demonstrating its generalization in unseen environments. The method is also deployed on a physical mobile robot and tested in real-world scenarios, further validating its practical effectiveness.

BackgroundBasic leucine zipper ATF-like transcription factor (BATF) is a nuclear basic leucine zipper protein affiliated with the AP-1/ATF superfamily. Previous research has confirmed that BATF expression plays a significant role in the tumour microenvironment. However, the associations between BATF expression and prognoses in acute myeloid leukaemia (AML) patients and their immunological effects remain unclear.MethodsGenomic and clinical AML data were got from the TCGA (TCGA-LAML) and GEO (GSE37642) databases for the subsequent analysis. The expression levels of BATF in AML patients were assessed using GEPIA, and the results were verified by qRT-PCR and Western blotting. In the meantime, the prognostic value of BATF was evaluated using univariate and multivariate analyses, receiver operating characteristic (ROC) curve (AUC) analysis, and Kaplan-Meier (KM) survival analysis. EdU, colony formation, and CCK-8 assays were employed to evaluate the proliferation of cells. Moreover, we detected the association of BATF expression with drug sensitivity through database analysis and in vivo experiments. To further investigate the mechanism of action of BATF in AML, RNA sequencing (RNA-seq) analysis was performed, followed by pathway enrichment analysis using Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis. Meanwhile, we detected the connection between BATF expression and the proportions of immune cells via flow cytometry in C1498 mouse model of AML. Finally, we investigated the association between BATF expression and cell-cell communication within the AML cell population using single-cell sequencing.ResultsIn this study, we thoroughly investigated the role of BATF in AML. First, we observed a significant elevation in the expression of BATF in patients and cells in AML. Further analysis revealed an association between high BATF expression and poor prognosis in AML. Additionally, BATF expression was found to promote the proliferative capacity of AML cells. Moreover, the results showed that the expression level of BATF dramatically affected the effect of chemotherapy in AML patients. We also discovered that BATF expression could activate multiple immune-related pathways, altering the proportions of CD8+T cells and NK cells, suggesting that BATF may be a regulator of immune cell infiltration. Finally, there were differences in receptor ligand pairs between AML cells with high and low expression of BATF and immune cells.ConclusionBioinformatics analysis and experimental verification revealed that BATF expression could alter the proportions of CD8+T cells and NK cells in AML and affect drug sensitivity, making it a potential treatment target for AML.

In the field of task planning for service robots, large language model (LLM)-based approaches have shown increasing potential but still struggle with responding to complex user demands and grounded task planning. In this letter, we propose a user-demand-oriented adaptive grounded task planning system that integrates three modules: User-Centric Adaptive Task Customization (UATC), Task-Driven Scene Graph Pruning (TSGP), and Grounded Task Planning (GTP). UATC utilizes semantic demand-task contrastive learning to develop a novel demand parsing model for user natural language, generating personalized tasks that align with user needs and environmental constraints. TSGP employs LLM-based scene graph pruning and recursive traversal strategies to extract task-relevant environmental information. GTP generates an initial task plan using LLMs through action selection and effect judgment, followed by a three-stage refinement incorporating environmental constraints to ensure precise grounding of actions. Experimental results demonstrate that our approach can customize executable tasks in response to diverse user demands, and achieves strong planning performance with an executability rate of 91.39% and a task success rate of 78.33% in VirtualHome, outperforming existing state-of-the-art baselines. Moreover, a real-world deployment on the TIAGo robot further validates the system's applicability in physical environments.

BackgroundDiffuse large B cell lymphoma (DLBCL) is a highly heterogeneous lymphoid neoplasm characterized by diverse gene expression profiles and genetic alterations, resulting in substantial variations in clinical features and response to therapy. The cyclic GMP-AMP synthase (cGAS)-stimulator of interferon genes (STING) pathway plays a central role in innate immune response and affects the development of DLBCL. However, the relationship between genetic polymorphisms in genes involved in the cGAS-STING-mediated signaling pathway and their role in DLBCL remains underexplored.MethodsA total of 147 patients with DLBCL and 247 healthy controls were recruited. Single nucleotide polymorphism (SNP) genotyping was conducted using the MassARRAY platform. We evaluated the associations between the selected SNPs and DLBCL susceptibility, clinical features, and survival.ResultsIn our study, TREX1 rs11797 and CXCL10 rs4508917 showed significant association with DLBCL susceptibility. IFNB1 rs1051922 was correlated with white blood cell (WBC) and monocyte count at diagnosis. TREX1 rs11797 and IFNB1 rs1051922 were associated with chemotherapy response in DLBCL. Moreover, PRMT1 rs975484 and CXCL10 rs8878 were associated with the overall survival of patients with DLBCL. Notably, PRMT1 rs975484 was also correlated with hemoglobin (HGB) level, and may serve as an independent favorable prognostic factor in DLBCL.ConclusionsOur findings suggest that SNPs involved in cGAS-STING-mediated type I interferon pathway may influence DLBCL susceptibility, treatment response, and prognosis, highlighting their potential as biomarkers for risk stratification and for guiding individualized disease monitoring.

