THBS1+ myeloid cells expand in SLD hepatocellular carcinoma and contribute to immunosuppression and unfavorable prognosis through TREM1.

Hepatocellular carcinoma (HCC) is an inflammation-associated cancer arising from viral or non-viral etiologies including steatotic liver diseases (SLDs). Expansion of immunosuppressive myeloid cells is a hallmark of inflammation and cancer, but their heterogeneity in HCC is not fully resolved and might underlie immunotherapy resistance. Here, we present a high-resolution atlas of innate immune cells from patients with HCC that unravels an SLD-associated contexture characterized by influx of inflammatory and immunosuppressive myeloid cells, including a discrete population of THBS1+ regulatory myeloid (Mreg) cells expressing monocyte- and neutrophil-affiliated genes. THBS1+ Mreg cells expand in SLD-associated HCC, populate fibrotic lesions, and are associated with poor prognosis. THBS1+ Mreg cells are CD163+ but distinguished from macrophages by high expression of triggering receptor expressed on myeloid cells 1 (TREM1), which contributes to their immunosuppressive activity and promotes HCC tumor growth in vivo. Our data support myeloid subset-targeted immunotherapies to treat HCC.

STIM2 is involved in the regulation of apoptosis and the cell cycle in normal and malignant monocytic cells.

Calcium is a ubiquitous messenger that regulates a wide range of cellular functions, but its involvement in the pathophysiology of acute myeloid leukemia (AML) is not widely investigated. Here, we identified, from an analysis of The Cancer Genome Atlas and genotype-tissue expression databases, stromal interaction molecule 2 (STIM2) as being highly expressed in AML with monocytic differentiation and negatively correlated with overall survival. This was confirmed on a validation cohort of 407 AML patients. We then investigated the role of STIM2 in cell proliferation, differentiation, and survival in two leukemic cell lines with monocytic potential and in normal hematopoietic stem cells. STIM2 expression increased at the RNA and protein levels upon monocyte differentiation. Phenotypically, STIM2 knockdown drastically inhibited cell proliferation and induced genomic stress with DNA double-strand breaks, as shown by increased levels of phosphorylate histone H2AXγ (p-H2AXγ), followed by activation of the cellular tumor antigen p53 pathway, decreased expression of cell cycle regulators such as cyclin-dependent kinase 1 (CDK1)-cyclin B1 and M-phase inducer phosphatase 3 (CDC25c), and a decreased apoptosis threshold with a low antiapoptotic/proapoptotic protein ratio. Our study reports STIM2 as a new actor regulating genomic stability and p53 response in terms of cell cycle and apoptosis of human normal and malignant monocytic cells.

Bromodomain-Containing Protein 9 Regulates Signaling Pathways and Reprograms the Epigenome in Immortalized Human Uterine Fibroid Cells.

Bromodomain-containing proteins (BRDs) are involved in many biological processes, most notably epigenetic regulation of transcription, and BRD dysfunction has been linked to many diseases, including tumorigenesis. However, the role of BRDs in the pathogenesis of uterine fibroids (UFs) is entirely unknown. The present study aimed to determine the expression pattern of BRD9 in UFs and matched myometrium and further assess the impact of a BRD9 inhibitor on UF phenotype and epigenetic/epitranscriptomic changes. Our studies demonstrated that the levels of BRD9 were significantly upregulated in UFs compared to matched myometrium, suggesting that the aberrant BRD expression may contribute to the pathogenesis of UFs. We then evaluated the potential roles of BRD9 using its specific inhibitor, I-BRD9. Targeted inhibition of BRD9 suppressed UF tumorigenesis with increased apoptosis and cell cycle arrest, decreased cell proliferation, and extracellular matrix deposition in UF cells. The latter is the key hallmark of UFs. Unbiased transcriptomic profiling coupled with downstream bioinformatics analysis further and extensively demonstrated that targeted inhibition of BRD9 impacted the cell cycle- and ECM-related biological pathways and reprogrammed the UF cell epigenome and epitranscriptome in UFs. Taken together, our studies support the critical role of BRD9 in UF cells and the strong interconnection between BRD9 and other pathways controlling the UF progression. Targeted inhibition of BRDs might provide a non-hormonal treatment option for this most common benign tumor in women of reproductive age.

Hormone Receptor Signaling and Breast Cancer Resistance to Anti-Tumor Immunity.

