Abstract:Objective: To evaluate the impact of individualized nutritional intervention by a multidisciplinary team (MDT) based on nutritional risk assessment on short-term (3 months) nutritional status, liver function, and serum liver fibrosis markers in patients with cirrhosis. Methods: A total of 60 patients with cirrhosis admitted to the People's Hospital of Gaoyou Affiliated to Yangzhou University from May 2023 to May 2025 were selected and divided into a control group (conventional nutritional management, n=30) and an intervention group (conventional nutritional management combined with MDT nutritional intervention based on nutritional risk assessment, n=30) using a random number table. Both groups of patients received appropriate nutritional intervention on day 1 of admission. To reduce the impact of fluid fluctuations and acute stress on laboratory indicators in the early stages of admission, blood samples taken in the morning of day 2 of admission and physical measurements were used as the baseline (T0). Follow-up was conducted at 1 month (T1) and 3 months (T2) after discharge. The observed indicators included nutritional status [body mass index (BMI), hemoglobin (Hb), serum albumin (ALB), prealbumin (PA)], liver function [alanine aminotransferase (ALT), prothrombin time (PT), total protein (TP)] and liver fibrosis indicators [collagen Ⅳ (Ⅳ-C), hyaluronic acid (HA), laminin (LN), prothrombin Ⅲ (PCⅢ)] were evaluated on the second day, one month and three months after admission. Mixed linear models were used to analyze the differences and changing trends of each indicator between the two groups at different time points. Results: The examination results of the patients showed that the levels of BMI, Hb, ALB and TP in the intervention group were higher than those in the control group (P<0.05); the levels of ALT, PT, Ⅳ-C, HA, LN and PC Ⅲ were lower than those in the control group (P < 0.05). Mixed linear model analysis showed that the intervention group improved significantly faster than the control group over time in terms of various indicators, but the response magnitudes of different indicators were not completely consistent. The interaction effects between groups and time were significant for TP, ALB, ALT, PT, Ⅳ-C, HA, LN, and PC Ⅲ (P < 0.05). Conclusion: Nutritional intervention based on nutritional risk assessment by MDT can effectively improve the short-term nutritional status, liver function, and serum liver fibrosis markers in patients with cirrhosis. However, the follow-up period of the study was short, and long-term clinical outcomes cannot be inferred at this time.

Abstract:The liver is a common site for tumor metastases. Liver metastases often occur secondary to malignancies including colorectal cancer, pancreatic cancer, melanoma, lung cancer, and breast cancer, accounting for nearly 25% of all cancer metastases, and the 1-year survival rate for all patients with liver metastases is 15.1%. Early diagnosis and precise treatment are crucial for improving patient prognosis. As an imaging technique that combines functional metabolism with anatomical structure, ¹⁸F-fluoro-deoxy-D-glucose(¹⁸F-FDG)PET/CT plays an increasingly important role in the diagnosis, staging, and treatment planning of liver metastases. This article reviews recent advances in ¹⁸F-FDG PET/CT research on liver metastases, including its applications in diagnosis, differential diagnosis, pre-treatment evaluation, efficacy assessment, and prognosis prediction; its advantages and disadvantages in comparative imaging; new technological developments such as delayed imaging and image optimization; diagnostic pitfalls in certain diseases; and its role in guiding radiotherapy. Finally, it summarizes the application of artificial intelligence technology in the diagnosis and treatment of liver metastases and provides prospects on the future development directions of ¹⁸F-FDG PET/CT in this field.

Abstract:Liver sinusoidal endothelial cells (LSEC) are liver-specific endothelial cells. Under physiological conditions, LSECs exhibit open fenestrae, facilitating the efficient bidirectional exchange of small molecules between hepatocytes and blood. They are crucial for maintaining hepatic filtration and metabolic homeostasis, suppressing the activation of hepatic stellate cells (HSC), and modulating hepatic immune responses and tolerance, thereby forming the principal barrier of the liver microcirculation. During liver injury, however, LSECs undergo phenotypic changes—including capillarization, cellular senescence, and endothelial-to-mesenchymal transition. These changes result in increased intrahepatic vascular resistance, impaired transvascular exchange, and enhanced secretion of pro-angiogenic and pro-inflammatory factors. In the absence of timely intervention, this pathological cascade can exacerbate liver damage. Therefore, therapeutic targeting of LSEC dysfunction has emerged as a novel strategy for chronic liver disease. This article systematically elaborates on the dynamic changes and specific mechanistic roles of LSECs in both physiological homeostasis and chronic liver disease, summarizes current therapeutic strategies aimed at LSECs, and also discusses existing research limitations and suggests future directions.

