Landscape of Bone Marrow Metastasis in Human Neuroblastoma Unraveled by Transcriptomics and Deep Multiplex Imaging

Radiomics models to predict bone marrow metastasis of neuroblastoma using CT

Background

Bone marrow is the leading site for metastasis from neuroblastoma and affects the prognosis of patients with neuroblastoma. However, the accurate diagnosis of bone marrow metastasis is limited by the high spatial and temporal heterogeneity of neuroblastoma. Radiomics analysis has been applied in various cancers to build accurate diagnostic models but has not yet been applied to bone marrow metastasis of neuroblastoma.

Methods

We retrospectively collected information from 187 patients pathologically diagnosed with neuroblastoma and divided them into training and validation sets in a ratio of 7:3. A total of 2632 radiomics features were retrieved from venous and arterial phases of contrast-enhanced computed tomography (CT), and nine machine learning approaches were used to build radiomics models, including multilayer perceptron (MLP), extreme gradient boosting, and random forest. We also constructed radiomics-clinical models that combined radiomics features with clinical predictors such as age, gender, ascites, and lymph gland metastasis. The performance of the models was evaluated with receiver operating characteristics (ROC) curves, calibration curves, and risk decile plots.

Results

The MLP radiomics model yielded an area under the ROC curve (AUC) of 0.97 (95% confidence interval [CI]: 0.95-0.99) on the training set and 0.90 (95% CI: 0.82-0.95) on the validation set. The radiomics-clinical model using an MLP yielded an AUC of 0.93 (95% CI: 0.89-0.96) on the training set and 0.91 (95% CI: 0.85-0.97) on the validation set.

Conclusions

MLP-based radiomics and radiomics-clinical models can precisely predict bone marrow metastasis in patients with neuroblastoma.

A human neural crest model reveals the developmental impact of neuroblastoma-associated chromosomal aberrations

Early childhood tumours arise from transformed embryonic cells, which often carry large copy number alterations (CNA). However, it remains unclear how CNAs contribute to embryonic tumourigenesis due to a lack of suitable models. Here we employ female human embryonic stem cell (hESC) differentiation and single-cell transcriptome and epigenome analysis to assess the effects of chromosome 17q/1q gains, which are prevalent in the embryonal tumour neuroblastoma (NB). We show that CNAs impair the specification of trunk neural crest (NC) cells and their sympathoadrenal derivatives, the putative cells-of-origin of NB. This effect is exacerbated upon overexpression of MYCN, whose amplification co-occurs with CNAs in NB. Moreover, CNAs potentiate the pro-tumourigenic effects of MYCN and mutant NC cells resemble NB cells in tumours. These changes correlate with a stepwise aberration of developmental transcription factor networks. Together, our results sketch a mechanistic framework for the CNA-driven initiation of embryonal tumours.

GM-CSF, G-CSF or no cytokine therapy with anti-GD2 immunotherapy for high-risk neuroblastoma

Colony-stimulating factors have been shown to improve anti-disialoganglioside 2 (anti-GD2) monoclonal antibody response in high-risk neuroblastoma by enhancing antibody-dependent cell-mediated cytotoxicity (ADCC). A substantial amount of research has focused on recombinant human granulocyte-macrophage colony-stimulating factor (GM-CSF) as an adjuvant to anti-GD2 monoclonal antibodies. There may be a disparity in care among patients as access to GM-CSF therapy and anti-GD2 monoclonal antibodies is not uniform. Only select countries have approved these agents for use, and even with regulatory approvals, access to these agents can be complex and cost prohibitive. This comprehensive review summarizes clinical data regarding efficacy and safety of GM-CSF, recombinant human granulocyte colony-stimulating factor (G-CSF) or no cytokine in combination with anti-GD2 monoclonal antibodies (ie, dinutuximab, dinutuximab beta or naxitamab) for immunotherapy of patients with high-risk neuroblastoma. A substantial body of clinical data support the immunotherapy combination of anti-GD2 monoclonal antibodies and GM-CSF. In contrast, clinical data supporting the use of G-CSF are limited. No formal comparison between GM-CSF, G-CSF and no cytokine has been identified. The treatment of high-risk neuroblastoma with anti-GD2 therapy plus GM-CSF is well established. Suboptimal efficacy outcomes with G-CSF raise concerns about its suitability as an alternative to GM-CSF as an adjuvant in immunotherapy for patients with high-risk neuroblastoma. While programs exist to facilitate obtaining GM-CSF and anti-GD2 monoclonal antibodies in regions where they are not commercially available, continued work is needed to ensure equitable therapeutic options are available globally.