Breast cancers regroup many heterogeneous diseases unevenly responding to currently available therapies. Approximately 70-80% of breast cancers express hormone (estrogen or progesterone) receptors. Patients with these hormone-dependent breast malignancies benefit from therapies targeting endocrine pathways. Nevertheless, metastatic disease remains a major challenge despite available treatments, and relapses frequently ensue. By improving patient survival and quality of life, cancer immunotherapies have sparked considerable enthusiasm and hope in the last decade but have led to only limited success in breast cancers. In addition, only patients with hormone-independent breast cancers seem to benefit from these immune-based approaches. The present review examines and discusses the current literature related to the role of hormone receptor signaling (specifically, an estrogen receptor) and the impact of its modulation on the sensitivity of breast cancer cells to the effector mechanisms of anti-tumor immune responses and on the capability of breast cancers to escape from protective anti-cancer immunity. Future research prospects related to the possibility of promoting the efficacy of immune-based interventions using hormone therapy agents are considered.

MED12 mutation activates the tryptophan/kynurenine/AHR pathway to promote growth of uterine leiomyomas.

Uterine leiomyomas cause heavy menstrual bleeding, anemia, and pregnancy loss in millions of women worldwide. Driver mutations in the transcriptional mediator complex subunit 12 (MED12) gene in uterine myometrial cells initiate 70% of leiomyomas that grow in a progesterone-dependent manner. We showed a distinct chromatin occupancy landscape of MED12 in mutant MED12 (mut-MED12) versus WT-MED12 leiomyomas. Integration of cistromic and transcriptomics data identified tryptophan 2,3-dioxygenase (TDO2) as the top mut-MED12 target gene that was significantly upregulated in mut-MED12 leiomyomas when compared with adjacent myometrium and WT-MED12 leiomyomas. TDO2 catalyzes the conversion of tryptophan to kynurenine, an aryl hydrocarbon receptor (AHR) ligand that we confirmed to be significantly elevated in mut-MED12 leiomyomas. Treatment of primary mut-MED12 leiomyoma cells with tryptophan or kynurenine stimulated AHR nuclear translocation, increased proliferation, inhibited apoptosis, and induced AHR-target gene expression, whereas blocking the TDO2/kynurenine/AHR pathway by siRNA or pharmacological treatment abolished these effects. Progesterone receptors regulated the expression of AHR and its target genes. In vivo, TDO2 expression positively correlated with the expression of genes crucial for leiomyoma growth. In summary, activation of the TDO2/kynurenine/AHR pathway selectively in mut-MED12 leiomyomas promoted tumor growth and may inform the future development of targeted treatments and precision medicine.

Systemic Inflammation and Normocytic Anemia in DOCK11 Deficiency.

BACKGROUND: Increasing evidence links genetic defects affecting actin-regulatory proteins to diseases with severe autoimmunity and autoinflammation, yet the underlying molecular mechanisms are poorly understood. Dedicator of cytokinesis 11 (DOCK11) activates the small Rho guanosine triphosphatase (GTPase) cell division cycle 42 (CDC42), a central regulator of actin cytoskeleton dynamics. The role of DOCK11 in human immune-cell function and disease remains unknown. METHODS: We conducted genetic, immunologic, and molecular assays in four patients from four unrelated families who presented with infections, early-onset severe immune dysregulation, normocytic anemia of variable severity associated with anisopoikilocytosis, and developmental delay. Functional assays were performed in patient-derived cells, as well as in mouse and zebrafish models. RESULTS: We identified rare, X-linked germline mutations in DOCK11 in the patients, leading to a loss of protein expression in two patients and impaired CDC42 activation in all four patients. Patient-derived T cells did not form filopodia and showed abnormal migration. In addition, the patient-derived T cells, as well as the T cells from Dock11-knockout mice, showed overt activation and production of proinflammatory cytokines that were associated with an increased degree of nuclear translocation of nuclear factor of activated T cell 1 (NFATc1). Anemia and aberrant erythrocyte morphologic features were recapitulated in a newly generated dock11-knockout zebrafish model, and anemia was amenable to rescue on ectopic expression of constitutively active CDC42. CONCLUSIONS: Germline hemizygous loss-of-function mutations affecting the actin regulator DOCK11 were shown to cause a previously unknown inborn error of hematopoiesis and immunity characterized by severe immune dysregulation and systemic inflammation, recurrent infections, and anemia. (Funded by the European Research Council and others.).

Artificial intelligence to empower diagnosis of myelodysplastic syndromes by multiparametric flow cytometry.