Abstract:Objective:To investigate whether secreted modular calcium-binding protein 2(SMOC2) participates in miR-125a/miR-150-mediated phenotypic switching of pulmonary artery smooth muscle cell (PASMC) in pulmonary hypertnesion (PH). Methods:Human lung tissues and lung sections were collected from pulmonary hypertension(PH) patients and healthy controls. PH animal models were established in mice exposed to chronic hypoxia and in rats treated with SU5416 combined with chronic hypoxia. SMOC2 expression levels in lung tissues from PH patients, mice, and rats were assessed by Western blot, while its cellular localization was determined by immunofluorescence staining. Human PASMC were transfected with SMOC2 siRNA and miRNA mimics, or treated with recombinant SMOC2 protein. PASMC proliferation and migration were evaluated using EdU incorporation and wound healing assays. Dual-luciferase reporter assays was performed to validate SMOC2 as a direct target of miR-125a/miR-150. Results:SMOC2 expression was significantly upregulated in the lung tissues fromhuman and animal PH samples, thus in hypoxic PASMC. Inhibition of SMOC2 attenuated proliferation, migration, and phenotype switching in PASMC, whereas treatment with recombinant SMOC2 protein promoted PASMC proliferation and migration. miR-125a and miR-150 recognized and inhibted SMOC2 mRNA translation in PASMC. miR-125a and miR-150 mimics inhibited hypoxia-induced proliferation and phenotype switch of PASMC. Conclusion:miR-125a/miR-150 regulate hypoxia-induced PASMC phenotypic switching by directly targeting SMOC2, providing a potential therapeutic target for PH treatment.

Abstract:Objective: This study aimed to develop a biodegradable esophageal stent based on polycaprolactone (PCL) blended with poly (lactic-co-glycolic acid) (PLGA), achieving controllable degradation rates by adjusting the PLGA ratio (0%−35%), balancing mechanical properties and degradation cycles, and exploring its in vitro degradation characteristics and clinical adaptation potential. Methods: PCL and PLGA were selected to prepare eight gradient blend ratios (0%, 5%, 10%, 15%, 20%, 25%, 30%, 35% PLGA) by extrusion-based high-temperature melt rotary 3D printing technology, combined with a honeycomb porous structure design (porosity 60%−70%). In vitro degradation experiments lasting 8 weeks were conducted using artificial saliva (pH≈6.6) and artificial gastric juice (pH≈4.0). Results: The PLGA ratio significantly affected the degradation rate, with increasing PLGA proportion accelerating stent degradation. The blending of PLGA and PCL produced a synergistic effect (for every 10% increase in PLGA, the time to reach the same weight loss rate was shortened by approximately 50%−70%), leading to stage-wise disintegration phenomena (when PLGA≥25%). The pH difference had no significant impact on degradation (P > 0.05), possibly due to the high-porosity design promoting diffusion of acidic products and the hydrophobic balance of PCL. Through ratio adjustment, degradation cycles of 4−24 weeks could be achieved. Conclusion: The PCL/PLGA blend stent achieves precise control of degradation rate through simple ratio adjustment, producing a unique “support-disintegration” mode that reduces residual risks and meets clinical needs ranging from congenital strictures to acute inflammation. In addition, the combination of 3D printing technology with biomimetic morphology and porous structure design significantly improves tissue conformity and anti-migration properties of the stent. This study provides theoretical basis and technical support for personalized biodegradable esophageal stent therapy.

