In a groundbreaking advancement that could redefine therapeutic paradigms for acute myeloid leukemia (AML), researchers have leveraged sophisticated single-cell and bulk transcriptomic analyses to unravel the enigmatic role of lactate metabolism and histone lactylation in this aggressive hematological malignancy. The study synthesizes high-resolution molecular data to illuminate how lactate-associated epigenetic remodeling intertwines with immune landscape alterations, ultimately influencing disease progression and patient prognosis.
AML’s notorious heterogeneity poses formidable challenges for clinicians seeking personalized treatment regimens. This study harnessed the power of single-cell RNA sequencing (scRNA-seq), coupled with bulk RNA sequencing techniques, to dissect the cellular and molecular underpinnings linked to lactate metabolism and histone lactylation. These molecular events, collectively referred to as lactate/lactylation-associated genes (LL-genes), encapsulate a network of genes that orchestrate both metabolic pathways and epigenetic modifications critical to tumor biology.
Utilizing the Seurat analytical framework for scRNA-seq data processing, the research team meticulously clustered diverse cell populations within AML specimens, cross-validating their annotations against the Tumor Immune Single Cell Hub 2 (TISCH2) database. This approach ensured precise identification of malignant progenitors exhibiting heightened expression of LL-genes. The application of Gene Set Variation Analysis (GSVA) then enabled quantification of the composite activity of lactate metabolism and lactylation, culminating in what the authors designate as the Lactate Metabolism-Lactylation Modification Combined Activity Score (LML-CAS).
The elevated LML-CAS signature was notably enriched in malignant progenitor cells, signaling a profound metabolic-epigenetic axis that fuels AML pathogenesis. This axis was concomitant with an amplified metabolic-inflammatory synergy, manifesting as immunosuppressive microenvironment features characterized by augmented proportions of regulatory T cells (Tregs) and M2-polarized macrophages. Such immune reprogramming underlines the tumor’s strategy to evade immune surveillance, potentially diminishing response to conventional interventions.
Beyond cellular characterization, the study employed ConsensusClusterPlus clustering algorithms on bulk RNA-seq datasets to stratify AML patients into two distinct molecular subtypes—Clusters A and B. These clusters bore stark differences in survival outcomes, with Cluster A harboring a significantly poorer prognosis. This bifurcation underscores the heterogeneity embedded within AML and the prognostic value embedded in metabolic-epigenetic profiling.
To translate these findings into a clinically actionable framework, the investigators deployed ten complementary machine learning algorithms to construct a robust prognostic model. The optimized seven-gene signature model achieved high accuracy in predicting overall survival and importantly, identified patients less likely to respond favorably to standard chemotherapy regimens. This model portends a shift towards precision oncology, tailoring therapies based on lactate-lactylation molecular signatures.
Transcriptomic dissection further revealed hallmarks of lactylation-associated immunosuppression. Notably, there was sharp downregulation of the CXCL9/10-CXCR3 chemokine axis, integral to T cell trafficking and effector functions, coupled with enrichment of exhaustion markers on T cells. These findings suggest that lactate-driven epigenetic mechanisms may subvert antitumor immunity, complicating therapeutic efforts.
Remarkably, in silico drug sensitivity analyses predicted heightened responsiveness of high-risk patients to BCL-2 and FGFR inhibitors, specifically ABT-737 and AZD4547. This insight paves the way for deploying targeted agents that exploit metabolic vulnerabilities conferred by aberrant lactate and lactylation pathways, potentially overcoming resistance to frontline treatments.
Validation efforts at the molecular level included qRT-PCR assays demonstrating altered expression of pivotal LL-genes, including IFI16, THOC2, HIST1H2BD, and ARPP19, mirroring bioinformatic predictions. Notably, differential expression of IFI16 and THOC2 was further corroborated at the protein level via Western blot analyses in AML patient specimens, reinforcing their functional dysregulation within the disease context.
This integrative multi-omics approach compellingly positions lactate metabolism and histone lactylation as central orchestrators of AML heterogeneity and immune escape. By illuminating the interplay between metabolic reprogramming and epigenetic control, the study opens avenues for novel biomarker development and targeted therapies aimed at modifying the tumor microenvironment.
Moreover, the deployment of machine learning to distill prognostic information from complex molecular datasets exemplifies the future trajectory of oncology research, where computational tools synergize with biological insights to inform clinical decision-making. The seven-gene prognostic model not only stratifies patients by survival risk but also offers predictive value regarding chemotherapeutic efficacy and potential sensitivity to emerging targeted treatments.
The discovery that lactylation influences immune exhaustion and suppression aligns with emerging concepts of metabolic-epigenetic crosstalk in cancer immunity. It highlights potential therapeutic modalities that could reinvigorate antitumor immune responses by modulating lactate-lactylation circuits, supplementing existing immunotherapies.
As AML treatment paradigms continue to evolve, this study’s findings herald a new frontier where metabolic and epigenetic features are leveraged to design precision medicine strategies. The prognostic model and molecular subtype classifications can facilitate risk-adapted therapies, improving outcomes by identifying patients who may benefit from metabolic inhibitors or immunomodulatory approaches.
In conclusion, the integration of single-cell and bulk transcriptomics has yielded transformative insights into the role of lactate and histone lactylation in AML. This knowledge not only deepens our understanding of leukemia biology but also propels the field towards more effective, individualized interventions grounded in metabolic and epigenetic vulnerabilities.
The innovative methodologies and clinical implications presented in this research underscore the potential to reshape AML management. By targeting lactate/lactylation networks, clinicians may soon have new tools to overcome therapeutic resistance and improve survival rates in a disease long challenged by heterogeneity and poor prognosis.
As precision oncology embraces multi-dimensional data integration, studies like this exemplify the power of combining metabolic insights with immune profiling. The promise of exploiting lactate-induced epigenetic modifications could transcend AML, offering strategies applicable to diverse malignancies driven by similar metabolic reprogramming.
Future research will undoubtedly build upon this foundation, refining prognostic models and therapeutic targets, advancing translational efforts to deliver on the promise of tailored treatments that address the metabolic and immunological complexity of leukemia.
Subject of Research: Acute Myeloid Leukemia, Lactate Metabolism, Histone Lactylation, Transcriptomics, Machine Learning Prognostic Models
Article Title: Deciphering lactate/lactylation networks in AML: integrated scRNA-seq and transcriptomics reveal functions and prognostic model
Article References:
Chen, X., Feng, A., Guo, H. et al. Deciphering lactate/lactylation networks in AML: integrated scRNA-seq and transcriptomics reveal functions and prognostic model. BMC Cancer 25, 1647 (2025). https://doi.org/10.1186/s12885-025-14938-8
Image Credits: Scienmag.com
DOI: https://doi.org/10.1186/s12885-025-14938-8
