Success of chimeric antigen receptor (CAR) T-cell therapy in lymphoid malignancies has not yet been recapitulated in acute myeloid leukemia (AML). We developed CAR T-cells targeting CD371 with a mutated CD28 costimulatory domain to limit T-cell exhaustion, and constitutive interleukin-18 secretion to enhance immune function (CD371/SAVVY/IL-18 CAR). We initiated a phase I trial (NCT06017258), successfully manufactured and administered CD371/SAVVY/IL-18 CAR T-cells in 5 patients with relapsed/refractory AML and observed expansion following a single infusion of 3x104 or 3x105 CAR T-cells/kg; three patients refractory to ≥5 lines of therapy and post-allogeneic transplant exhibited AML clearance and no evidence of graft-versus-host disease. Dose-limiting toxicity in the two patients treated with 3x105 CAR T-cells/kg dose (prolonged cytopenias with marrow hypoplasia; severe cytokine release syndrome) led to dose reduction to 3x104 CAR T-cells/kg in the following three patients. Single-cell analyses revealed that circulating CAR T-cells in responders included predominantly cytotoxic CD8+ effector T-cells 2 weeks post-infusion while co-existing NK-cells expressed markers of activation. This pilot study highlights the activity of low-dose IL-18 "armored" CAR T-cells against refractory AML and their potential to promote CAR T-cell cytotoxicity and innate endogenous anti-tumor immunity. NCT06017258.

Congenital thrombotic thrombocytopenic purpura (cTTP) is an ultra-rare thrombotic microangiopathy mediated through inherited deficiency in ADAMTS13 (a disintegrin and metalloprotease with a thrombospondin type 1 motif, member 13). To date more than 200 ADAMTS13 genetic variants have been associated with cTTP. We report longitudinal follow up from the UK TTP registry in 104 confirmed cTTP cases (91 consented for follow-up) in a large multi-ethnic national cTTP patient cohort, including a large Black African cTTP cohort. 71 ADAMTS13 variants were identified, with N-terminal variants associated with earlier age at presentation. During the follow-up period (median 63 months (range 1-179), 80.2% of patients received regular (plasma derived) prophylaxis which reduced end-organ damage, including stroke/TIA (19.0% to 1.5%) and renal impairment during follow-up. Post presentation acute TTP episodes were reduced with prophylaxis (0.68 vs 0.06 acute episodes/follow-up year) Despite regular prophylaxis, symptom control remained apparent on plasma-derived therapy (including headache 42.6%, depression/anxiety 13.2%, fatigue 16.2% and abdominal pain 13.2%). The majority of cTTP patients in the UK have now switched to recombinant ADAMTS13 (n=43, 58.9%), owing to inadequate symptom control (53.5%), plasma reactions (30.2%), or subclinical disease activity (16.3%). This work shows the breadth of ADAMTS13 genetic variants in cTTP, and demonstrates efficacy of regular prophylaxis in (i) reducing acute TTP episodes, and (ii) preventing end organ damage, but despite advances, cTTP related symptoms and the use of blood products, remained problematic.

Venetoclax-obinutuzumab (Ven-Obi) is a standard first-line therapy for chronic lymphocytic leukemia (CLL). The impact of age, fitness, and dose reductions remains unclear. We analyzed patients treated with Ven-Obi in the CLL13 and CLL14 trials, excluding patients with TP53 aberrations. Fitness was assessed using the cumulative illness rating scale (CIRS, >6) and creatinine clearance (≤70 ml/min). Among 410 patients (median age 67), 55.7% were unfit (median age 72) and 44.3% fit (58). Overall response rate (ORR) was 89.5% in unfit and 96.1% in fit patients. Undetectable minimal residual disease (uMRD) <10-4 rates were 80.3% in unfit and 85.1% in fit patients. Progression-free survival (PFS) at 3 years was 86.4% vs 87.5% (HR 1.12, 95%-CI 0.70-1.81, p=0.63). Overall survival at 3 years was 91.8% vs 96.9% (HR 2.02, 95%-CI 0.90-4.55, p=0.088). Adverse events included neutropenia (62.7% unfit, 56.9% fit), infusion-related-reactions (44.3% unfit, 56.9% fit), fatigue (15.8% unfit, 35.9% fit) and infections (57.5% unfit, 69.6% fit). Venetoclax dose reductions <80% occurred in 39.6% of unfit and 17.6% of fit patients, with lower ORR (83.3% vs 98.2%) and uMRD rates (74.2% vs 87.9%) in those with reduced dose intensities, but similar PFS. Dose reductions <70% were associated with shorter PFS. Overall, this study shows comparable efficacy and toxicity of Ven-Obi in fit and unfit patients with CLL.