Abstract:Objective:To investigate the associations between frailty status and the risks of mortality and vascular complications in adults with type 1 diabetes (T1D). Methods:This prospective cohort study was based on the UK Biobank (UKB) and included participants with T1D at baseline. Frailty status was assessed using physical frailty (PF) and the frailty index (FI). The primary outcomes included all-cause mortality,T1D-related and complication-related mortality,and incident macrovascular and microvascular complications. Cox proportional hazards regression models were used to evaluate the associations between frailty status and each outcome. Dose-response analyses,subgroup analyses,and multiple sensitivity analyses were further conducted. Results:A total of 1 571 participants with T1D at baseline were included in the mortality analyses,with a median follow-up of 13.4 years;among them,1 207 participants without baseline vascular complications were included in the complication analyses. Based on PF and FI,12.7% and 28.5% of participants were classified as frail,respectively. After multivariable adjustment,compared with non-frail participants,frail individuals had a significantly higher risk of all-cause mortality[PF:hazard ratio(HR)=2.76,95% confidence interval(CI):1.92-3.98;FI:HR=2.71,95% CI:1.71-4.31]. The risk of T1D-related and complication-related mortality was also markedly increased among frail participants(PF:HR 4.02,95% CI:2.21-7.28;FI:HR 4.14,95% CI:1.73-9.89). Among participants without baseline vascular complications,frailty was significantly associated with higher risks of incident macrovascular complications(PF:HR=2.17,95% CI:1.46-3.22;FI:HR=2.39,95% CI:1.61-3.54)and microvascular complications(PF:HR=1.66,95% CI:1.18-2.32;FI:HR=2.05,95% CI:1.47-2.86). Dose-response analyses showed monotonic increases in the risks of multiple outcomes with increasing frailty severity,whereas a non-linear association was observed between PF scores and the risk of microvascular complications. Subgroup analyses showed that the association between FI and all-cause mortality was more pronounced among participants with lower systolic blood pressure (SBP) or lower body mass index (BMI),whereas the predictive effect of PF was relatively consistent across subgroups. Sensitivity analyses yielded results generally consistent with the main analyses. Conclusion:In adults with T1D,frailty was significantly associated with increased risks of mortality and incident vascular complications. Frailty assessment may help identify individuals at higher risk of poor prognosis and provide additional support for risk stratification and management in this population.

Abstract:Objective: The impact of sex on the prognosis of childhood immune thrombocytopenia(ITP) remains controversial, and the view that female sex is a high-risk factor for chronicity has not reached a universal consensus. This study aimed to construct sex-specific prediction models for chronic ITP through sex stratification, providing a basis for individualized prognosis assessment and early clinical intervention. Methods: This retrospective study enrolled 224 children initially diagnosed with ITP who were hospitalized in the Department of Pediatrics at The Affiliated Huai'an No.1 People's Hospital of Nanjing Medical University, between 1 January 2019 and 1 January 2023. The cohort included 128 males and 96 females. Demographic and clinical data were collected, with a follow-up period of at least one year. Multivariate logistic regression analysis was used to develop prediction models for chronic ITP separately for males and females. The performance of the models was evaluated using receiver operating characteristic(ROC) curves, calibration curves, and clinical decision curve analysis. Results: This study found that sex was an independent factor influencing the prognosis of childhood ITP(P<0.05). Sex-stratified prediction models were constructed. In the male model, absolute neutrophil count, neutrophil-to-lymphocyte ratio, platelet count on day 7 of treatment, complement C4, and bleeding grade were independent predictors of chronicity (P<0.05). The area under the ROC curve(AUC) for the combination of these five indicators was 0.879(95% CI: 0.819–0.938). However, in the female model, absolute lymphocyte count, complement C3, platelet count on day 7 of treatment, and immunoglobulin G were independent predictors of chronicity(P<0.05), with a combined AUC of 0.902(95% CI: 0.842–0.961). The calibration curves for both models were close to the ideal curve, and the clinical decision curves indicated a positive net clinical benefit within a threshold probability range of 0.10–0.70. Conclusion: Predictive factors for chronicity in childhood ITP showed significant sex differences. The sex-stratified prediction models demonstrated good predictive performance.