The diagnosis of myelodysplastic syndromes (MDS) might be challenging and relies on the convergence of cytological, cytogenetic, and molecular factors. Multiparametric flow cytometry (MFC) helps diagnose MDS, especially when other features do not contribute to the decision-making process, but its usefulness remains underestimated, mostly due to a lack of standardization of cytometers. We present here an innovative model integrating artificial intelligence (AI) with MFC to improve the diagnosis and the classification of MDS. We develop a machine learning model through an elasticnet algorithm directed on a cohort of 191 patients, only based on flow cytometry parameters selected by the Boruta algorithm, to build a simple but reliable prediction score with five parameters. Our AI-assisted MDS prediction score greatly improves the sensitivity of the Ogata score while keeping an excellent specificity validated on an external cohort of 89 patients with an Area Under the Curve of 0.935. This model allows the diagnosis of both high- and low-risk MDS with 91.8% sensitivity and 92.5% specificity. Interestingly, it highlights a progressive evolution of the score from clonal hematopoiesis of indeterminate potential (CHIP) to highrisk MDS, suggesting a linear evolution between these different stages. By significantly decreasing the overall misclassification of 52% for patients with MDS and of 31.3% for those without MDS (P=0.02), our AI-assisted prediction score outperforms the Ogata score and positions itself as a reliable tool to help diagnose MDS.

HDAC6 regulates human erythroid differentiation through modulation of JAK2 signalling.

Among histone deacetylases, HDAC6 is unusual in its cytoplasmic localization. Its inhibition leads to hyperacetylation of non-histone proteins, inhibiting cell cycle, proliferation and apoptosis. Ricolinostat (ACY-1215) is a selective inhibitor of the histone deacetylase HDAC6 with proven efficacy in the treatment of malignant diseases, but anaemia is one of the most frequent side effects. We investigated here the underlying mechanisms of this erythroid toxicity. We first confirmed that HDAC6 was strongly expressed at both RNA and protein levels in CD34+ -cells-derived erythroid progenitors. ACY-1215 exposure on CD34+ -cells driven in vitro towards the erythroid lineage led to a decreased cell count, an increased apoptotic rate and a delayed erythroid differentiation with accumulation of weakly hemoglobinized immature erythroblasts. This was accompanied by drastic changes in the transcriptomic profile of primary cells as shown by RNAseq. In erythroid cells, ACY-1215 and shRNA-mediated HDAC6 knockdown inhibited the EPO-dependent JAK2 phosphorylation. Using acetylome, we identified 14-3-3ζ, known to interact directly with the JAK2 negative regulator LNK, as a potential HDAC6 target in erythroid cells. We confirmed that 14-3-3ζ was hyperacetylated after ACY-1215 exposure, which decreased the 14-3-3ζ/LNK interaction while increased LNK ability to interact with JAK2. Thus, in addition to its previously described role in the enucleation of mouse fetal liver erythroblasts, we identified here a new mechanism of HDAC6-dependent control of erythropoiesis through 14-3-3ζ acetylation level, LNK availability and finally JAK2 activation in response to EPO, which is crucial downstream of EPO-R activation for human erythroid cell survival, proliferation and differentiation.

Targeting Class I Histone Deacetylases in Human Uterine Leiomyosarcoma.

Uterine leiomyosarcoma (uLMS) is the most frequent subtype of uterine sarcoma that presents a poor prognosis, high rates of recurrence, and metastasis. Currently, the molecular mechanism of the origin and development of uLMS is unknown. Class I histone deacetylases (including HDAC1, 2, 3, and 8) are one of the major classes of the HDAC family and catalyze the removal of acetyl groups from lysine residues in histones and cellular proteins. Class I HDACs exhibit distinct cellular and subcellular expression patterns and are involved in many biological processes and diseases through diverse signaling pathways. However, the link between class I HDACs and uLMS is still being determined. In this study, we assessed the expression panel of Class I HDACs in uLMS and characterized the role and mechanism of class I HDACs in the pathogenesis of uLMS. Immunohistochemistry analysis revealed that HDAC1, 2, and 3 are aberrantly upregulated in uLMS tissues compared to adjacent myometrium. Immunoblot analysis demonstrated that the expression levels of HDAC 1, 2, and 3 exhibited a graded increase from normal and benign to malignant uterine tumor cells. Furthermore, inhibition of HDACs with Class I HDACs inhibitor (Tucidinostat) decreased the uLMS proliferation in a dose-dependent manner. Notably, gene set enrichment analysis of differentially expressed genes (DEGs) revealed that inhibition of HDACs with Tucidinostat altered several critical pathways. Moreover, multiple epigenetic analyses suggested that Tucidinostat may alter the transcriptome via reprogramming the oncogenic epigenome and inducing the changes in microRNA-target interaction in uLMS cells. In the parallel study, we also determined the effect of DL-sulforaphane on the uLMS. Our study demonstrated the relevance of class I HDACs proteins in the pathogenesis of malignant uLMS. Further understanding the role and mechanism of HDACs in uLMS may provide a promising and novel strategy for treating patients with this aggressive uterine cancer.

