Mitochondrial DNA (mtDNA) mutations are frequently observed in cancer, but their clinical and functional significance in chronic myeloid leukemia (CML) remains incompletely defined. Here, we show that a distinct mtDNA mutational landscape is associated with mitochondrial metabolic programs and response to imatinib therapy in CML. We performed comprehensive profiling of somatic mtDNA mutations in 120 patients with chronic-phase CML. At diagnosis, 241 somatic mtDNA mutations were identified in 92 patients, including 29 homoplasmic mutations. In a clinically annotated cohort of 79 imatinib-treated patients, a higher number of mtDNA mutations (≥3 mutations) and higher variant allele frequency were associated with superior molecular responses, and remained significant in multivariable analyses. mtDNA mutational patterns were associated with distinct metabolic phenotypes in CD34+ leukemic stem/progenitor cells. Suboptimal responders exhibited increased mitochondrial respiration, spare respiratory capacity, mitochondrial content, and enrichment of mitochondrial biogenesis and lipid metabolic programs, consistent with enhanced oxidative phosphorylation dependence. In contrast, favorable responders displayed higher mtDNA mutational burden together with reduced respiratory reserve and increased mitophagy-related programs. Pharmacologic Complex I inhibition reduced clonogenic potential and enhanced imatinib sensitivity. Collectively, these findings identify mtDNA mutational states as a biomarker of metabolic fitness and therapeutic response in CML, while supporting further investigation of mitochondrial metabolism as a potential therapeutic vulnerability in CML.
Neurodegeneration (ND) is a severe complication of Langerhans cell histiocytosis (LCH), yet its underlying biology and reliable biomarkers remain poorly defined. The aim of this study was to (1) gain insight into neuroimmunological mechanisms governing ND and (2) assess the clinical value of established and novel biomarkers for ND-LCH. We applied targeted proteomics and neurofilament light chain (NFL) assays to cerebrospinal fluid (CSF) and plasma from LCH patients with and without ND, with control cohorts. ND-LCH exhibited a distinct CSF proteomic profile characterized by increased decoy receptors, including interleukin-1 receptor type 2 (IL-1RT2), macrophage activation markers, and cytotoxic and immunoregulatory proteins. CSF IL-1RT2 showed higher specificity for ND-LCH than CSF NFL (0.99 [0.99-1.0] vs. 0.88 [0.80-0.96]) and tracked longitudinal disease changes. Plasma analyses revealed a complementary immune signature in which NFL correlated with CSF markers, including IL-1RT2, and distinguished ND-LCH from controls. In independent plasma NFL cohorts from France and Japan, elevated levels were confirmed in adults with histiocytosis-associated ND. These findings define a neuroinflammatory signature of ND-LCH and identify IL-1RT2 as a novel candidate biomarker for diagnosis and monitoring.
Relapsed/refractory (R/R) B-cell lymphomas (B-NHL) remain incurable, with limited predictive biomarkers to guide immunotherapy. Preclinical platforms that faithfully preserve the immune context and the viability of patient samples are critically needed for relevant high-throughput drug testing. Here, we present a 3D ex vivo platform for immunotherapy screening that generates patient-derived lymphoma spheroids (PDLS) a scaffold- and matrix-free 3D model derived from peripheral blood or lymph node biopsies of patients, including R/R follicular lymphoma, transformed follicular lymphoma, or diffuse large B-cell lymphoma. PDLS maintain the cellular composition, T-cell activation, and immune escape profiles of the original tumors, supporting multiplexed screening of single-agent and combination therapies within clinically relevant time frames. This scalable model recapitulates immune-tumor cell interactions, provides rapid assessment of immunotherapy efficacy, and bridges preclinical research with clinical application in aggressive B-cell lymphomas. By integrating spatial profiling and single-cell RNA sequencing, PDLS can further evolve into patient-specific 3D tumor avatars that support precision oncology approaches in aggressive B-NHL.