Bispecific T cell engagers (TCE) targeting BCMA and CD3 induce deep hematologic responses in approximately 60% of heavily pre-treated multiple myeloma (MM) patients. We and others found that high tumor burden leads to resistance to TCE and novel strategies are urgently needed to improve responses in this setting. Ikaros-degraders, including IMiDs and CELMoDs, represent logical partners for TCE due to their direct anti-MM effects and additional immune stimulatory activity; however, it is unclear how to optimally combine them with TCE. Taking advantage of the immunocompetent IMiD-sensitive Vk*MYChCRBN murine model of MM, we optimized strategies to overcome primary resistance to BCMA-TCE and achieve sustained remission, while maintaining a manageable safety profile. The addition of anti-PD1 and pomalidomide reduced the T cell exhaustion that occurs in response to TCE in high tumor burden settings. This allowed for a higher degree of T cell activation and significant improvement in response rates but also increased risk of lethal cytokine release syndrome (CRS). To moderate the response and prevent CRS, we evaluated Ikaros-degraders and dexamethasone with step-up dosed TCE. Pre-treatment with iberdomide and dexamethasone reshaped the bone marrow T cell compartment, promoted infiltration of naïve T cells, and generated 100% response rates and the longest survival in subjects with high tumor burden. This was accompanied by more favorable T cell profiling, with limited expansion of regulatory T cells and exhaustion. In total, administering TCE following dexamethasone and iberdomide treatments provided deeper and more durable responses with reduced risk of CRS.

Obesity is a major health issue and a risk factor for venous thromboembolic disease (VTE). Plasminogen activator inhibitor 1 (PAI-1), encoded by the gene SERPINE1, is a negative regulator of fibrinolysis and has been associated with obesity. The liver, which senses obesity-induced metabolic stress, is a key determinant of circulating PAI-1 levels. However, the mechanisms underlying the increased PAI-1 expression in obesity are unclear. This study investigated the upstream regulation of PAI-1 and its role in fibrinolysis and deep vein thrombosis (DVT). Compared with lean mice, diet-induced obesity (DIO) mice presented significantly shorter fibrinolysis times and larger venous thrombi, due largely to increased hepatocyte expression of PAI-1. A public single-cell RNA-seq dataset from the livers of obese individuals suggested that increased PAI-1 expression may be related to reduced hepatocyte FXR signaling. FXR activation also suppressed Serpine1 mRNA and PAI-1 protein expression levels in both mice and primary mouse hepatocytes (MPHs), but a decrease in PAI-1 in MPHs of Fxr-null mice after FXR activation was not observed. Both Fxr-null mice and Fxrfl/fl mice with AAV8-TBG-Cre exhibited significantly elevated plasma PAI-1, resulting in further impaired fibrinolysis and increased DVT burden. Dual-luciferase reporter assays and chromatin immunoprecipitation (Ch-IP) suggested that FXR activation directly represses Serpine1 transcription. Importantly, tropifexor treatment of obese mice lowered plasma PAI-1 levels and further alleviated fibrinolysis and the DVT load. These findings suggest that targeting FXR in hepatocytes may improve fibrinolysis and reduce DVT risk.