Comparative transcriptomics coupled to developmental grading via transgenic zebrafish reporter strains identifies conserved features in neutrophil maturation

Neutrophils are evolutionarily conserved innate immune cells playing pivotal roles in host defense. Zebrafish models have contributed substantially to our understanding of neutrophil functions but similarities to human neutrophil maturation have not been systematically characterized, which limits their applicability to studying human disease. Here we show, by generating and analysing transgenic zebrafish strains representing distinct neutrophil differentiation stages, a high-resolution transcriptional profile of neutrophil maturation. We link gene expression at each stage to characteristic transcription factors, including C/ebp-β, which is important for late neutrophil maturation. Cross-species comparison of zebrafish, mouse, and human samples confirms high molecular similarity of immature stages and discriminates zebrafish-specific from pan-species gene signatures. Applying the pan-species neutrophil maturation signature to RNA-sequencing data from human neuroblastoma patients reveals association between metastatic tumor cell infiltration in the bone marrow and an overall increase in mature neutrophils. Our detailed neutrophil maturation atlas thus provides a valuable resource for studying neutrophil function at different stages across species in health and disease.

Comprehensive review of solid tumor bone marrow metastasis

Bone marrow metastasis (BMM) of solid tumors refers to a group of diseases that originate from non-hematopoietic malignant tumor cells invading the bone marrow (BM) through complex metastatic patterns. If BMM identification is delayed, the disease will rapidly develop into disseminated carcinogenesis of the BM, which manifests as a series of hematological disorders and microangiopathic hemolytic anemia, leading to serious life-threatening conditions. Although the study of solid tumor BMM is receiving increasing attention, study remains limited, and most descriptions are derived from case reports. Currently, clinicians have insufficient understanding of BMM, and BMM occurrence is often not recognized early or treated effectively, resulting in high mortality rates. In this article, we review the epidemiology, molecular mechanisms, clinical diagnosis, treatment, and prognosis of solid tumor BMM.

Three-dimensional in vitro model of bone metastases of neuroblastoma as a tool for pharmacological evaluations

In vitro metastatic models are foreseen to introduce a breakthrough in the field of preclinical screening of more functional small-molecule pharmaceuticals and biologics. To achieve this goal, the complexity of current in vitro systems requests an appropriate upgrade to approach the three-dimensional (3D) in vivo metastatic disease. Here, we explored the potential of our 3D β-tricalcium phosphate (β-TCP) model of neuroblastoma bone metastasis for drug toxicity assessment. Tailor-made scaffolds with interconnected channels were produced by combining 3D printing and slip casting method. The organization of neuroblastoma cells into a mesenchymal stromal cell (MSC) network, cultured under bioactive conditions provided by β-TCP, was monitored by two-photon microscopy. Deposition of extracellular matrix protein Collagen I by MSCs and persistent growth of tumor cells confirmed the cell-supportive performance of our 3D model. When different neuroblastoma cells were treated with conventional chemotherapeutics, the β-TCP model provided the necessary reproducibility and accuracy of experimental readouts. Drug efficacy evaluation was done for 3D and 2D cell cultures, highlighting the need for a higher dose of chemotherapeutics under 3D conditions to achieve the expected cytotoxicity in tumor cells. Our results confirm the importance of 3D geometry in driving native connectivity between nonmalignant and tumor cells and sustain β-TCP scaffolds as a reliable and affordable drug screening platform for use in the early stages of drug discovery.