Abstract:Objective: This study employed lipidomics analysis to systematically screen potential biomarkers of childhood obesity comorbid with central precocious puberty (CO-CPP), and explored the feasibility of combining these biomarkers with disease-related clinical indicators to provide new evidence for early identification and warning of CO-CPP. Methods: A total of 20 obese female children aged 6 to 8 years who first visited the Pediatric Endocrinology Clinic of the Affiliated Huai'an No. 1 People's Hospital of Nanjing Medical University between January 2024 and June 2025 were enrolled. They were divided into the CO-CPP group (n=10) and the comorbidity of childhood obesity comorbid with peripheral precocious puberty (CO-PPP) group (n=10) based on the presence of central precocious puberty (CPP) or peripheral precocious puberty (PPP). General clinical data were compared between the two groups, and serum samples were collected for non-targeted lipidomics analysis. Differential lipids were screened using receiver operating characteristic (ROC) curve analysis, and a combined predictive model was constructed with clinical indicators via binary logistic regression, followed by further combined ROC analysis. Results: Lipidomics analysis identified 42 lipid molecules with significant differences. Among them, the expression of ceramide (Cer) and phosphatidylcholine (PC) were up-regulated, while the expression of phosphatidylethanolamine (PE) was down-regulated. The area under the curve (AUC) of ROC for predicting CO-CPP by Cer, PC, and PE alone was 0.810, 0.798, and 0.834, respectively, all of which showed good discriminative efficacy (all P < 0.05). The AUC of the combined prediction model of Cer and the basal level of luteinizing hormone (LH) was 0.970. Conclusion: In girls with CPP, the expressions of Cer and PC, the key lipids associated with pubertal development, are up-regulated, while the expression of PE is down-regulated. These lipids may serve as potential biological markers for CPP. The discriminative efficacy of Cer combined with basal LH is superior to that of single lipid indicators, showing good potential for clinical application.

Abstract:Objective: To investigate the metabolic profiles of blood amino acids, carnitines, and succinylacetone in neonates, analyze their influencing factors, and identify differential metabolites and metabolic pathways between preterm and full-term infants, thereby providing a basis for clinical intervention in preterm infants. Methods: A total of 19 599 neonates delivered at medical and healthcare institutions in Huai' an City, Jiangsu Province from July 2023 to June 2024 were enrolled. According to gestational age, they were divided into two groups: the preterm group (28 weeks≤gestational age<37 weeks, n=1 107) and the full-term group (37 weeks ≤ gestational age < 42 weeks, n=18 492). Dried blood spot samples were collected within 7 days after birth. Tandem mass spectrometry was used to measure blood metabolic parameters. Spearman rank correlation analysis was applied to examine the correlations among sex, gestational age, birth weight, and metabolic parameters. The t test and partial least squares discriminant analysis (PLS-DA) were used to identify significantly different metabolites between the two groups, and the intersection of the differential metabolites obtained by the two methods was selected as potential biologically significant biomarkers. Pathway enrichment analysis was performed using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database to identify differential metabolic pathways. Results: Multiple blood amino acids and carnitine metabolic parameters were significantly correlated with neonatal sex, gestational age, and birth weight(P<0.05). A total of 18 significantly different metabolites were identified between preterm and full-term infants. Of them, 8 metabolites were upregulated in the preterm group, including free carnitine (C0), arginine (Arg), and phenylalanine (Phe); 10 metabolites were downregulated, including alanine (Ala), glycine (Gly), and decanoylcarnitine (C10). These differential metabolites were involved in 17 metabolic pathways, including arginine and proline metabolism; phenylalanine, tyrosine and tryptophan biosynthesis; phenylalanine metabolism; biosynthesis of unsaturated fatty acids; fatty acid degradation; and fatty acid biosynthesis. Conclusion: Blood amino acid and carnitine levels in neonates are significantly correlated with sex, gestational age, and birth weight. There are distinct metabolic differences between preterm and full-term infants, which may affect amino acid and fatty acid metabolic processes.

Abstract:Objective:To explore the association between infant visual acuity and neurodevelopment at 3 years of age. Methods:We employed a prospective cohort study design. A total of 1 122 children from 1 096 families enrolled in the Jiangsu Birth Cohort between August 2016 and June 2019 were included. At 1 year of age, visual acuity was measured using the Teller Acuity Card Ⅱ. At 3 years of age, neurodevelopmental status was assessed using the Bayley Scales of Infant and Toddler Development, Version Ⅲ Screening Test, covering five domains: cognition, receptive communication, expressive communication, fine motor, and gross motor. Poisson regression models were used to examine the associations between infant visual acuity and neurodevelopment at 3 years of age. Results:Among the 1 122 children, 94 (8.4% ) showed abnormal visual acuity. The prevalence of noncompetent neurodevelopment across five domains ranged from 3.5% to 10.3%. Compared with the group with normal visual acuity in infancy, the group with abnormal visual acuity had a 96% increased risk of being noncompetent development in gross motor domain at age 3 (RR=1.96, 95% CI 1.03- 3.70). This association remained stable after excluding twins, preterm infants, low birth weight infants and children conceived via assisted reproductive technology. Conclusion:Abnormal visual acuity in infancy was associated with an increased risk of being noncompetent in gross motor development at 3 years of age.