Acute myeloid leukemia (AML) is characterized by a low five-year survival rate. Despite having many clinical metrics to assess patient prognosis, there remain opportunities to improve risk stratification. We hypothesized that an underexplored resource to examine AML patient prognosis is the plasma metabolome. Circulating metabolites are influenced by patients' clinical status and can serve as accessible cancer biomarkers. To establish a resource of circulating metabolites in genetically diverse AML patients, we performed an unbiased metabolomic and lipidomic analysis of 231 diagnostic AML plasma samples prior to treatment with intensive chemotherapy. Intriguingly, circulating metabolites were highly associated with the mutation status within the AML cells. Further, lipids were associated with refractory status. We established a machine learning algorithm trained on chemo-refractory associated lipids to predict patient survival. Cox regression and Kaplan-Meier analysis demonstrated that the high-risk lipid signature predicted overall survival in this patient cohort. Impressively, the top lipid in the high-risk lipid signature, sphingomyelin (d44:1), was sufficient to predict overall survival in the original and an independent validation dataset. Overall, this research underscores the potential of circulating metabolites to capture AML heterogeneity and lipids to be used as potential AML biomarkers.

The discovery of calreticulin (CALR) mutations in patients with myeloproliferative neoplasms (MPN) has paved the way for the elucidation of a unique disease mechanism that is particularly well-suited to targeting by biologics. All MPN-associated pathogenic CALR mutations are characterized by a frameshift, resulting in translation of the same neoantigen peptide. This neoantigen directly activates the thrombopoietin receptor, leading to uncontrolled neoplastic cell proliferation. Current therapeutic approaches for MPN are focused primarily on blood count control. Furthermore, current approaches are neither disease-modifying nor clonally selective. However, as the mutant CALR neoantigen peptide is functional and not expressed in normal physiology, it is an ideal drug target. Here, we review the structure and function of mutant CALR, including the subtle yet clinically and therapeutically relevant differences between the two most commonly occurring types of mutation. We also review the current therapeutic landscape for CALR-mutated MPN, highlighting the areas in which current approaches are inadequate. Finally, we review ongoing clinical and preclinical experimental approaches for targeting mutant CALR in MPN in a clonally selective manner using monoclonal antibodies, bispecific antibodies, cancer vaccination, chimeric antigen receptor T cells, and antibody-drug conjugates. Taken together, we expect that ongoing developments in mutant CALR-targeted therapeutics will lead to promising novel strategies for long-term disease control.

T-cell acute lymphoblastic leukemia (T-ALL) represents a group of aggressive hematological malignancies characterized by unfavorable prognosis, urging the need for innovative therapeutic strategies. LEF1 is a member of the lymphoid enhancer factor/T-cell factor (LEF/TCF) family of DNA-binding transcription factors, known for their interaction with nuclear β-catenin in the context of the Wnt signaling pathway. Although LEF1's implication in colon cancer is well-documented, its clinical relevance and functional consequences remain elusive in T-ALL. In this study, we provide valuable insights into the prevalence and significance of LEF1 alterations in a comprehensive cohort of 474 pediatric and adult T-ALL patients enrolled in the FRALLE 2000 and GRAALL-2003-2005 trials (NCT00222027; NCT00327678) respectively. LEF1 alterations were detected in 63 cases (13%), including 9 point-mutations (14.3%), 18 large deletions (28.6%), and - strikingly - 36 focal deletions (57.1%), which emerge as the most recurrent subtype. LEF1-altered cases were associated with increased central nervous system involvement and improved initial treatment response. Importantly, we unveil the existence of a previously undescribed dominant-negative LEF1 isoform resulting from focal deletions of the exons 2-3. This novel truncated protein, previously unreported in the literature, is associated with the disruption of the Wnt pathway and TCR signaling, which can be exploited as a therapeutic strategy to enhance chemosensitivity in LEF1-deleted T-ALL cases.

Fetal hemoglobin (HbF) reactivation is a promising therapy for β-hemoglobinopathies. We developed a prime-editing strategy that introduces multiple mutations in the fetal γ-globin promoters that are expected to increase their activity. We tested multiple targets and optimized a variety of parameters to achieve ~50% of precise edits in a hematopoietic cell line, with minimal off-targets effects. This work improved our understanding of the complex DNA repair mechanisms involved in prime editing. We tested this strategy in patients' hematopoietic stem/progenitor cells (HSPCs). Although the editing efficiency was variable amongst donors, erythroid clones carrying multiple mutations express a significantly higher γ-globin level compared to cells carrying individual mutations, confirming the potential therapeutic benefit of our combined strategy for patients with β-hemoglobinopathies.