Prediction of the outcome of large B-cell lymphomas/high-grade B-cell lymphomas (DLBCLs/HGBCLs) is based on clinical parameters and molecular testing, for example, for rearrangements of MYC (MYC-R) and MYC-R in combination with BCL2 and BCL6 translocations (double/triple hit). However, the group of DLBCL/HGBCL with poor outcome is not confined to MYC-R lymphomas, and fluorescence in situ hybridization (FISH) testing for MYC-R misses several high-risk lymphomas. We aimed to understand if artificial intelligence (AI) trained to identify MYC-R will delineate a poor prognostic subgroup. We generated a collection of digital hematoxylin and eosin (H&E)-stained slides of DLBCL/HGBCL (N = 2018) annotated for MYC, BCL2, and BCL6 translocations. A multiple-instance deep learning AI model for the identification of MYC-R alone or as double/triple hit was established using 1035 H&E-stained slides and evaluated on an external test cohort (N = 499). A pretrained tumor tissue classifier improved reliability and interpretability by focusing the model on tumor areas. Our model score reflects a morphological "MYCness" in DLBCL/HGBCL and demonstrates a strong association with overall survival (OS) and progression-free survival (PFS) in the external test cohort. This was confirmed with an additional clinical test cohort (N = 484) without FISH labels. The AI model scores correlate with various molecular features of DLBCL/HGBCL, including BCL2 and MYC gene expression, and the high-grade gene expression signature, but also features of the tumor microenvironment. Multivariate analysis, adjusted for International Prognostic Index (IPI) factors, demonstrated the prognostic significance of our model in identifying high-risk cases for both PFS and OS.
Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is a curative treatment option for a significant proportion of patients with acute myeloid leukemia (AML), and it is generally recommended when the relapse risk without allo-HSCT outweighs the estimated non-relapse mortality significantly. While current recommendations for allo-HSCT are based on risk groups, there is considerable heterogeneity within these individual categories. We analyzed 2550 intensively treated AML patients aged 18-70 with cytogenetic and next-generation sequencing data from the HARMONY Alliance database, who did not receive allo-HSCT in first complete remission (CR1). A non-parametric machine learning (ML) model based on Bayesian Additive Regression Trees (BART) integrated clinical variables and genomic aberrations to provide individualized outcome estimations. External validation was performed in a cohort of 714 patients enrolled in UK-NCRI trials. The predictive performance of the HARMONY ML model, measured by the area under the time-dependent receiver operating curve (AUC(t)), was superior to European LeukemiaNet (ELN)2022 risk classification in estimating 5-year overall survival (0.741 vs. 0.700), 5-year relapse-free survival (0.752 vs. 0.705), and 5-year cumulative incidence of relapse (0.742 vs. 0.708), which was confirmed in the external validation cohort. Notably, the model revealed substantial heterogeneity within ELN2022 risk groups, identifying a significant proportion of favorable-risk patients with a predicted 5-year CIR > 40%, who could potentially benefit from allo-HSCT in CR1. The HARMONY ML model provides individualized risk prediction in intensively treated adult AML patients and supports more tailored therapeutic decisions regarding allo-HSCT in CR1, which should be further advanced by integrating measurable residual disease in the future.