Targeting the myeloid microenvironment in neuroblastoma

Myeloid cells (granulocytes and monocytes/macrophages) play an important role in neuroblastoma. By inducing a complex immunosuppressive network, myeloid cells pose a challenge for the adaptive immune system to eliminate tumor cells, especially in high-risk neuroblastoma. This review first summarizes the pro- and anti-tumorigenic functions of myeloid cells, including granulocytes, monocytes, macrophages, and myeloid-derived suppressor cells (MDSC) during the development and progression of neuroblastoma. Secondly, we discuss how myeloid cells are engaged in the current treatment regimen and explore novel strategies to target these cells in neuroblastoma. These strategies include: (1) engaging myeloid cells as effector cells, (2) ablating myeloid cells or blocking the recruitment of myeloid cells to the tumor microenvironment and (3) reprogramming myeloid cells. Here we describe that despite their immunosuppressive traits, tumor-associated myeloid cells can still be engaged as effector cells, which is clear in anti-GD2 immunotherapy. However, their full potential is not yet reached, and myeloid cell engagement can be enhanced, for example by targeting the CD47/SIRPα axis. Though depletion of myeloid cells or blocking myeloid cell infiltration has been proven effective, this strategy also depletes possible effector cells for immunotherapy from the tumor microenvironment. Therefore, reprogramming of suppressive myeloid cells might be the optimal strategy, which reverses immunosuppressive traits, preserves myeloid cells as effectors of immunotherapy, and subsequently reactivates tumor-infiltrating T cells.

Machine‐learning radiomics to predict bone marrow metastasis of neuroblastoma using magnetic resonance imaging

Background

Neuroblastoma is one common pediatric malignancy notorious for high temporal and spatial heterogeneities. More than half of its patients develop distant metastases involving vascularized organs, especially the bone marrow. It is thus necessary to have an economical, noninvasive method without much radiation for follow‐ups. Radiomics has been used in many cancers to assist accurate diagnosis but not yet in bone marrow metastasis in neuroblastoma.

Methods

A total of 182 patients with neuroblastoma were retrospectively collected and randomly divided into the training and validation sets. Five‐hundred and seventy‐two radiomics features were extracted from magnetic resonance imaging, among which 41 significant ones were selected via T‐test for model development. We attempted 13 machine‐learning algorithms and eventually chose three best‐performed models. The integrative performance evaluations are based on the area under the curves (AUCs), calibration curves, risk deciles plots, and other indexes.

Results

Extreme gradient boosting, random forest (RF), and adaptive boosting were the top three to predict bone marrow metastases in neuroblastoma while RF was the most accurate one. Its AUC was 0.90 (0.86–0.93), F1 score was 0.82, sensitivity was 0.76, and negative predictive value was 0.79 in the training set. The values were 0.82 (0.71–0.93), 0.80, 0.75, and 0.92 in the validation set, respectively.

Conclusions

Radiomics models are likely to contribute more to metastatic diagnoses and the formulation of personalized healthcare strategies in clinics. It has great potential of being a revolutionary method to replace traditional interventions in the future.

The molecular basis of tumor metastasis and current approaches to decode targeted migration-promoting events in pediatric neuroblastoma

Cell motility is a crucial biological process that plays a critical role in the development of multicellular organisms and is essential for tissue formation and regeneration. However, uncontrolled cell motility can lead to the development of various diseases, including neoplasms. In this review, we discuss recent advances in the discovery of regulatory mechanisms underlying the metastatic spread of neuroblastoma, a solid pediatric tumor that originates in the embryonic migratory cells of the neural crest. The highly motile phenotype of metastatic neuroblastoma cells requires targeting of intracellular and extracellular processes, that, if affected, would be helpful for the treatment of high-risk patients with neuroblastoma, for whom current therapies remain inadequate. Development of new potentially migration-inhibiting compounds and standardized preclinical approaches for the selection of anti-metastatic drugs in neuroblastoma will also be discussed.

Case series on clinical applications of liquid biopsy in pediatric solid tumors: towards improved diagnostics and disease monitoring

Background and aims

Solid tumors account for about 30% of all pediatric cancers. The diagnosis is typically based on histological and molecular analysis of a primary tumor biopsy. Liquid biopsies carry several advantages over conventional tissue biopsy. However, their use for genomic analysis and response monitoring of pediatric solid tumors is still in experimental stages and mostly performed retrospectively without direct impact on patient management. In this case series we discuss six clinical cases of children with a solid tumor for whom a liquid biopsy assay was performed and demonstrate the potential of liquid biopsy for future clinical decision making.

Methods

We performed quantitative real-time PCR (RT-qPCR), droplet digital PCR (ddPCR) or reduced representation bisulphite sequencing of cell-free DNA (cfRRBS) on liquid biopsies collected from six pediatric patients with a solid tumor treated between 2017 and 2023 at the Princess Máxima Center for Pediatric Oncology in the Netherlands. Results were used to aid in clinical decision making by contribution to establish a diagnosis, by prognostication and response to therapy monitoring.