Abstract:Objective: To clarify the definition of pseudolock phenomenon during locking plate osteosynthesis, propose a novel clinically oriented classification, and explore the incidence and distribution characteristics of this phenomenon. Methods: According to the insertion state of locking head screws in the threaded holes of locking plates, the pseudolock phenomenon was classified into three types: Type A, the screw depth is insufficient, but the orientation is normal in a precontoured plate; Type B, both the screw depth and orientation are abnormal in a precontoured plate; Type C, the screw depth and orientation are either/both abnormal in an overbent plate. Demographic and radiographic data of 1 796 patients with upper limb long bone fractures (clavicle, humerus, ulna, and radius) who underwent locking plate osteosynthesis from June 2017 to April 2024 were collected and analyzed. A Chi-square test with Bonferroni correction was used to assess the incidence of pseudolock incidences across different genders, ages, sites, sides, locations, and diameters. Results: A total of 1 878 locking plates were identified in the long bones of the upper limbs, distributed as follows: clavicle 1 129(60.1%), humerus 364(19.4%), ulna 137(7.3%), and radius 248(13.2%). In total, 11 636 locking head screws placed in these plates were included in the analysis, of which 2 564(22.0%)were pseudolocked. Type A pseudolock was the most common with an incidence of 14.1%(1 646/11 636), followed by Type B at 7.8%(911/11 636), while Type C was rare with an incidence of 0.1%(7/11 636). Conclusion: The pseudolock phenomenon in locking plate osteosynthesis is defined and a new classification system is established. The incidence of pseudolock exceeds 20% in osteosynthesis of upper extremity long bones including clavicle, humerus, ulna and radius. Mastering the preset insertion depth and orientation of screws for locking plates and identifying pseudolock timely via tangential fluoroscopy during operation are crucial measures to improve the fixation stability of locking plates.

Abstract:Objective: To evaluate the predictive efficacy of spectral CT quantitative parameters for central lymph node metastasis (CLNM) in preoperative clinical node-negative (cN0) papillary thyroid carcinoma (PTC) patients. Methods: Clinical and imaging data of 86 surgical pathologically confirmed PTC patients who underwent preoperative spectral CT imaging, were retrospectively collected. Differences in clinical and spectral CT features between the CLNM and non-CLNM groups were compared. Receiver operating characteristic (ROC) curve analysis was used to assess the predictive performance of clinical features, spectral CT features, and a combined model, with area under the curve (AUC), sensitivity, and specificity calculated. Results: Among clinical indicators, sex, thyroglobulin (Tg) level, and cystic changes showed statistically significant differences between CLNM and non-CLNM groups (all P < 0.05). Among non-contrast (NC) spectral CT parameters, 40 keV_HU and 70 keV_HU values were significantly lower in CLNM group (all P < 0.05). Among arterial phase (AP) spectral CT parameters, all parameters except arterial normalized effective atomic number (NZ_eff) differed significantly between two groups (all P < 0.05). Among venous phase spectral CT parameters, 70 keV_HU value was significantly lower in CLNM group (P=0.014). Combined model demonstrated the highest predictive efficacy (AUC=0.896, sensitivity=86.7%, specificity=80.4% ), followed by spectral CT quantitative parameter model (AUC=0.835, sensitivity=73.3%, specificity=85.7% ), and clinical model (AUC=0.591, sensitivity=20.0%, specificity=98.2% ; all P < 0.05). Conclusion: Spectral CT quantitative parameters based on lesion characteristics improve the preoperative prediction of CLNM in cN0 PTC patients, and aid in treatment plan.

Abstract:Artificial intelligence (AI) is driving the paradigm shift of modern drug delivery systems towards intelligence and precision. To address the challenges of long cycles, high costs, and insufficient targeting in traditional research and development, this article systematically reviews the empowering role of AI technology throughout the entire drug delivery chain. By analyzing application examples of AI technologies such as machine learning and deep learning in the development of various dosage forms (including traditional dosage forms, sustained-release systems, microspheres/microcapsules, nanocarriers, and 3D-printed formulations), the article elucidates the significant advantages of AI models in predicting drug release behavior, optimizing formulation design, intelligently identifying formulation defects, and achieving personalized drug delivery. The article points out that despite challenges such as data quality and model interpretability, AI has undoubtedly become the core driving force propelling drug delivery towards the era of intelligent response and precision medicine, demonstrating immense potential in the field of pharmaceutical research and development transformation.