Diffuse large B-cell lymphoma (DLBCL), the most common B-cell non-Hodgkin lymphoma (B-NHL), is characterized by strong aggression, high heterogeneity, and poor prognosis. Consequently, there is an urgent need to identify crucial therapeutic targets. Here, we found that the transcription factor zinc-finger and homeobox 2 (ZHX2) was highly expressed in DLBCL. Subsequently, ZHX2 was proven to be critical for promoting DLBCL cell proliferation by inhibiting ferroptosis. Mechanistically, ZHX2 bound to the promoter region of the solute carrier family 3-member 2 (SLC3A2) gene through liquid-liquid phase separation (LLPS) and activated its function to negatively regulate ferroptosis. Furthermore, we constructed lipid nanoparticles ZHX2-siRNA@LNP targeting DLBCL, which effectively inhibited the growth of the tumors in vivo. In summary, our study indicated that the LLPS of ZHX2 protected DLBCL against ferroptosis through induction of SLC3A2, and disturbing it with ZHX2-siRNA@LNP could significantly repress DLBCL, providing a promising therapeutic strategy for DLBCL.

The study aimed to integrate genes associated with endoplasmic reticulum stress and immune responses to create a validated prognostic model for DLBCL. The research identified novel prognostic model and provided insights for developing innovative immunotherapies and anti-tumor strategies for DLBCL. The study utilized GEO data to identify prognostically significant genes via Cox regression and developed a Lasso-Cox-based prognostic model. Patients were divided into high- and low-risk groups, with survival differences analyzed using Kaplan–Meier. Independent prognostic factors were identified through univariate and multivariate Cox analyses, and model accuracy was evaluated with ROC curves. A nomogram was developed utilizing the "rms" R package. Immune infiltration and microenvironment differences were assessed with CIBERSORT, ssGSEA, and ESTIMATE. Our study developed a prognostic model using 10 genes that effectively categorized DLBCL patients into high- and low-risk cohorts, and the high-risk group revealed poorer outcomes (P < 0.001). Predictive AUCs for survival at 1, 3, and 5 years were 0.667, 0.727, and 0.729. Significant differences were observed in DLBCL samples stratified by stage, ECOG score, and LDH level. In patients aged < 65 years, with < 2 extranodal sites, ECOG < 2, GCB or non-GCB subtype, normal LDH levels, or stage III-IV, high-risk patients had worse survival outcomes (P < 0.05). The risk score model outperformed other clinicopathological factors in predicting OS (P < 0.0001, HR = 3.373) and was an independent risk factor alongside age, ECOG score, and COO classification. Significant differences were observed in immune cell infiltration, immune functions, checkpoints, and tumor priming between risk groups. In this study, we developed a risk scoring model based on endoplasmic reticulum stress and immunity to stratify DLBCL patients into low- and high-risk groups. Integrating this prognostic model with clinical parameters, we constructed a comprehensive nomogram. The identification of 10 key genes offers valuable prognostic targets for treatment and provides new perspectives for advancing therapeutic strategies in DLBCL.

Acute myeloid leukemia is a highly heterogeneous hematopoietic malignancy, and we constructed a prognostic signature combining disulfidptosis-related genes and ferroptosis-related genes to predict the prognosis, immunotherapy response, and drug sensitivity of acute myeloid leukemia (AML) patients. The TCGA-LAML datasets underwent random partitioning into training and validation sets. Subsequently, a prognostic risk signature was formulated using the least absolute shrinkage and selection operator algorithm. Kaplan-Meier survival analysis and receiver operating characteristic curve analysis were employed to assess the clinical significance of the signature. The results of the immune infiltration difference analyses are displayed, and the drug sensitivity analyses were used to identify potentially effective drugs for AML patients. qPCR was employed to validate the expression levels of the signature genes and compared the signature with existing signatures and mutated genes. Univariate and multivariate Cox regression analyses have underscored the signature as an autonomous prognostic risk determinant. Scrutiny into immune infiltration has unveiled significant associations: the risk score exhibits a favorable correlation with monocyte and M2 macrophage counts but an adverse correlation with resting mast cell counts. The expression patterns of immune checkpoint genes diverge between the distinct risk cohorts. Patients categorized as high-risk demonstrate enhanced benefits from cyclopamine, 443654, and 770041, whereas those classified as low-risk exhibit more pronounced advantages from cytarabine and AZD6244. The risk signature demonstrates superior prognostic accuracy compared to established signatures and mutated genes. In summary, our study may provide potential prognostic biomarkers and individualized precision therapy for AML patients.