Therapy-related Myeloid Neoplasms Following PARP Inhibitors: Real-life Experience.

PURPOSE: To provide insights into the diagnosis and management of therapy-related myeloid neoplasms (t-MN) following PARP inhibitors (PARPi). EXPERIMENTAL DESIGN: In a French cancer center, we identified and described the profiles of 13 t-MN diagnosed among 37 patients with ovarian cancer referred to hematology consultation for cytopenia under PARPi. Next, we described these 13 t-MN post-PARPi among 37 t-MN post ovarian cancer according to PARPi exposure. Finally, we described 69 t-MN post-PARPi in a national cohort. RESULTS: From 2016 to 2021, cumulative incidence of t-MN was 3.5% (13/373) among patients with ovarian cancer treated with PARPi. At time of hematologic consultation, patients with t-MN had a longer PARPi exposure (9 vs. 3 months, P = 0.01), lower platelet count (74 vs. 173 G/L, P = 0.0005), and more cytopenias (2 vs. 1, P = 0.0005). Compared with t-MN not exposed to PARPi, patients with t-MN-PARPi had more BRCA1/2 germline mutation (61.5% vs. 0%, P = 0.03) but similar overall survival (OS). In the national cohort, most t-MN post-PARPi had a complex karyotype (61%) associated with a high rate of TP53 mutation (71%). Median OS was 9.6 months (interquartile range, 4-14.6). In multivariate analysis, a longer time between end of PARPi and t-MN (HR, 1.046; P = 0.02), olaparib compared with other PARPi (HR, 5.82; P = 0.003) and acute myeloid leukemia (HR, 2.485; P = 0.01) were associated with shorter OS. CONCLUSIONS: In a large series, we described a high incidence of t-MN post-PARPi associated with unfavorable cytogenetic and molecular abnormalities leading to poor OS. Early detection is crucial, particularly in cases of delayed cytopenia.

CDK8 and CDK19 regulate intestinal differentiation and homeostasis via the chromatin remodeling complex SWI/SNF.

Initiation and maintenance of transcriptional states are critical for controlling normal tissue homeostasis and differentiation. The cyclin dependent kinases CDK8 and CDK19 (Mediator kinases) are regulatory components of Mediator, a highly conserved complex that orchestrates enhancer-mediated transcriptional output. While Mediator kinases have been implicated in the transcription of genes necessary for development and growth, its function in mammals has not been well defined. Using genetically defined models and pharmacological inhibitors, we showed that CDK8 and CDK19 function in a redundant manner to regulate intestinal lineage specification in humans and mice. The Mediator kinase module bound and phosphorylated key components of the chromatin remodeling complex switch/sucrose non-fermentable (SWI/SNF) in intestinal epithelial cells. Concomitantly, SWI/SNF and MED12-Mediator colocalized at distinct lineage-specifying enhancers in a CDK8/19-dependent manner. Thus, these studies reveal a transcriptional mechanism of intestinal cell specification, coordinated by the interaction between the chromatin remodeling complex SWI/SNF and Mediator kinase.

Combining imaging flow cytometry and machine learning for high-throughput schistocyte quantification: A SVM classifier development and external validation cohort.

