Lidan Hu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Lidan Hu has a total of 300 co-authors. Here are the top co-authors by number of shared publications:
- Xiangjun Chen
Eye Center of the Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China
18 shared publications - Jianhua Mao
Department of Nephrology, Children’s Hospital, Zhejiang University School of Medicine, National Clinical Research Center for Child Health, National Children’s Regional Medical Center, Hangzhou, China
16 shared publications Ying Zhang
Eye Center of the Second Affiliated Hospital, Zhejiang University School of Medicine Hangzhou China
11 shared publicationsGuannan Bai
Department of Child Health Care, Children’s Hospital, Zhejiang University School of Medicine, National Clinical Research Center for Child Health, Hangzhou, China
9 shared publicationsHaidong Fu
Department of Nephrology The Children's Hospital, Zhejiang University School of Medicine, National clinical research center for child health Hangzhou China
9 shared publicationsWei Wu
9 shared publications
Latest Publications
- Open Access
- 30 June 2026
- Published by Elsevier in SSRN Electronic Journal
Apolipoprotein-Informed Explainable Machine Learning for Predicting Dialysis Initiation in Diabetic Kidney Disease
Background: Diabetic kidney disease (DKD) is a leading cause of kidney failure and represents a metabolically complex, cardiorenal disease state. Existing kidney failure prediction models are largely derived from broad chronic kidney disease (CKD) cohorts and may not adequately capture the diabetes-specific metabolic, cardiovascular, and lipid-related determinants of progression. We aimed to develop and externally validate an explainable machine learning (ML) model for predicting dialysis initiation in patients with DKD, and to test whether apolipoprotein profiling adds prognostic information beyond conventional renal markers.Methods: We conducted a retrospective inpatient cohort study using electronic medical record (EMR) data from a tertiary hospital. Among 5,644 hospitalized patients with DKD identified between January 2015 and December 2023, 3,078 were included after predefined exclusions. The primary outcome was dialysis initiation, defined as hemodialysis or peritoneal dialysis. Prediction horizons were evaluated at 2, 5, and 10 years. After preprocessing, 36 routinely available variables were retained. Candidate predictors were selected using univariate and multivariable logistic and Cox regression analyses. Six ML algorithms were compared: decision tree, random forest, support vector machine, XGBoost, LightGBM, and multilayer perceptron. Model performance was assessed by discrimination, calibration, and decision curve analysis, and Shapley Additive Explanations (SHAP) were used for model interpretation. Two independent hospital cohorts were used for external validation. In a subset with available apolipoprotein measurements (n = 1,855), we further assessed the incremental contribution of ApoB and ApoAI.Results: Among 3,078 patients with DKD, 227 (7.4%) initiated dialysis during follow-up. eGFR and UACR were the dominant renal predictors across all prediction horizons. The final seven-variable clinical model included eGFR, UACR, cardiovascular disease (CVD), hemoglobin (HGB), sex, age, and diabetes duration; serum phosphate showed horizon-specific associations but was not retained to maintain a consistent predictor set. XGBoost provided the best overall performance, with AUCs of 0.9274, 0.9154, and 0.9095 for the 2-, 5-, and 10-year cohorts, respectively. SHAP analysis confirmed eGFR and UACR as the principal risk drivers, with CVD, HGB, age, sex, and diabetes duration contributing additional cardiorenal context. In the internal apolipoprotein subset, adding ApoB and ApoAI improved AUC from 0.923 to 0.955 at 2 years, from 0.905 to 0.928 at 5 years, and from 0.904 to 0.923 at 10 years (P < 0.0001, P < 0.0001, and P = 0.0018, respectively). External validation in two independent cohorts supported model transportability, with AUCs consistently exceeding 0.91, although the incremental apolipoprotein benefit was not reproduced externally.Conclusions: We developed and externally validated an explainable ML model for predicting dialysis initiation in DKD using routinely available clinical variables. The model integrates renal, cardiovascular, hematologic, and metabolic risk signals and may support structured risk review after prospective testing. Internally, ApoB and ApoAI improved discrimination, suggesting that apolipoprotein profiling may identify a lipoprotein-related DKD progression phenotype; however, this exploratory signal requires confirmation in larger, prospective, and more diverse cohorts.