Diffuse large B-cell lymphoma (DLBCL) is a highly aggressive and heterogeneous haematological malignancy with poor outcomes, underscoring the need for novel therapeutic targets to improve remission rates. SUMO (small ubiquitin-related modifier)-specific protease 2 (SENP2) has been demonstrated to exert pleiotropic functions across diverse physiological processes and oncogenic transformation. Nevertheless, its precise function in DLBCL has rarely been investigated. Our research revealed that SENP2 was markedly overexpressed in DLBCL and its elevated levels were correlated with unfavourable prognosis. Furthermore, we constructed an integrative nomogram incorporating SENP2 expression and the International Prognostic Index score, which demonstrated robust predictive performance. Subsequently, SENP2 was validated to critically promote DLBCL cell proliferation both in vitro and in vivo. To investigate the underlying mechanisms, we performed RNA sequencing coupled with tumour infiltrating immune cells analysis, revealing that SENP2 knockout reduced myeloid-derived suppressor cell accumulation while simultaneously enhancing both the infiltration and functional activation of CD8+ T cells. Finally, we performed virtual screening of Food and Drug Administration-approved drugs against SENP2, followed by re-docking analysis and identified four of the most promising candidates. Collectively, our findings characterized SENP2 as a novel prognostic biomarker and a promising therapeutic target in DLBCL.

Acute myeloid leukemia (AML) is a clonal malignancy originating from leukemia stem cells, characterized by a poor prognosis, underscoring the necessity for novel therapeutic targets and treatment methodologies. This study focuses on Ras homolog family member F, filopodia associated (RHOF), a Rho guanosine triphosphatase (GTPase) family member. We found that RHOF is overexpressed in AML, correlating with an adverse prognosis. Our gain- and loss-of-function experiments revealed that RHOF overexpression enhances proliferation and impedes apoptosis in AML cells in vitro. Conversely, genetic suppression of RHOF markedly reduced the leukemia burden in a human AML xenograft mouse model. Furthermore, we investigated the synergistic effect of RHOF downregulation and chemotherapy, demonstrating significant therapeutic efficacy in vivo. Mechanistically, RHOF activates the AKT/β-catenin signaling pathway, thereby accelerating the progression of AML. Our findings elucidate the pivotal role of RHOF in AML pathogenesis and propose RHOF inhibition as a promising therapeutic approach for AML management.

With the increase in the number of hearing-impaired people in the world, sign language recognition (SLR) has attracted extensive attention from scholars. Given the problems existing in the current research of SLR, such as parameter expansion and unsatisfactory performance of feature extraction, a novel skeleton-based method is proposed in this paper. The Asymmetric Multi-branch Graph Convolution Network (AM-GCN), composed of a spatial graph convolution and an Asymmetric Multi-branch Temporal Convolution (MTC), is constructed to achieve the acquisition and processing of graph structure. MTC utilizes multi-branch dilated convolution to expand the receptive field and enhance information dependence. To effectively extract discriminative spatiotemporal information from a large amount of information, the Spatial and Temporal Fusion Attention module (STFA) is proposed. The STFA maintains spatiotemporal consistency and obtains the fused attention map, which substantially facilitates spatiotemporal feature learning. In this article, Asymmetric Convolution Channel Attention (ACCA) is used as channel attention. Some experiments are carried out on a processed dataset obtained from video transformation, confirming the robustness of the ACCA for image flipping and rotation. The STFA and ACCA jointly form a spatial-temporal-channel attention module to extract distinguishing features and enhance the model representation. Eventually, the attention module is inserted into the AM-GCN, attaining AM-GCN-A, which is experimented on the WLASL2000, AUTSL, and CSL datasets. The top 1 accuracy is 57.01 % , 96.27 % , and 98.20 % , respectively. The results are competitive with the state-of-the-art methods and prove the effectiveness of the model.