Adoptive transfer of NK cells can induce complete remissions in 30-50% of patients with refractory acute myeloid leukemia and lymphoma. While blood chimerism occurs, attaining functional homing to the site of tumor without exhaustion has been elusive. During chronic infections and tumorigenesis, exposure to activating stimuli weakens the effector activity of NK cells. Despite this knowledge, there is little known about the mechanisms that govern this dysregulation and whether these disparate activating stimuli use distinct pathways to downregulate effector immunity. In this study, we reveal that chronic NK cell activating receptor (NKAR) stimulation and chronic IL-15 exposure impart distinct modes of dysregulation, with NKAR stimulation inducing a tissue resident-like state that resembles that of tumor-infiltrating NK cells in cancer patients. Using loss- and gain-of-function studies, we identify the transcription factor KLF2 as a master regulator of the NK cell response to chronic activation and provide evidence that KLF2 overexpression promotes NK cell cytotoxicity, cytokine production, and chemotaxis, and inhibits the development of dysfunctional, tissue resident-like features. Using KLF2 reporter mice, we show that in certain tissues, tissue resident NK cells are predominantly KLF2-negative while circulating NK cells in these tissues are overwhelmingly KLF2+. Lastly, using mixed bone marrow chimeras, we demonstrate that conditional KLF2 deficiency in NK cells leads to altered homing and the acquisition of tissue resident-like features in vivo. Together, these findings highlight the profound changes NK cells undergo during prolonged activation and advance our understanding of how some NK cell therapies fail during malignant relapse.

Platelets are dynamic cells that perform vital roles both in thrombosis and in immune regulation. Platelets are activated through well described signaling pathways following injury to the vascular wall to prevent excess bleeding and initiate vascular repair. However, platelet functions extend beyond this initial physical phase of platelet activation, through their expression and secretion of angiogenic factors and immune mediators that initiate wound healing and recruit leukocytes to prevent infection. Not only do activated platelets induce inflammation and vascular repair but circulating resting platelets maintain vascular integrity and immune homeostasis. Furthermore, many studies have shown that platelets are not homogenous, demonstrating transcriptomic and proteomic differences that associate with platelet phenotypes and functions in multiple disease states. Platelets are produced by megakaryocytes that are found not only in the bone marrow but also in the lungs and spleen. Megakaryocytes within the bone marrow are heterogeneous, which may contribute to platelet heterogeneity. Additionally, lung megakaryocytes are immune-differentiated compared to those in other tissues and proportionately produce more platelets at times of increased demand, perhaps adding to platelet heterogeneity in disease contexts. In this review, we will discuss underlying mechanisms that may lead to platelet heterogeneity impacting both health maintenance and disease.

Artificial intelligence (AI) and its sub-discipline, machine learning (ML), have the potential to revolutionize healthcare, including hematology. The diagnosis and treatment of hematologic disorders depend on the integration of diverse data sources, such as imaging, pathology, omics, and laboratory parameters. The increasing volume and complexity of patient data have made clinical decision-making more challenging. AI/ML hold significant potential for enhancing diagnostic accuracy, risk stratification, and treatment response prediction through advanced modeling techniques. Generative AI, a recent advancement within the broader field of AI, is poised to have a profound impact on healthcare and hematology. Generative AI can enhance the development of novel therapeutic strategies, improve diagnostic workflows by generating high-fidelity images or pathology reports, and facilitate more personalized approaches to patient management. Its ability to augment clinical decision-making and streamline research represents a significant leap forward in the field. However, despite this potential, few AI/ML tools have been fully implemented in clinical practice due to challenges related to data quality, equity, advanced infrastructure, and the establishment of robust evaluation metrics. Despite its promise, AI implementation in hematology faces critical challenges, including bias, data quality issues, and a lack of regulatory frameworks and safety standards that keep pace with rapid technological advancements. In this review, we provide an overview of the current state of AI/ML in hematology as of 2025, identify existing gaps, and offer insights into future developments.