BACKGROUND: Schistocyte counts are a cornerstone of the diagnosis of thrombotic microangiopathy syndrome (TMA). Their manual quantification is complex and alternative automated methods suffer from pitfalls that limit their use. We report a method combining imaging flow cytometry (IFC) and artificial intelligence for the direct label-free and operator-independent quantification of schistocytes in whole blood. METHODS: We used 135,045 IFC images from blood acquisition among 14 patients to extract 188 features with IDEAS® software and 128 features from a convolutional neural network (CNN) with Keras framework in order to train a support vector machine (SVM) blood elements' classifier used for schistocytes quantification. FINDING: Keras features showed better accuracy (94.03%, CI: 93.75-94.31%) than ideas features (91.54%, CI: 91.21-91.87%) in recognising whole-blood elements, and together they showed the best accuracy (95.64%, CI: 95.39-95.88%). We obtained an excellent correlation (0.93, CI: 0.90-0.96) between three haematologists and our method on a cohort of 102 patient samples. All patients with schistocytosis (>1% schistocytes) were detected with excellent specificity (91.3%, CI: 82.0-96.7%) and sensitivity (100%, CI: 89.4-100.0%). We confirmed these results with a similar specificity (91.1%, CI: 78.8-97.5%) and sensitivity (100%, CI: 88.1-100.0%) on a validation cohort (n=74) analysed in an independent healthcare centre. Simultaneous analysis of 16 samples in both study centres showed a very good correlation between the 2 imaging flow cytometers (Y=1.001x). INTERPRETATION: We demonstrate that IFC can represent a reliable tool for operator-independent schistocyte quantification with no pre-analytical processing which is of most importance in emergency situations such as TMA. FUNDING: None.

The Functional Role and Regulatory Mechanism of Bromodomain-Containing Protein 9 in Human Uterine Leiomyosarcoma.

Uterine leiomyosarcoma (uLMS) is the most common type of uterine sarcoma associated with poor prognosis, high rates of recurrence, and metastasis. There is currently limited information about uLMS molecular mechanisms of origin and development. Bromodomain (BRD)-containing proteins are involved in many biological processes, most notably epigenetic regulation of transcription, and BRD protein dysfunction has been linked to many diseases including tumorigenesis. However, the role of BRD proteins in the pathogenesis of uLMS is unknown. Here, we show for the first time that BRD9 is aberrantly overexpressed in uLMS tissues compared to adjacent myometrium. BRD9 expression is also upregulated in uLMS cell lines compared to benign uterine fibroid and myometrium cell lines. Inhibition of BRD9 using the specific inhibitor (TP-472) suppressed uLMS cell proliferation via inducing apoptosis and cell cycle arrest. To further characterize the mechanistic basis for TP-472 inhibition of uLMS cell growth, we performed a comparative RNA-seq analysis of vehicle-treated and TP-472-treated uLMS cells (n = 4 each). Bioinformatics analysis revealed that TP-472 treatment distinctly altered the uLMS cell transcriptome. Gene set enrichment analysis identified critical pathways altered by BRD9 inhibition, including interferon-alpha response, KRAS signaling, MYC targets, TNF-a signaling via NFkB, and MTORC1 signaling. Parsimonious gene correlation network analysis identified nine enriched modules, including cell cycle and apoptosis modules. Moreover, the ENCODE Histone Modifications gene set and TargetScan microRNA analysis in Enrichr suggested that TP-472-induced BRD9 inhibition may alter the uLMS cell transcriptome by reprograming the oncogenic epigenome and inducing miRNA-mediated gene regulation. Therefore, BRD9 constitutes a specific vulnerability in malignant uLMS, and targeting non-BET BRD proteins in uLMS may provide a promising and novel strategy for treating patients with this aggressive uterine cancer.

Aberrant R-loop-induced replication stress in MED12-mutant uterine fibroids.

Uterine fibroid (UF) driver mutations in Mediator complex subunit 12 (MED12) trigger genomic instability and tumor development through unknown mechanisms. Herein, we show that MED12 mutations trigger aberrant R-loop-induced replication stress, suggesting a possible route to genomic instability and a novel therapeutic vulnerability in this dominant UF subclass. Immunohistochemical analyses of patient-matched tissue samples revealed that MED12 mutation-positive UFs, compared to MED12 mutation-negative UFs and myometrium, exhibited significantly higher levels of R-loops and activated markers of Ataxia Telangiectasia and Rad3-related (ATR) kinase-dependent replication stress signaling in situ. Single molecule DNA fiber analysis revealed that primary cells from MED12 mutation-positive UFs, compared to those from patient-matched MED12 mutation-negative UFs and myometrium, exhibited defects in replication fork dynamics, including reduced fork speeds, increased and decreased numbers of stalled and restarted forks, respectively, and increased asymmetrical bidirectional forks. Notably, these phenotypes were recapitulated and functionally linked in cultured uterine smooth muscle cells following chemical inhibition of Mediator-associated CDK8/19 kinase activity that is known to be disrupted by UF driver mutations in MED12. Thus, Mediator kinase inhibition triggered enhanced R-loop formation and replication stress leading to an S-phase cell cycle delay, phenotypes that were rescued by overexpression of the R-loop resolving enzyme RNaseH. Altogether, these findings reveal MED12-mutant UFs to be uniquely characterized by aberrant R-loop induced replication stress, suggesting a possible basis for genomic instability and new avenues for therapeutic intervention that involve the replication stress phenotype in this dominant UF subtype.