Abstract Thrombotic events (TEs) occur in up to 40% of patients with vacuoles, E1 enzyme, X‐linked, autoinflammatory, and somatic (VEXAS) syndrome, but data on its clinical‐genomics features and anticoagulation strategies are limited. To gain more insight into this, we conducted a two‐step study evaluating the prevalence and outcome of TE in VEXAS. First, among 1086 patients followed for TEs, 198 were men aged >40 years with unprovoked thrombosis and no known thrombophilia; 21 also had at least one VEXAS‐compatible feature and underwent UBA1 exon 3 testing. No UBA1 mutation was detected in these 21 patients. Next, we leveraged our Italian VEXAS network, and we accrued 87 molecularly confirmed Italian VEXAS cases (median age 70 years). Any history of TE was documented in 43/87 patients (49%), deep vein thrombosis being the most common (71%). Because follow‐up varied, incident thrombosis was analyzed using a time‐to‐first‐event framework from molecular VEXAS diagnosis, with death without prior TE treated as a competing event. Among 49 patients without prior/concomitant TE, five developed incident post‐diagnosis TE; the 24‐month cumulative incidence was 18.3%. Thrombophilia testing revealed a 15% co‐occurrence, including heterozygous Factor V Leiden, Factor II G20210A, and anti‐cardiolipin antibodies. Treatments comprised direct oral anticoagulants (DOACs) (51%), low molecular weight heparin (LMWH) (28%), Fondaparinux (14%), and vitamin K antagonists (AVKs) (7%). Notably, 27% experienced multiple TEs, of which 22% occurring despite anticoagulation during disease flares. Our findings provide an updated cartography of VEXAS‐related TE, suggesting early screening for thrombophilia in these patients to inform both personalized anticoagulation and disease‐control strategies.
We conducted a prospective, multicenter, Phase II study to evaluate the safety and efficacy of venetoclax/azacitidine/low-dose cytarabine/aclarubicin/granulocyte colony-stimulating factor (G-CSF) (VA-CAG) in young patients with newly diagnosed acute myeloid leukemia (ND-AML). The VA-CAG regimen included venetoclax (100 mg/day, Days 1-2; 200 mg, Day 3; 400 mg/day, Days 4-21), azacitidine (75 mg/m2, Days 1-7), cytarabine (10 mg/m2/12 h, Days 1-7), aclarubicin (12 mg/m2, Days 1, 3, 5, and 7), and G-CSF (5 μg/kg/day, Days 0-8). The primary endpoint was the complete remission (CR) rate after Cycle 1; the secondary endpoints included measurable residual disease (MRD)-negative remission rate, adverse events, and duration of remission (DOR). A total of 120 subjects were enrolled. The median age was 50 years (interquartile range, IQR, 36-57). The CR and composite complete response (CRc) rates were 91% (95% CI 86%-96%) and 95% (95% CI 91%-99%), respectively; 84% (95% CI 81%-88%) of the CRc patients achieved MRD-negative remission. Common Grade ≥4 adverse events included neutropenia (96%), thrombocytopenia (86%), febrile neutropenia (24%), pneumonia (4%), and sepsis (4%). The median times to recovery of absolute neutrophil count (ANC) ≥ 0.5 × 109/L and platelet count ≥ 20 × 109/L for responding patients were 14 days (IQR, 9-19) and 10 days (IQR, 4-16), and the 60-day mortality rate was 0%. With a median follow-up of 22.1 months (95% CI 20.4-23.7 months), 34 subjects subsequently received hematopoietic stem cell transplantation (HSCT). The median DOR values of all patients and the patients who did not receive HSCT were not reached. Hence, the VA-CAG regimen is a safe and effective first-line induction chemotherapy for ND-AML patients.
Due to its high clinical and biological complexity, multiple myeloma (MM) is a malignancy of heterogeneous prognoses and outcomes. The International Myeloma Society (IMS)/International Myeloma Working Group (IMWG) formulated a consensus genomic staging (CGS) of high-risk MM (HRMM), aiming to refine and homogenize the classification of high-risk disease. We evaluated the new HRMM criteria in a cohort of 1209 consecutive newly diagnosed MM patients with complete CGS data (except TP53 mutations), treated in a single center with contemporary regimens, including triplets and quadruplets, between 2010 and 2024. Based on CGS criteria, 25.2% of our cohort's patients were classified as HRMM. High-risk status was associated with significantly inferior outcomes: median overall survival (OS) was 44 months for HRMM versus 93 months for standard-risk (SR) patients (hazard ratio [HR] 1.81; P < 0.001), corresponding to 5-year OS rates of 40% versus 61%. Similarly, median progression-free survival was 21.3 months for HRMM versus 36 months for SR (HR 1.64; P < 0.001). The presence of ≥2 high-risk features identified an ultra-high-risk group (4.4% of patients) with a threefold increased risk of death, compared to SR patients. CGS remained prognostic irrespective of renal dysfunction, stratifying patients with sCr > 1.2 mg/dL (HR 1.71; P < 0.001) and retained prognostic significance across different age groups, transplant eligibility, and treatment eras (pre- and post-2020). The IMS/IMWG CGS effectively stratified patients in a large cohort treated at a tertiary academic center, identifying a substantially high-risk population with poor outcomes despite modern therapies. This validation supports its deployment in clinical practice and research to enhance the precision of risk-adapted management.