- 30 June 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessApolipoprotein-Informed Explainable Machine Learning for Predicting Dialysis Initiation in Diabetic Kidney Disease
Background: Diabetic kidney disease (DKD) is a leading cause of kidney failure and represents a metabolically complex, cardiorenal disease state. Existing kidney failure prediction models are largely derived from broad chronic kidney disease (CKD) cohorts and may not adequately capture the diabetes-specific metabolic, cardiovascular, and lipid-related determinants of progression. We aimed to develop and externally validate an explainable machine learning (ML) model for predicting dialysis initiation in patients with DKD, and to test whether apolipoprotein profiling adds prognostic information beyond conventional renal markers.Methods: We conducted a retrospective inpatient cohort study using electronic medical record (EMR) data from a tertiary hospital. Among 5,644 hospitalized patients with DKD identified between January 2015 and December 2023, 3,078 were included after predefined exclusions. The primary outcome was dialysis initiation, defined as hemodialysis or peritoneal dialysis. Prediction horizons were evaluated at 2, 5, and 10 years. After preprocessing, 36 routinely available variables were retained. Candidate predictors were selected using univariate and multivariable logistic and Cox regression analyses. Six ML algorithms were compared: decision tree, random forest, support vector machine, XGBoost, LightGBM, and multilayer perceptron. Model performance was assessed by discrimination, calibration, and decision curve analysis, and Shapley Additive Explanations (SHAP) were used for model interpretation. Two independent hospital cohorts were used for external validation. In a subset with available apolipoprotein measurements (n = 1,855), we further assessed the incremental contribution of ApoB and ApoAI.Results: Among 3,078 patients with DKD, 227 (7.4%) initiated dialysis during follow-up. eGFR and UACR were the dominant renal predictors across all prediction horizons. The final seven-variable clinical model included eGFR, UACR, cardiovascular disease (CVD), hemoglobin (HGB), sex, age, and diabetes duration; serum phosphate showed horizon-specific associations but was not retained to maintain a consistent predictor set. XGBoost provided the best overall performance, with AUCs of 0.9274, 0.9154, and 0.9095 for the 2-, 5-, and 10-year cohorts, respectively. SHAP analysis confirmed eGFR and UACR as the principal risk drivers, with CVD, HGB, age, sex, and diabetes duration contributing additional cardiorenal context. In the internal apolipoprotein subset, adding ApoB and ApoAI improved AUC from 0.923 to 0.955 at 2 years, from 0.905 to 0.928 at 5 years, and from 0.904 to 0.923 at 10 years (P < 0.0001, P < 0.0001, and P = 0.0018, respectively). External validation in two independent cohorts supported model transportability, with AUCs consistently exceeding 0.91, although the incremental apolipoprotein benefit was not reproduced externally.Conclusions: We developed and externally validated an explainable ML model for predicting dialysis initiation in DKD using routinely available clinical variables. The model integrates renal, cardiovascular, hematologic, and metabolic risk signals and may support structured risk review after prospective testing. Internally, ApoB and ApoAI improved discrimination, suggesting that apolipoprotein profiling may identify a lipoprotein-related DKD progression phenotype; however, this exploratory signal requires confirmation in larger, prospective, and more diverse cohorts.
- Open Access
- 21 June 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Lipopolysaccharide hydrolysis-targeting nano-chimeras detoxify endotoxin through specific adsorption and efficient degradation
Lipopolysaccharide (LPS), a potent immunogenic component of the outer membrane of Gram-negative bacteria, triggers severe inflammation and organ failure even at nanomolar concentrations. However, neutralizing LPS in vivo remains challenging due to the high abundance of other biomolecules in biological fluids, which interfere with LPS detoxification. Brassicaceae species produce a transmembrane protein termed lipooligosaccharide-specific reduced elicitation (LORE) that specifically recognizes and binds LPS. Here, we prepare plant-derived nanovesicles naturally presenting LORE on their surface and integrate them with cerium-based nanozymes exhibiting LPS hydrolysis activity. We show that these hybrid nanostructures (Atv@Ce) neutralize LPS through two coordinated mechanisms: LORE captures LPS specifically, while nanozymes chemically degrade the phosphate groups and glycosidic bonds in the lipid A moiety. This dual strategy effectively attenuates both local and systemic inflammation, offering a biocompatible detoxification strategy with translational potential. Our work provides an insight into nanomaterial-mediated detoxification.