Results

In three patients cfRRBS helped to establish the diagnosis of a rhabdomyosarcoma, an Ewing sarcoma and a neuroblastoma (case 1-3). In two patients, liquid biopsies were used for prognostication, by MYCN ddPCR in a patient with neuroblastoma and by RT-qPCR testing rhabdomyosarcoma-specific mRNA in bone marrow of a patient with a rhabdomyosarcoma (case 4 and 5). In case 6, mRNA testing demonstrated disease progression and assisted clinical decision making.

Conclusion

This case series illustrates the value of liquid biopsy. We further demonstrate and recommend the use of liquid biopsies to be used in conjunction with conventional methods for the determination of metastatic status, prognostication and monitoring of treatment response in patients with pediatric solid tumors.

Single-cell transcriptomics and epigenomics unravel the role of monocytes in neuroblastoma bone marrow metastasis

Metastasis is the major cause of cancer-related deaths. Neuroblastoma (NB), a childhood tumor has been molecularly defined at the primary cancer site, however, the bone marrow (BM) as the metastatic niche of NB is poorly characterized. Here we perform single-cell transcriptomic and epigenomic profiling of BM aspirates from 11 subjects spanning three major NB subtypes and compare these to five age-matched and metastasis-free BM, followed by in-depth single cell analyses of tissue diversity and cell-cell interactions, as well as functional validation. We show that cellular plasticity of NB tumor cells is conserved upon metastasis and tumor cell type composition is NB subtype-dependent. NB cells signal to the BM microenvironment, rewiring via macrophage mgration inhibitory factor and midkine signaling specifically monocytes, which exhibit M1 and M2 features, are marked by activation of pro- and anti-inflammatory programs, and express tumor-promoting factors, reminiscent of tumor-associated macrophages. The interactions and pathways characterized in our study provide the basis for therapeutic approaches that target tumor-to-microenvironment interactions.

Utilization of temporal autoencoder for semi-supervised intracranial EEG clustering and classification

Manual visual review, annotation and categorization of electroencephalography (EEG) is a time-consuming task that is often associated with human bias and requires trained electrophysiology experts with specific domain knowledge. This challenge is now compounded by development of measurement technologies and devices allowing large-scale heterogeneous, multi-channel recordings spanning multiple brain regions over days, weeks. Currently, supervised deep-learning techniques were shown to be an effective tool for analyzing big data sets, including EEG. However, the most significant caveat in training the supervised deep-learning models in a clinical research setting is the lack of adequate gold-standard annotations created by electrophysiology experts. Here, we propose a semi-supervised machine learning technique that utilizes deep-learning methods with a minimal amount of gold-standard labels. The method utilizes a temporal autoencoder for dimensionality reduction and a small number of the expert-provided gold-standard labels used for kernel density estimating (KDE) maps. We used data from electrophysiological intracranial EEG (iEEG) recordings acquired in two hospitals with different recording systems across 39 patients to validate the method. The method achieved iEEG classification (Pathologic vs. Normal vs. Artifacts) results with an area under the receiver operating characteristic (AUROC) scores of 0.862 ± 0.037, 0.879 ± 0.042, and area under the precision-recall curve (AUPRC) scores of 0.740 ± 0.740, 0.714 ± 0.042. This demonstrates that semi-supervised methods can provide acceptable results while requiring only 100 gold-standard data samples in each classification category. Subsequently, we deployed the technique to 12 novel patients in a pseudo-prospective framework for detecting Interictal epileptiform discharges (IEDs). We show that the proposed temporal autoencoder was able to generalize to novel patients while achieving AUROC of 0.877 ± 0.067 and AUPRC of 0.705 ± 0.154.

