Yunqing Qiu

Background: Adult-onset Still’s disease (AOSD) is a systemic autoinflammatory disorder lacking a gold-standard diagnostic criterion. Objective: To develop and validate a clinically applicable model for identifying AOSD among patients with fever of unknown origin (FUO) who have clinical suspicion for AOSD. Methods: Clinical data (2010–2020) were divided into training and internal test set (7:3) using stratified random sampling according to disease status (AOSD vs non-AOSD). Feature selection was performed using Boruta, recursive feature elimination and least absolute shrinkage and selection operator algorithms. Selected features were used to train logistic regression (LR), random forest and extreme gradient boosting models with fivefold cross-validation. Model performance was evaluated using area under the curve (AUC), receiver operating characteristic curves, sensitivity, specificity and accuracy. External validation was performed at another centre using the same adjudication procedure. Results: A total of 847 patients were included, comprising a derivation cohort of 771 patients and an independent external validation cohort of 75 patients. Six features—age, neutrophil percentage, white blood cell count, infection indicator, ferritin and ‘AOSD-related clinical presentation score’—were consistently selected by at least two algorithms and used to build the model. LR achieved the highest AUC in both training (0.969; 95% CI 0.956 to 0.983) and test sets (0.960; 95% CI 0.934 to 0.985). A nomogram based on the LR model demonstrated good real-world performance in the independent validation cohort, with an AUC of 0.906. Conclusions: We developed a machine learning model using electronic medical records to identify AOSD among patients with FUO with clinical suspicion for AOSD. The model shows high accuracy and potential for early identification of AOSD in this specific clinical setting.

Current mRNA-LNP systems face challenges in efficient co-delivery of nucleic acid adjuvants, limiting their capacity to elicit robust cellular immunity. To

Liver cirrhosis remains a major global health challenge, with liver transplantation currently representing the most effective treatment. Elucidating its molecular pathogenesis is therefore critical for the development of novel therapeutic strategies. Emerging evidence indicates that mitophagy plays a pivotal role in cirrhosis progression. In this study, we employed integrative bioinformatics to explore the association between mitophagy and liver cirrhosis, aiming to identify potential therapeutic targets. Differentially expressed genes (DEGs) were obtained from GSE77627 and GSE139602 data sets, followed by functional enrichment analysis. Mitophagy-related differentially expressed genes (mito-DEGs) were screened and assessed through receiver operating characteristic (ROC) analysis, immune cell infiltration profiling, and protein–protein interaction (PPI). Weighted gene coexpression network analysis (WGCNA) identified key modules, and intersecting genes with mito-DEGs highlighted RAB7B as a candidate hub gene, which was validated in external data sets. Experimental verification confirmed elevated RAB7B expression in activated hepatic stellate cells (HSCs) and a mouse model of liver cirrhosis. Functional assays demonstrated that RAB7B inhibition attenuated HSC activation, and molecular docking revealed strong binding affinity between RAB7B and predicted anticirrhosis compounds. We found that RAB7B knockdown restored TGF-β-induced mitophagy inhibition, enhanced mitochondria-lysosome colocalization, and showed dynamic regulation by mitophagy status, indicating its role as a negative and responsive regulator of mitochondrial clearance. Collectively, our findings identify RAB7B as a mitophagy-related hub gene driving liver cirrhosis progression and provide novel insights into its therapeutic potential.

Voriconazole, a first-line therapy for invasive aspergillosis, carries hepatotoxicity risks lacking comprehensive epidemiological characterization. Current evidence is limited by heterogeneity in diagnostic criteria (e.g., varying thresholds for liver function abnormalities across studies), insufficient sample sizes, and inadequate adjustment for comorbidities, which compromises risk assessment accuracy. This study aims to systematically characterize the epidemiological profile of voriconazole-induced liver injury and identify modifiable independent risk factors. We analyzed 7,659 voriconazole-treated adults (2007–2020) from the First Affiliated Hospital of Zhejiang University to determine incidence and injury subtypes. Univariable and multivariable logistic regression identified risk factors, with model performance evaluated by ROC (receiver-operating characteristic) analysis; sensitivity analyses addressed potential confounders. In our results, voriconazole-induced liver injury occurred in 11.32% of patients, predominantly manifesting as hepatocellular (5.01%, 279/630), cholestatic (5.19%, 289/630), and mixed-pattern injuries (1.11%, 62/630). Key modifiable risk factors included intravenous administration (odds ratio [OR] = 1.88, 95% confidence interval [CI]: 1.43–2.46). Non-modifiable predictors comprised baseline elevated urea (OR = 1.47, 95% CI: 1.08–2.02), elevated triglycerides (TG) (cholestatic injury: OR = 1.47, 95% CI: 1.08–2.02), elevated C-reactive protein (CRP) (cholestatic injury: OR = 1.47, 95% CI: 1.08–2.02), pre-existing liver disease (liver cirrhosis, OR = 2.16, 95% CI: 1.58–2.96), diabetes (OR = 2.19, 95% CI: 1.70–2.80), and transplant status (hematopoietic stem cell transplantation, OR = 2.93, 95% CI: 1.85–4.65). Using this clinically substantial cohort ( n =7,659), we proposed a risk-stratification framework incorporating administration route optimization and comorbidity-specific monitoring, supporting personalized risk-benefit analysis in antifungal therapy.

Background: ASC10, an oral double prodrug of the antiviral ribonucleoside analog ASC10-A (also referred to as NHC), is currently in clinical trials for the treatment of COVID-19. Upon administration, ASC10 undergoes rapid biotransformation into the monoprodrug molnupiravir, which then swiftly converts to the active metabolite ASC10-A. Alternatively, ASC10 can directly transform into ASC10-A without forming molnupiravir as an intermediate. This study aimed to describe the population pharmacokinetics (popPK) of ASC10-A in healthy Chinese subjects and to inform clinical drug development. Methods: We performed popPK modeling for ASC10-A using data from a Phase I clinical trial involving 57 healthy subjects and 1634 observations. The M3 method was employed to handle data below the quantification limit (BQL), while the Laplacian algorithm was implemented for popPK modeling. The final model was evaluated through goodness-of-fit (GOF) plots, non-parametric bootstrap method, and visual predictive check (VPC). The model-based simulations were performed to generate concentration-time profiles for ASC-10 or NHC across different studies, evaluate the exposure levels of ASC10-A under an 800 mg twice-daily dosing regimen (consistent with the recommended dose of molnupiravir), and quantify the covariate effects on the pharmacokinetic parameters of ASC10-A. Results: The ASC10-A pharmacokinetics was described using a two-compartment model with first-order elimination and transit compartment absorption. Food intake and body weight were identified as influential variables on ASC10-A pharmacokinetics. The simulation results revealed that the ASC10-A exposure decreased with increasing body weight, and participants fasting could result in a higher peak concentration while retaining a similar exposure to those fed. The ASC10 dosing regimen of 800 mg every 12 hours could provide desirable exposure associated with clinical response. Conclusion: This study provides a comprehensive population pharmacokinetic profile of ASC10-A in healthy Chinese volunteers. These findings support the clinical development and identification of effective regimens for ASC10.