Abstract:Clinical studies have indicated that the disease progression of uveitis may be associated with multiple factors, including immune cell subsets, vitamin D levels, gut microbiota, and inflammatory cytokines levels. However, traditional clinical observational studies are susceptible to confounding factors when inferring causality, making it difficult to establish definitive causal relationships. Therefore, it is necessary to introduce alternative causal inference methods to validate the true associations between these potential factors and the onset and progression of the disease. Mendelian randomization (MR), as an emerging causal inference approach, has been widely applied in recent years to investigate the etiology of uveitis. By using genetic variants as instrumental variables, this method effectively overcomes the confounding bias and reverse causation issues inherent in traditional observational studies. In the context of uveitis research, MR analysis has also demonstrated significant value in mediation analysis, multivariable integration, and drug target prioritization.

Abstract:Peptide compounds exhibit the advantages of high target affinity, strong specificity, and low immunogenicity. Therefore, after clarifying the therapeutic target, peptide compounds can be obtained as lead compounds via the protein fragment truncation strategy. However, the inherent structural limitations of peptide compounds result in several drawbacks, such as poor stability, low oral bioavailability, and high production costs. In contrast, small-molecule compounds possess the prominent advantages of high oral bioavailability, metabolic stability, and low production costs. Thus, it is particularly important to screen small-molecule compounds that can mimic the action mode of peptides and retain biological activity by analyzing the interaction modes between peptide compounds and target proteins. Focusing on intractable brain diseases such as Alzheimer's disease, this paper systematically elaborates on the research and development pathway of deriving active peptide compounds from functional proteins, deciphering the interaction modes between peptides and target proteins, and further obtaining targeted small-molecule compounds through five typical transformation cases, which can provide a valuable reference for subsequent compound design.

Abstract:Tumor drug resistance is the main cause of clinical treatment failure. According to conventional wisdom, resistance is mostly caused by greater drug efflux, accelerated DNA repair, and decreased apoptosis. In recent years, more and more studies have shown that glucose metabolic reprogramming, as a key mechanism for tumor cells to adapt to the unique tumor microenvironment and obtain the advantages of survival and proliferation, also plays a very important role in regulating the process of tumor drug resistance. In addition to providing sustained material and energy supplies for their own rapid growth and proliferation, the abnormally active aerobic glycolysis and pentose phosphate pathway in tumor cells can also affect drug tolerance through pathways like regulating redox homeostasis, enhancing cellular stemness, and reshaping the tumor microenvironment. Additionally, the combination treatments that target glycose metabolic reprogramming combined with anticancer medications have shown progressively greater therapeutic benefits. This article reviews the research progress of glucose metabolic reprogramming in the resistance to various anti-tumor drugs, in order to provide a new breakthrough for the increasingly prominent status quo of tumor cell drug resistance.

Abstract:Acute myeloid leukemia (AML) is a highly heterogeneous and aggressive hematologic malignancy whose pathogenesis and progression are influenced by a variety of molecular mechanisms. These encompass genetic mutations, aberrant signaling pathways, epigenetic dysregulation, and complex interactions within the tumor microenvironment. Hypoxia-inducible factor- 1α (HIF-1α), a master regulator of cellular response to low oxygen tension, is frequently overexpressed in AML, and its abnormal expression is closely associated with poor patient prognosis. The histone methyltransferase enhancer of zeste homolog 2(EZH2), recognized as a high-risk molecular marker in the Chinese adult AML diagnosis and treatment guidelines, exhibits a context-dependent dual regulatory role. It can lose its tumor-suppressive function through loss-of-function mutations, while in specific disease stages, its overexpression contributes to maintaining leukemia stem cell properties and mediating chemotherapy resistance. Research indicates that HIF- 1α and EZH2 may engage in synergistic crosstalk in AML, potentially forming a "metabolism-epigenetics" axis. This interaction creates a positive feedback loop that cooperatively promotes leukemia cell proliferation, blocks differentiation, facilitates metabolic reprogramming, and aids in immune evasion. This article provides a comprehensive review of the expression patterns, biological functions, interaction mechanisms, and clinical translational potential of HIF-1α and EZH2 in AML. The aim is to offer a theoretical foundation and suggest research directions for a deeper understanding of AML pathogenesis and the development of novel combination targeted therapy strategies.