High-dose cytarabine (HDAC) is commonly used for consolidation therapy in young acute myeloid leukemia (AML) patients, but the dosage of cytarabine is still controversial in the clinic due to its obvious post-chemotherapy adverse effects. The aim of this study was to contrast the efficacy in different dose groups of cytarabine after consolidation therapy in Chinese AML patients. AML patients treated with cytarabine consolidation at Qilu Hospital, Shandong University from January 2010 to September 2022 were retrospectively analyzed, from which 346 AML patients with relatively complete follow-up data were selected for this study. We compared the patients’ overall survival (OS) rate, relapse-free survival (RFS) rate, and hematologic adverse events in terms of their general characteristics, cytarabine consolidation therapy dose, consolidation course, 2022 European Leukemia Net (ELN) risk stratification, and transplantation. In AML patients under 60 years of age, the 5-year RFS rate with high-dose cytarabine consolidation therapy was superior to that of small-dose cytarabine (P = 0.024), while the 5-year RFS rate was comparable in the high-dose and intermediate-dose groups, and there was no obvious difference among the three groups in the 5-year OS rate (P > 0.05). OS and RFS of those given more than 3 courses of cytarabine consolidation therapy were better than those in the 1–2 courses group (P = 0.060, P = 0.040). OS and RFS were better in patients with cumulative dose of cytarabine ≥ 36g than in patients with cumulative dose < 36g (P < 0.05), but cumulative dose ≥ 54g was comparable in OS and RFS with ≥ 36–< 54g group (P > 0.05). There was no significant difference in hematologic adverse effects among the three treatment groups. In the latest ELN risk stratification favorable-risk group, the cumulative dose of cytarabine ≥ 36g had a better 5-year RFS rate than the < 36g group (P = 0.038), and in the intermediate-risk group the 5-year OS rate and RFS rate were better in the ≥ 36g group than the < 36g group (P = 0.012, 0.025). In addition, the prognosis of transplanted patients was better than that of non-transplanted patients, whereas in non-transplanted patients, consolidation therapy with ≥ 36g cytarabine can effectively improve outcomes. Multivariate analysis indicated that age, fibrinogen (FIB) and the cumulative dose of cytarabine of ≥ 36–< 54g were predictors of OS, while age, white blood cell (WBC) and HDAC were predictors of RFS. The results of the study showed that consolidation therapy with cytarabine up to a cumulative dose of ≥ 36–< 54g in AML patients who did not undergo transplantation significantly improved patient prognosis. In the latest ELN risk stratification, cumulative doses of cytarabine ≥ 36g had a better prognosis in favorable and intermediate-risk patients.

Low power consumption and stable performance insensitive to power supply are highly required for field-effect transistors integrated in portable technologies. Here, we report a mechanism of bias-independent sub-60 mV/dec subthreshold swing (SS) in ballistic cold-source field-effect transistors (CS-FETs) for portable electronics. Our first-principles and quantum-transport simulations demonstrate that, in the ballistic-transport regime, the energy alignment of the number of conduction modes (NOCM) between the drain and source electrodes is critical to achieving bias-independent SS of C-31/MoS2-based CS-FETs. By revealing the connection between NOCM and density of states (DOS), we propose a device model to demonstrate how similar slopes of the NOCM and DOS in the drain falling into the gate window can stabilize the SS of the devices under different bias. This study underscores the significance of drain DOS engineering in the design of bias-insensitive CS-FETs for portable electronic applications.

Within convolutional neural networks, convolutional operations are good at extracting local features, but have difficulty in capturing global representations. For Vision Transformer, multi-head self-attention can capture feature dependencies over long distance, but can destruct local feature details. Based on this, we propose a novel lightweight model, named HybridNet, based on MobileNet-v2 and Vision Transformer, capable of combining the advantages of both CNNs and Vision Transformer. In addition, to enhance the capability of HybridNet for temporal information interaction, we incorporate temporal-channel attention in HybridNet. We conducted experiments on Kinetics-400, Jester, and EgoGesture datasets to validate the effectiveness of HybridNet. The experimental results demonstrate that the lightweight model HybridNet achieves 96.3% and 93.9% accuracy on Jester and EgoGesture, respectively, obtaining the performance close to or even comparable with the state-of-the-art methods. Last but not least, we take HybridNet as the real-time gesture recognition model and use the recognition results as commands to control robots in the simulation environment to achieve human–robot interaction. The use of gesture interaction between humans and robots improves communication, facilitates physical collaboration, enables non-verbal expression, enhances accessibility, and creates a more engaging user experience. It adds a dimension of intuitiveness and efficiency to human–robot interaction, making it more dynamic and interactive.