Desmopressin is often administered to correct factor VIII (FVIII) levels in female hemophilia A carriers (HAC). However, the post-desmopressin FVIII pharmacokinetic profiles have been reported only in small case series in this population. Therefore, this study analyzed the post-desmopressin FVIII and von Willebrand factor response in 361 HAC included at 14 French hemophilia treatment centers. A population pharmacokinetic/pharmacodynamic model was developed to analyze the VWF and FVIII levels by taking into account the F8 gene variants (n=143, 39.6%, with null, and n=218, 60.4%, with non-null variants), demographic and laboratory covariates. The prior/after desmopressin mean basal, peak and recovery FVIII:C levels were 0.34 IU.mL-1 (0.08-0.65), 1.13 IU.mL-1 (0.19-2.69), and 2.85 (1.06-7.13), respectively. Peak FVIII:C was ≥0.5 IU.mL-1 in 95.6% (345/361) and ≥0.8IU.mL-1 in 78.7% (284/361) of patients. The covariate analysis showed a poorer desmopressin FVIII response for null than non-null F8 variants: lower mean peak (1.04 vs 1.23 IU.mL-1, p<5.10-5) and lower percentage of patients with normalized FVIII:C (91.6% vs 98.1%, p=0.0068), higher mean clearance (5898.83 vs 2704.06 IU.h-1, p=1.58.10-15), lower mean AUC0-12h (7.73 vs 9.06 IU.mL-1.h, p<5.10-6), and shorter mean time with FVIII:C ≥0.8 IU.mL-1 (1.9 vs 4.1h, p<6.10-6). HAC with body weight <35 kg had lower peak FVIII:C, higher FVIII clearance, lower AUC0-12h, and shorter time with FVIII:C ≥0.8 IU.mL-1 than HAC with body weight 35-70 kg. The ABO blood type did not influence the response to desmopressin. In HAC, the post-desmopressin FVIII response is strongly influenced by the F8 genotype and body weight.

Transfusion-related acute lung injury (TRALI) is a leading cause of transfusion-related mortality. Though the precise mechanism is not fully understood, a two-hit model is widely accepted, involving both a predisposing patient condition and the transfusion itself. Mitochondrial damage-associated molecular patterns (mtDAMPs), such as mitochondrial DNA (mtDNA) and N-formylated peptides (NFPs), are elevated in trauma patients and stored blood products, and have been implicated in adverse transfusion outcomes, prompting us to investigated whether mtDAMPs could serve as a priming "first-hit" in TRALI. Using a murine model, we found that injection of purified mitochondria followed by a monoclonal anti-major histocompatibility complex class I (MHC class I) antibody (34-1-2s) induced significantly greater lung injury compared to the isotype control. This was evidenced by increased pulmonary edema, elevated plasma macrophage inflammatory protein-2 (MIP-2), enhanced neutrophil lung infiltration, hypothermia, and respiratory distress. Similar effects were observed using a Toll-Like Receptor 9 agonist (ODN 2395), purified mtDNA, and a synthetic NFP (WKYMVm), agonist of Formyl Peptide Receptor (FPR). Notably, a TLR9 antagonist blocked the mtDAMP-induced TRALI response, whereas two FPR antagonists did not, underscoring a key role for mtDNA and TLR9 signaling in disease priming. These findings suggest that mtDAMPs, particularly mtDNA, present in both transfusion products and recipient plasma may predispose patients to antibody-mediated TRALI. Targeting mtDAMPs or their receptors may offer a novel therapeutic strategy to mitigate TRALI risk.

Until recently, the JAK1/2 inhibitor, ruxolitinib (Jakafi), was the only therapy for steroid-refractory acute graft-versus-host disease (SR-aGVHD) approved by the U.S. Food and Drug Administration (FDA) for use in patients older than 12 years. Now the FDA has approved a potent mesenchymal stromal cell product (MSC), remestemcel-L-rknd (Ryoncil), for children 18 and younger, showing 70% response rates and nearly 70% six-month survival. In this Spotlight, we highlight this important advance in the field.