- 21 June 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Open AccessLipopolysaccharide hydrolysis-targeting nano-chimeras detoxify endotoxin through specific adsorption and efficient degradation
Lipopolysaccharide (LPS), a potent immunogenic component of the outer membrane of Gram-negative bacteria, triggers severe inflammation and organ failure even at nanomolar concentrations. However, neutralizing LPS in vivo remains challenging due to the high abundance of other biomolecules in biological fluids, which interfere with LPS detoxification. Brassicaceae species produce a transmembrane protein termed lipooligosaccharide-specific reduced elicitation (LORE) that specifically recognizes and binds LPS. Here, we prepare plant-derived nanovesicles naturally presenting LORE on their surface and integrate them with cerium-based nanozymes exhibiting LPS hydrolysis activity. We show that these hybrid nanostructures (Atv@Ce) neutralize LPS through two coordinated mechanisms: LORE captures LPS specifically, while nanozymes chemically degrade the phosphate groups and glycosidic bonds in the lipid A moiety. This dual strategy effectively attenuates both local and systemic inflammation, offering a biocompatible detoxification strategy with translational potential. Our work provides an insight into nanomaterial-mediated detoxification.
- Open Access
- 17 June 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Attention problems statistically mediate the association between mobile phone addiction and mental health and procrastination in Chinese college students
Mobile phone addiction is a significant public health issue among college students, associated with poor mental health and procrastination, yet the underlying mechanisms involving attention problems are not well understood. A cross-sectional survey was conducted among college students in Zhejiang Province. Linear regression models were employed to assess the relationships between mobile phone addiction, mental health, and procrastination. Statistical mediation analysis was performed to evaluate the role of attention problems. Data from 755 college students were analyzed. After adjusting for covariates, mobile phone addiction scores were significantly associated with poorer mental health (β = 0.44, p < 0.001) and more severe procrastination (β = 0.23, p < 0.001). In the pathway from mobile phone addiction to mental health, the indirect effect via attention problems (0.159, 95% CI 0.104–0.225) accounted for 36.55% of the total effect, indicating statistical indirect effects consistent with partial mediation. Similarly, in the pathway from mobile phone addiction to procrastination, attention problems showed a statistically significant indirect effect (0.076, 95% CI 0.041–0.115) accounting for 33.33% of the total effect. In this cross-sectional, self-report sample of Chinese college students, attention problems consistent with a partial mediating role in the associations between mobile phone addiction and both mental health issues and procrastination. The results generate testable hypotheses for preventive interventions targeting attention problems. Future longitudinal or experimental research is needed to confirm these findings.
- 17 June 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Open AccessAttention problems statistically mediate the association between mobile phone addiction and mental health and procrastination in Chinese college students
Mobile phone addiction is a significant public health issue among college students, associated with poor mental health and procrastination, yet the underlying mechanisms involving attention problems are not well understood. A cross-sectional survey was conducted among college students in Zhejiang Province. Linear regression models were employed to assess the relationships between mobile phone addiction, mental health, and procrastination. Statistical mediation analysis was performed to evaluate the role of attention problems. Data from 755 college students were analyzed. After adjusting for covariates, mobile phone addiction scores were significantly associated with poorer mental health (β = 0.44, p < 0.001) and more severe procrastination (β = 0.23, p < 0.001). In the pathway from mobile phone addiction to mental health, the indirect effect via attention problems (0.159, 95% CI 0.104–0.225) accounted for 36.55% of the total effect, indicating statistical indirect effects consistent with partial mediation. Similarly, in the pathway from mobile phone addiction to procrastination, attention problems showed a statistically significant indirect effect (0.076, 95% CI 0.041–0.115) accounting for 33.33% of the total effect. In this cross-sectional, self-report sample of Chinese college students, attention problems consistent with a partial mediating role in the associations between mobile phone addiction and both mental health issues and procrastination. The results generate testable hypotheses for preventive interventions targeting attention problems. Future longitudinal or experimental research is needed to confirm these findings.