Beyond Immunosuppression: The Multifaceted Functions of Tumor-Promoting Myeloid Cells in Breast Cancers.

Breast cancers are commonly associated with an immunosuppressive microenvironment responsible for tumor escape from anti-cancer immunity. Cells of the myeloid lineage account for a major part of this tumor-promoting landscape. These myeloid cells are composed of heterogeneous subsets at different stages of differentiation and have traditionally been described by their cardinal ability to suppress innate and adaptive anticancer immunity. However, evidence has accumulated that, beyond their immunosuppressive properties, breast cancer-induced myeloid cells are also equipped with a broad array of "non-immunological" tumor-promoting functions. They therefore represent major impediments for anticancer therapies, particularly for immune-based interventions. We herein analyze and discuss current literature related to the versatile properties of the different myeloid cell subsets engaged in breast cancer development. We critically assess persisting difficulties and challenges in unequivocally discriminate dedicated subsets, which has so far prevented both the selective targeting of these immunosuppressive cells and their use as potential biomarkers. In this context, we propose the concept of IMCGL, "pro-tumoral immunosuppressive myeloid cells of the granulocytic lineage", to more accurately reflect the contentious nature and origin of granulocytic cells in the breast tumor microenvironment. Future research prospects related to the role of this myeloid landscape in breast cancer are further considered.

Influence of the Metabolism on Myeloid Cell Functions in Cancers: Clinical Perspectives.

Tumor metabolism plays a crucial role in sustaining tumorigenesis. There have been increasing reports regarding the role of tumor metabolism in the control of immune cell functions, generating a potent immunosuppressive contexture that can lead to immune escape. The metabolic reprogramming of tumor cells and the immune escape are two major hallmarks of cancer, with several instances of crosstalk between them. In this paper, we review the effects of tumor metabolism on immune cells, focusing on myeloid cells due to their important role in tumorigenesis and immunosuppression from the early stages of the disease. We also discuss ways to target this specific crosstalk in cancer patients.

Comprehensive Review of Uterine Fibroids: Developmental Origin, Pathogenesis, and Treatment.

Uterine fibroids are benign monoclonal neoplasms of the myometrium, representing the most common tumors in women worldwide. To date, no long-term or noninvasive treatment option exists for hormone-dependent uterine fibroids, due to the limited knowledge about the molecular mechanisms underlying the initiation and development of uterine fibroids. This paper comprehensively summarizes the recent research advances on uterine fibroids, focusing on risk factors, development origin, pathogenetic mechanisms, and treatment options. Additionally, we describe the current treatment interventions for uterine fibroids. Finally, future perspectives on uterine fibroids studies are summarized. Deeper mechanistic insights into tumor etiology and the complexity of uterine fibroids can contribute to the progress of newer targeted therapies.

Accurate classification of plasma cell dyscrasias is achieved by combining artificial intelligence and flow cytometry.

Monoclonal gammopathy of unknown significance (MGUS), smouldering multiple myeloma (SMM), and multiple myeloma (MM) are very common neoplasms. However, it is often difficult to distinguish between these entities. In the present study, we aimed to classify the most powerful markers that could improve diagnosis by multiparametric flow cytometry (MFC). The present study included 348 patients based on two independent cohorts. We first assessed how representative the data were in the discovery cohort (123 MM, 97 MGUS) and then analysed their respective plasma cell (PC) phenotype in order to obtain a set of correlations with a hypersphere visualisation. Cluster of differentiation (CD)27 and CD38 were differentially expressed in MGUS and MM (P < 0·001). We found by a gradient boosting machine method that the percentage of abnormal PCs and the ratio PC/CD117 positive precursors were the most influential parameters at diagnosis to distinguish MGUS and MM. Finally, we designed a decisional algorithm allowing a predictive classification ≥95% when PC dyscrasias were suspected, without any misclassification between MGUS and SMM. We validated this algorithm in an independent cohort of PC dyscrasias (n = 87 MM, n = 41 MGUS). This artificial intelligence model is freely available online as a diagnostic tool application website for all MFC centers worldwide (https://aihematology.shinyapps.io/PCdyscrasiasToolDg/).