Cytomorphological evaluation of non-haematopoietic malignancies metastasizing to the bone marrow

Bone marrow (BM) is one of the rare but important site of metastasis of solid tumors. The key steps of metastasis include invasion, intravasation, circulation, extravasation, and colonization. Tumor cells may express some adhesion molecules that promote the transmigration to the marrow space and link them to the marrow stroma with subsequent engraftment. It is important to detect the bone marrow metastasis for initial clinical staging, therapeutic selection, prognostic risk stratification, assessment of response to therapy and predicting relapse. Prognosis of non-hematopoietic malignancies with BM metastasis is dismal. Due to occulting and atypical clinical manifestations, bone marrow metastases can be easily missed or misdiagnosed, leading to higher mortality rates. The important factors on which the prognosis of patients with bone marrow metastases depends are primary tumor site, performance status, platelet count, and therapeutic regimens (systemic chemotherapy or palliative/supportive care). Further, in cases with BM metastasis with unknown primary sites, misdiagnosis can lead to delayed initiation of therapy and increased mortality. BM metastasis is seen in less than 10% of patients with metastatic cancer and is common in lung, breast or prostate carcinoma. Bone marrow metastasis can be presented as the initial presentation with hematological changes and may be misdiagnosed as a primary haematopoietic disorder. Leucoerythoblastic blood picture is the most common peripheral blood smear finding indicating BM metastasis, may be an indicator of associated BM fibrosis. Bone marrow aspiration and biopsy with immunohistochemistry (IHC) is an easy, cost effective and gold standard method of detection of BM metastasis. BM biopsy is superior to bone marrow aspirate for detection of metastasis. Morphology of metastatic cells is as per the primary site of tumor. Immunohistochemistry is a useful adjunct to morphology in reaching a definitive diagnosis even in case with carcinoma unknown primary (CUP) and also in diagnosing case of unsuspected malignancies. Though bone marrow is not among the most common site of involvement in CUP, which are liver, bone, lymph nodes and lung. But BM, if involved, the site of origin is determined using the immunohistochemistry panel applied to the metastatic deposits based on the morphology The aim of the review is to discuss the hematological findings of non-haematopoietic malignancies metastasizing to the bone marrow, emphasizing on histomorphology with IHC and its significance in establishing primary diagnosis in clinically unsuspected cases.

Expression of ADRB2 in children with neuroblastoma and its influence on prognosis

Objective

Neuroblastoma (NB), originating from sympathetic spinal tissue, is a serious threat to the life of children. Especially in the high-risk group, an overall five-year survival rate less than 50% indicates an extremely poor clinical outcome. Here, the expression the of β-2 adrenergic (ADRB2) receptor gene in tumor tissues of children with NB was detected and the correlation between its expression and clinical characteristics and prognosis was analyzed.

Methods

Forty-five tumor tissue samples and forty-eight paraffin sections of NB were obtained from Children’s Hospital of Chongqing Medical University from 2015 to 2021. Real-time fluorescence quantitative polymerase chain reaction (RT–qPCR) was utilized to detect the expression of ADRB2 at the mRNA level and immunohistochemistry (IHC) at the protein level.

Results

For the RT–qPCR, the analysis showed that the expression of ADRB2 in the high-risk group was significantly lower (P = 0.0003); in addition, there were also statistically significant differences in Shimada classification (P = 0.0025) and N-MYC amplification (P = 0.0011). Survival prognosis analysis showed that the prognosis was better with high ADRB2 expression (P = 0.0125), and the ROC curve showed that ADRB2 has a certain accuracy in predicting prognosis (AUC = 0.707, CI: 0.530–0.884). Moreover, the expression of ADRB2, N-MYC amplification and bone marrow metastasis were the factors that independently affected prognosis, and at the protein level, the results showed that the differential expression of ADRB2 was conspicuous in risk (P = 0.0041), Shimada classification (P = 0.0220) and N-MYC amplification (P = 0.0166). In addition, Kaplan–Meier curves showed that the prognosis in the group with high expression of ADRB2 was better (P = 0.0287), and the ROC curve showed that the score of ADRB2 had poor accuracy in predicting prognosis (AUC = 0.662, CI: 0.505–0.820).

Conclusion

ADRB2 is a protective potential biomarker and is expected to become a new prognostic biomolecular marker of NB.

Reduction of LPAR1 Expression in Neuroblastoma Promotes Tumor Cell Migration

Neuroblastoma is the most common extracranial solid tumor in children. Tumor metastasis in high-risk NB patients is an essential problem that impairs the survival of patients. In this study, we aimed to use a comprehensive bioinformatics analysis to identify differentially expressed genes between NB and control cells, and to explore novel prognostic markers or treatment targets in tumors. In this way, FN1, PIK3R5, LPAR6 and LPAR1 were screened out via KEGG, GO and PPI network analysis, and we verified the expression and function of LPAR1 experimentally. Our research verified the decreased expression of LPAR1 in NB cells, and the tumor migration inhibitory effects of LPA on NB cells via LPAR1. Moreover, knockdown of LPAR1 promoted NB cell migration and abolished the migration-inhibitory effects mediated by LPA-LPAR1. The tumor-suppressing effects of the LPA-LPAR1 axis suggest that LPAR1 might be a potential target for future treatment of NB.