- 17 June 2026
- Published by Elsevier in Ageing Research Reviews
- Vol. 120, 103209
Biomarkers and therapies associated with Hutchinson-Gilford Progeria Syndrome
- 17 June 2026
- Published by Elsevier in Ageing Research Reviews
- Vol. 120, 103209
Biomarkers and therapies associated with Hutchinson-Gilford Progeria Syndrome
- Open Access
- 20 March 2026
- Published by Wiley in Journal of Cellular and Molecular Medicine
- Vol. 30 (6) , e71108
Molecular Mechanisms of CLCN5 Missense Mutations in Dent Disease Type 1: A Comprehensive Computational Analysis and Clinical Correlations in a Chinese Cohort
Dent's disease, an X‐linked recessive disorder predominantly affecting males, is characterized by nephrocalcinosis, nephrolithiasis, and a high risk of progression to end‐stage renal disease. Dent's disease type 1, accounting for 60% of cases, caused by mutations in the CLCN5 gene encoding the chloride ion channel protein ClC‐5, exhibits significant clinical heterogeneity and variability in disease progression. The lack of hotspot mutations poses challenges for genetic diagnosis and counselling, complicating the prediction of disease outcomes. This study systematically evaluated the functional and structural impacts of 181 CLCN5 missense mutations using computational tools, including PredictSNP, MAGPIE, and molecular dynamics simulations, to propose a robust method for improving genetic counselling and prognosis prediction. Our analysis identified mutations at the dimer interface and chloride selectivity filter as critical disruptors of ClC‐5 function and stability. Notably, molecular dynamics simulations of L200R, P213L, and G512R mutations revealed that L200R significantly destabilized the protein structure. Clinical data from a multicentre cohort of Chinese patients with CLCN5 mutations corroborated our computational predictions, highlighting the essential role of helix O in ClC‐5 function. By integrating bioinformatics analyses with clinical validation, this study provides molecular insights into Dent's disease heterogeneity and proposes a framework for enhancing genetic counselling and prognostic assessment for affected patients.
- 20 March 2026
- Published by Wiley in Journal of Cellular and Molecular Medicine
- Vol. 30 (6) , e71108
Open AccessMolecular Mechanisms of CLCN5 Missense Mutations in Dent Disease Type 1: A Comprehensive Computational Analysis and Clinical Correlations in a Chinese Cohort
Dent's disease, an X‐linked recessive disorder predominantly affecting males, is characterized by nephrocalcinosis, nephrolithiasis, and a high risk of progression to end‐stage renal disease. Dent's disease type 1, accounting for 60% of cases, caused by mutations in the CLCN5 gene encoding the chloride ion channel protein ClC‐5, exhibits significant clinical heterogeneity and variability in disease progression. The lack of hotspot mutations poses challenges for genetic diagnosis and counselling, complicating the prediction of disease outcomes. This study systematically evaluated the functional and structural impacts of 181 CLCN5 missense mutations using computational tools, including PredictSNP, MAGPIE, and molecular dynamics simulations, to propose a robust method for improving genetic counselling and prognosis prediction. Our analysis identified mutations at the dimer interface and chloride selectivity filter as critical disruptors of ClC‐5 function and stability. Notably, molecular dynamics simulations of L200R, P213L, and G512R mutations revealed that L200R significantly destabilized the protein structure. Clinical data from a multicentre cohort of Chinese patients with CLCN5 mutations corroborated our computational predictions, highlighting the essential role of helix O in ClC‐5 function. By integrating bioinformatics analyses with clinical validation, this study provides molecular insights into Dent's disease heterogeneity and proposes a framework for enhancing genetic counselling and prognostic assessment for affected patients.