First-line (1L) bendamustine plus rituximab (BR) leads to high response rates in follicular lymphoma (FL), but maintaining durable remissions remains challenging. We report the 3-year follow-up from arm 3 of the phase 1b/2 EPCORE NHL-2 trial (NCT04663347) of epcoritamab, a subcutaneously administered CD3×CD20 bispecific antibody, combined with BR in patients with newly diagnosed FL. Twenty-five patients received epcoritamab plus BR, followed by epcoritamab monotherapy for up to 2 years. At a median follow-up of 41.3 months, the best overall response and CR rates were both 96%. The median time to CR was 1.5 (range 1-6) months. At 3 years, 87% of responders maintained CR. High CR rates were observed across subgroups, including 100% of patients with bulky disease (≥7 cm), 93% with Follicular Lymphoma International Prognostic Index score ≥3, and 100% with bone marrow involvement. The three-year progression-free survival and overall survival rates were 83% and 96%, respectively. Three patients progressed within 24 months of initiating treatment. Long-term data were consistent with prior reports and the known safety profiles of the individual agents, with no high-grade cytokine release syndrome or immune effector cell-associated neurotoxicity syndrome events. Infections occurred in 92% of patients; COVID-19 was the most common (84%). Overall, 1L FL treatment with epcoritamab plus BR resulted in deep, durable responses beyond 3 years with a consistent safety profile. These results compare favorably with BR alone, although they require confirmation in further studies, and highlight the versatility of epcoritamab in combination with various standards of care and in improving outcomes in FL.
The rarer p190 (e1a2) transcript in chronic myeloid leukemia (CML) is associated with atypical presentations; yet, its biological basis remains poorly understood. Using a cohort of 60 patients including 42 chronic phase patients age-matched 1:1 with 42 e13a2/e14a2 patients in the chronic phase, we investigated the clinical, genomic, and clonal features of e1a2 BCR::ABL1 CML. We identified 60 e1a2 BCR::ABL1 CML patients showing distinctive hematologic features including lower leukocyte and platelet counts and higher monocytosis (12.3% vs. 2.0%, P < 0.001). Additional somatic mutations were detected in 37/42 (88%) e1a2 BCR::ABL1 cases compared with 7/42 (17%) e13a2/e14a2 BCR::ABL1 cases. The mutational spectrum was dominated by ASXL1 and TET2, and closely resembled CMML-like profiles. Genomic breakpoint sequences of 34 e1a2 BCR::ABL1 cases showed that BCR and ABL1 coordinates were similar to those observed in 394 B-ALL. Longitudinal mutational tracking revealed two distinct clonal architectures. In 71% of patients, mutations disappeared with molecular response, consistent with BCR::ABL1 as the founding event. In contrast, 29% of patients had mutations with stable VAFs, while BCR::ABL1 transcript levels decreased after treatment, indicating that the fusion had been acquired within a pre-existing mutated clone. Single-cell genotyping experiments confirmed these clonal architectures. These patients frequently developed cytopenias under tyrosine kinase inhibitor therapy and half required red blood cell transfusions, reflecting persistence of the ancestral clone rather than BCR::ABL1-driven disease. These findings show that e1a2 BCR::ABL1 CML frequently arises within complex, premutated clonal backgrounds, providing a biological basis for its atypical presentation and heterogeneous treatment response.