Cuproptosis is a newly defined form of programmed cell death that relies on mitochondria respiration. Long noncoding RNAs (lncRNAs) play crucial roles in tumorigenesis and metastasis. However, whether cuproptosis-related lncRNAs are involved in the pathogenesis of diffuse large B cell lymphoma (DLBCL) remains unclear. This study aimed to identify the prognostic signatures of cuproptosis-related lncRNAs in DLBCL and investigate their potential molecular functions. RNA-Seq data and clinical information for DLBCL were collected from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO). Cuproptosis-related lncRNAs were screened out through Pearson correlation analysis. Utilizing univariate Cox, least absolute shrinkage and selection operator (Lasso) and multivariate Cox regression analysis, we identified seven cuproptosis-related lncRNAs and developed a risk prediction model to evaluate its prognostic value across multiple groups. GO and KEGG functional analyses, single-sample GSEA (ssGSEA), and the ESTIMATE algorithm were used to analyze the mechanisms and immune status between the different risk groups. Additionally, drug sensitivity analysis identified drugs with potential efficacy in DLBCL. Finally, the protein-protein interaction (PPI) network were constructed based on the weighted gene co-expression network analysis (WGCNA). We identified a set of seven cuproptosis-related lncRNAs including LINC00294, RNF139-AS1, LINC00654, WWC2-AS2, LINC00661, LINC01165 and LINC01398, based on which we constructed a risk model for DLBCL. The high-risk group was associated with shorter survival time than the low-risk group, and the signature-based risk score demonstrated superior prognostic ability for DLBCL patients compared to traditional clinical features. By analyzing the immune landscapes between two groups, we found that immunosuppressive cell types were significantly increased in high-risk DLBCL group. Moreover, functional enrichment analysis highlighted the association of differentially expressed genes with metabolic, inflammatory and immune-related pathways in DLBCL patients. We also found that the high-risk group showed more sensitivity to vinorelbine and pyrimethamine. A cuproptosis-related lncRNA signature was established to predict the prognosis and provide insights into potential therapeutic strategies for DLBCL patients.

POEMS (polyneuropathy, organomegaly, endocrinopathy, monoclonal protein, skin changes) syndrome is a paraneoplastic syndrome associated with an underlying plasma cell neoplasm. According to the current diagnostic criteria for POEMS syndrome, the presence of characteristic polyneuropathy and clonal plasma cell disorder are required for diagnosis. We report a case of a Castleman disease variant of POEMS syndrome without monoclonal protein (M protein) expression, which presented with polyneuropathy, organomegaly, endocrinopathy, skin lesions, and sclerotic bone lesions. The patient was treated with lenalidomide and dexamethasone (RD), after which her symptoms improved. The findings in this case suggest that the diagnostic criteria for POEMS syndrome might require reconsideration.

Purpose The aim of this study was to explore the relationships between single-nucleotide polymorphisms (SNPs) of crucial molecules in the cGAS-STING signalling pathway and the susceptibility to, induction chemotherapy response of, and prognosis of acute myeloid leukaemia (AML) in Chinese patients. Methods Thirteen SNPs of crucial molecules in the cGAS-STING signalling pathway were genotyped in 262 AML patients using the Sequenom MassARRAY system. The associations of SNPs with susceptibility, and induction chemotherapy response were analysed using the chi-square test or Fisher’s exact test and univariate binary logistic regression, the connection of SNPs with prognosis of AML was analysed using the log-rank test, and Kaplan–Meier curves were applied for survival estimation. Results In our study, gene polymorphisms of cGAS-STING signalling pathway molecules could be vitally associated with AML. In the recessive model, the cGAS rs311678 gene polymorphism could be closely related to AML susceptibility (CC vs. TT + TC, odds ratio (OR) = 0.480, 95% confidence interval (CI) = 0.260–0.889, p = 0.020). Moreover, IKKA rs3808917 might be associated with the WBC count, cGAS rs311678 could be associated with the bone marrow (BM) blast percentage, and NF-κB rs1056890 under codominant and recessive models could be connected with the HGB level. Patients who were STING rs7380272 TT/CT carriers was likely to have higher insensitivity to induction chemotherapy than CC carriers (TT + CT vs. CC, OR = 2.917, 95% CI = 1.073–7.929, p = 0.036). Survival analysis indicated that the IKKB rs3747811 TT genotype might be associated with decreased overall survival (OS) ( p < 0.05). Conclusions SNPs of molecules in the cGAS-STING signalling pathway could be significantly associated with AML. The cGAS rs311678 gene polymorphism could be associated with AML susceptibility, the STING rs7380272 variant might be related to induction chemotherapy response, and IKKB rs3747811 tended to be associated with AML overall survival. Moreover, IKKA rs3808917 could be associated with the WBC count, cGAS rs311678 could be associated with the BM blast percentage, and NF-κB rs1056890 might be related to the HGB level.