- Open Access
- 1 January 2026
- Published by Maximum Academic Press in Visual Neuroscience
- Vol. 43 (1)
Troxerutin targets c-RAF to inhibit ERK signaling and protect against diabetic retinopathy
Diabetic retinopathy (DR) remains a leading cause of preventable blindness in working-age adults worldwide. Current clinical interventions are mostly invasive and ineffective in halting early DR progression, highlighting an urgent need for safe and orally active therapeutic agents. Troxerutin (Trx), a clinically approved flavonoid derivative with well-established safety, exhibits antioxidant, anti-inflammatory, and vascular-protective properties. This study used a streptozotocin-induced diabetic rat model to investigate the protective effect of Trx against DR and elucidate its underlying molecular mechanism; retinal injury was evaluated by histology, immunofluorescence, qPCR, and biochemical assays, while network pharmacology, molecular docking, and molecular dynamics were applied to identify targets, and Western blotting was used for validation. Results showed that Trx effectively ameliorated retinal edema and glial activation in diabetic rats, improved glutathione redox balance, and reduced the expression of inflammatory and pro-fibrotic factors. Our work reveals c-RAF as a candidate target for troxerutin. Trx alleviates c-RAF accumulation and inhibits ERK phosphorylation, while computational analysis predicts that Trx can stably bind to c-RAF and exacerbate its conformational instability. It is concluded that Trx exerts protective effects against DR by targeting c-RAF and inhibiting the RAF/ERK signaling pathway, which provides a new theoretical basis for its application as a potential oral drug in the treatment of diabetic retinopathy.
- 1 January 2026
- Published by Maximum Academic Press in Visual Neuroscience
- Vol. 43 (1)
Open AccessTroxerutin targets c-RAF to inhibit ERK signaling and protect against diabetic retinopathy
Diabetic retinopathy (DR) remains a leading cause of preventable blindness in working-age adults worldwide. Current clinical interventions are mostly invasive and ineffective in halting early DR progression, highlighting an urgent need for safe and orally active therapeutic agents. Troxerutin (Trx), a clinically approved flavonoid derivative with well-established safety, exhibits antioxidant, anti-inflammatory, and vascular-protective properties. This study used a streptozotocin-induced diabetic rat model to investigate the protective effect of Trx against DR and elucidate its underlying molecular mechanism; retinal injury was evaluated by histology, immunofluorescence, qPCR, and biochemical assays, while network pharmacology, molecular docking, and molecular dynamics were applied to identify targets, and Western blotting was used for validation. Results showed that Trx effectively ameliorated retinal edema and glial activation in diabetic rats, improved glutathione redox balance, and reduced the expression of inflammatory and pro-fibrotic factors. Our work reveals c-RAF as a candidate target for troxerutin. Trx alleviates c-RAF accumulation and inhibits ERK phosphorylation, while computational analysis predicts that Trx can stably bind to c-RAF and exacerbate its conformational instability. It is concluded that Trx exerts protective effects against DR by targeting c-RAF and inhibiting the RAF/ERK signaling pathway, which provides a new theoretical basis for its application as a potential oral drug in the treatment of diabetic retinopathy.
- Open Access
- 1 January 2026
- Published by Maximum Academic Press in Visual Neuroscience
- Vol. 43 (1)
Deciphering the gut–eye axis: evidence from observational and genetic analyses linking inflammatory bowel disease with ocular disorders
Ocular comorbidities in inflammatory bowel disease (IBD) have gained increasing attention, with the gut–eye axis hypothesized as a potential biological bridge. However, the genetic architecture and underlying mechanisms remain underexplored. In this study, we integrated longitudinal cohort data from the UK Biobank with genome-wide summary statistics to systematically decode these associations. Observational analyses revealed that IBD, including the subtypes of Crohn's disease (CD) and ulcerative colitis (UC), significantly elevated the risks of multiple ocular comorbidities, specifically uveitis, cataracts, age-related macular degeneration (AMD), glaucoma, and diabetic retinopathy (DR). Mediation analyses further demonstrated that these risks were partially explained by systemic inflammatory biomarkers. Genetically, we identified significant global and local genetic correlations between IBD and various ocular comorbidities. A cross-trait meta-analysis pinpointed 49 novel pleiotropic variants within ρ-heritability estimation from summary statistics (HESS) significant regions, highlighting shared pathways in immune regulation. Finally, two-sample Mendelian randomization established robust causal relationships between IBD (or its subtypes) and uveitis and AMD. Our findings elucidated a shared genomic foundation and identified systemic inflammation as a contributing link within the gut–eye axis. These results provided novel insights into the co-pathogenesis of intestinal and ocular comorbidities, suggesting that early ophthalmic screening and targeted immunomodulatory strategies are essential for the comprehensive management of IBD patients.