MIF/CXCR4 signaling axis contributes to survival, invasion, and drug resistance of metastatic neuroblastoma cells in the bone marrow microenvironment

BACKGROUND

The bone marrow (BM) is the most common site of dissemination in patients with aggressive, metastatic neuroblastoma (NB). However, the molecular mechanisms underlying the aggressive behavior of NB cells in the BM niche are still greatly unknown. In the present study, we explored biological mechanisms that play a critical role in NB cell survival and progression in the BM and investigated potential therapeutic targets.

METHODS

Patient-derived bone marrow (BM) primary cultures were generated using fresh BM aspirates obtained from NB patients. NB cell lines were cultured in the presence of BM conditioned media containing cell-secreted factors, and under low oxygen levels (1% O₂) to mimic specific features of the BM microenvironment of high-risk NB patients. The BM niche was explored using cytokine profiling assays, cell migration-invasion and viability assays, flow cytometry and analysis of RNA-sequencing data. Selective pharmacological inhibition of factors identified as potential mediators of NB progression within the BM niche was performed in vitro and in vivo.

RESULTS

We identified macrophage migration inhibitory factor (MIF) as a key inflammatory cytokine involved in BM infiltration. Cytokine profiling and RNA-sequencing data analysis revealed NB cells as the main source of MIF in the BM, suggesting a potential role of MIF in tumor invasion. Exposure of NB cells to BM-conditions increased NB cell-surface expression of the MIF receptor CXCR4, which was associated with increased cell viability, enhanced migration-invasion, and activation of PI3K/AKT and MAPK/ERK signaling pathways. Moreover, subcutaneous co-injection of NB and BM cells enhanced tumor engraftment in mice. MIF inhibition with 4-IPP impaired in vitro NB aggressiveness, and improved drug response while delayed NB growth, improving survival of the NB xenograft model.

CONCLUSIONS

Our findings suggest that BM infiltration by NB cells may be mediated, in part, by MIF-CXCR4 signaling. We demonstrate the antitumor efficacy of MIF targeting in vitro and in vivo that could represent a novel therapeutic target for patients with disseminated high-risk NB.

Elevated circulating myeloid-derived suppressor cells associated with poor prognosis in B-cell non-Hodgkin's lymphoma patients

Introduction

Myeloid‐derived suppressor cells (MDSCs) are a heterogeneous cell population with the ability to suppress immune responses. MDSCs usually cluster in cancer, inflammation, and autoimmune diseases. Although there have been some studies on MDSCs in non‐Hodgkin lymphoma (NHL), the correlation between the peripheral levels of MDSCs in patients with various subtypes of B cell NHL and clinical features and prognosis remains inconclusive. This study aimed at the issue.

Methods

101 patients with B cell NHL and 15 age‐matched healthy controls were included in this study. Flow cytometric detection of monocytic‐MDSCs (M‐MDSCs) and granulocytic‐MDSCs (G‐MDSCs) was done.

Results

In this study, we found that counts of circulating M‐MDSCs and G‐MDSCs were significantly increased in different clinical statuses of B‐NHL patients compared to healthy controls. Similarly, a significant increase in the levels of M‐MDSCs and G‐MDSCs was found among the diverse types of B‐NHL compared with healthy donors. Stratification studies indicated MDSCs expansion was closely associated with disease progression (tumor stage, LDH levels and B syndromes). Moreover, the overall survival time of patients with G‐MDSCs (%) ≥ 98.70% was shorter than patients with G‐MDSCs (%) < 98.70% in newly diagnosed B‐NHL subgroup, meanwhile, there was a significant difference in survival of patients with M‐MDSCs (%) ≥ 7.19% compared to patients with M‐MDSCs (%) < 7.19% in relapsed B‐NHL subgroup.

Conclusion

Our results suggested that M‐MDSCs and G‐MDSCs may be a potential and efficient index to evaluate the prognosis of B‐NHL patients.