- 1 January 2026
- Published by Maximum Academic Press in Visual Neuroscience
- Vol. 43 (1)
Open AccessDeciphering the gut–eye axis: evidence from observational and genetic analyses linking inflammatory bowel disease with ocular disorders
Ocular comorbidities in inflammatory bowel disease (IBD) have gained increasing attention, with the gut–eye axis hypothesized as a potential biological bridge. However, the genetic architecture and underlying mechanisms remain underexplored. In this study, we integrated longitudinal cohort data from the UK Biobank with genome-wide summary statistics to systematically decode these associations. Observational analyses revealed that IBD, including the subtypes of Crohn's disease (CD) and ulcerative colitis (UC), significantly elevated the risks of multiple ocular comorbidities, specifically uveitis, cataracts, age-related macular degeneration (AMD), glaucoma, and diabetic retinopathy (DR). Mediation analyses further demonstrated that these risks were partially explained by systemic inflammatory biomarkers. Genetically, we identified significant global and local genetic correlations between IBD and various ocular comorbidities. A cross-trait meta-analysis pinpointed 49 novel pleiotropic variants within ρ-heritability estimation from summary statistics (HESS) significant regions, highlighting shared pathways in immune regulation. Finally, two-sample Mendelian randomization established robust causal relationships between IBD (or its subtypes) and uveitis and AMD. Our findings elucidated a shared genomic foundation and identified systemic inflammation as a contributing link within the gut–eye axis. These results provided novel insights into the co-pathogenesis of intestinal and ocular comorbidities, suggesting that early ophthalmic screening and targeted immunomodulatory strategies are essential for the comprehensive management of IBD patients.
- Open Access
- 7 December 2025
- Published by Springer Nature in Virology Journal
- Vol. 23 (1)
Epidemiological characteristics of hepatitis B virus infection in adult patients before, during and after COVID-19 in China
The onset of COVID-19 and subsequent restrictive measures have impacted various infectious diseases, including hepatitis B virus (HBV). This study explored
- 7 December 2025
- Published by Springer Nature in Virology Journal
- Vol. 23 (1)
Open AccessEpidemiological characteristics of hepatitis B virus infection in adult patients before, during and after COVID-19 in China
The onset of COVID-19 and subsequent restrictive measures have impacted various infectious diseases, including hepatitis B virus (HBV). This study explored
- Open Access
- 30 July 2025
- Published by Wiley in Smart Medicine
- Vol. 4 (3)
Identification of Differential Metabolites in Chronic Suppurative Otitis Media With Non‐Targeted and Targeted Metabolomics Approach
Chronic suppurative otitis media (CSOM) is a leading cause of hearing loss and otorrhea, and when associated with cholesteatoma, it can pose a serious threat to patients' lives. This study aims to identify differences in tissue metabolites between patients with CSOM, both with and without cholesteatoma. Metabolomic profiles were measured in tissue samples from 42 surgically treated CSOM patients (35 with cholesteatoma, 7 without cholesteatoma). Significantly altered metabolites associated with CSOM were identified using a non‐targeted metabolomics approach and a targeted metabolomics approach. The 42 patients were divided into screening and validation sets. The non‐targeted analysis revealed 484 distinct differential metabolites and 32 metabolic pathways that differed between CSOM with and without cholesteatoma in the screening set. Targeted metabolomics confirmed that levels of azobenzene and marimastat in the validation set exhibited trends similar to those observed in the non‐targeted analysis. Azobenzene and marimastat were found to be associated with the differences between CSOM with and without cholesteatoma, as well as with bone erosion in the middle ear. This study identified novel potential metabolic pathways and metabolites, providing insights into their possible roles in the inflammatory processes and bone erosion associated with CSOM and cholesteatoma.
- 30 July 2025
- Published by Wiley in Smart Medicine
- Vol. 4 (3)
Open AccessIdentification of Differential Metabolites in Chronic Suppurative Otitis Media With Non‐Targeted and Targeted Metabolomics Approach
Chronic suppurative otitis media (CSOM) is a leading cause of hearing loss and otorrhea, and when associated with cholesteatoma, it can pose a serious threat to patients' lives. This study aims to identify differences in tissue metabolites between patients with CSOM, both with and without cholesteatoma. Metabolomic profiles were measured in tissue samples from 42 surgically treated CSOM patients (35 with cholesteatoma, 7 without cholesteatoma). Significantly altered metabolites associated with CSOM were identified using a non‐targeted metabolomics approach and a targeted metabolomics approach. The 42 patients were divided into screening and validation sets. The non‐targeted analysis revealed 484 distinct differential metabolites and 32 metabolic pathways that differed between CSOM with and without cholesteatoma in the screening set. Targeted metabolomics confirmed that levels of azobenzene and marimastat in the validation set exhibited trends similar to those observed in the non‐targeted analysis. Azobenzene and marimastat were found to be associated with the differences between CSOM with and without cholesteatoma, as well as with bone erosion in the middle ear. This study identified novel potential metabolic pathways and metabolites, providing insights into their possible roles in the inflammatory processes and bone erosion associated with CSOM and cholesteatoma.
- Open Access
- 1 July 2025
- Published by Wiley in Advanced Science
- Vol. 12 (36) , e04184
SF3A3 Drives Tumorigenesis in Endometrial Cancer by Enhancing c‐FOS Expression and Represents a Potential Therapeutic Target
Aberrant alternative splicing plays a crucial role in tumorigenesis. Here, splicing factor 3A subunit 3 (SF3A3) is significantly upregulated in endometrial cancer (EC) tissues and associated with poor prognosis. Functionally, SF3A3 drives tumor progression by promoting cell proliferation, suppressing apoptosis, and enhancing cisplatin resistance in vitro and in vivo. Mechanistically, SF3A3 regulates alternative splicing of fos proto‐oncogene, AP‐1 transcription factor subunit (c‐FOS), an oncogene linked to chemoresistance, resulting in a ≈2‐fold increase in full‐length c‐FOS expression and activation of downstream anti‐apoptotic pathways. Notably, phenylethyl isothiocyanate (PEITC) as a direct inhibitor of SF3A3 through database screening and biophysical validation via surface plasmon resonance and mass spectrometry is identified. PEITC reduces c‐FOS expression and induces apoptosis in EC cells. Moreover, encapsulating PEITC in a hydrogel delivery system significantly enhances its therapeutic efficacy by enabling controlled release, reducing dosing frequency, and improving clinical applicability. The therapeutic potential of SF3A3 inhibition is further validated using patient‐derived tumor‐like cell clusters (PTCs), where PEITC and the c‐FOS inhibitor T‐5224 exhibit synergistic effects in suppressing EC. Collectively, our findings establish SF3A3 as a novel oncogenic regulator in EC and highlight PEITC, particularly in its hydrogel formulation, as a promising therapeutic strategy for improving clinical outcomes in EC patients.
- 1 July 2025
- Published by Wiley in Advanced Science
- Vol. 12 (36) , e04184
Open AccessSF3A3 Drives Tumorigenesis in Endometrial Cancer by Enhancing c‐FOS Expression and Represents a Potential Therapeutic Target
Aberrant alternative splicing plays a crucial role in tumorigenesis. Here, splicing factor 3A subunit 3 (SF3A3) is significantly upregulated in endometrial cancer (EC) tissues and associated with poor prognosis. Functionally, SF3A3 drives tumor progression by promoting cell proliferation, suppressing apoptosis, and enhancing cisplatin resistance in vitro and in vivo. Mechanistically, SF3A3 regulates alternative splicing of fos proto‐oncogene, AP‐1 transcription factor subunit (c‐FOS), an oncogene linked to chemoresistance, resulting in a ≈2‐fold increase in full‐length c‐FOS expression and activation of downstream anti‐apoptotic pathways. Notably, phenylethyl isothiocyanate (PEITC) as a direct inhibitor of SF3A3 through database screening and biophysical validation via surface plasmon resonance and mass spectrometry is identified. PEITC reduces c‐FOS expression and induces apoptosis in EC cells. Moreover, encapsulating PEITC in a hydrogel delivery system significantly enhances its therapeutic efficacy by enabling controlled release, reducing dosing frequency, and improving clinical applicability. The therapeutic potential of SF3A3 inhibition is further validated using patient‐derived tumor‐like cell clusters (PTCs), where PEITC and the c‐FOS inhibitor T‐5224 exhibit synergistic effects in suppressing EC. Collectively, our findings establish SF3A3 as a novel oncogenic regulator in EC and highlight PEITC, particularly in its hydrogel formulation, as a promising therapeutic strategy for improving clinical outcomes in EC patients.