Cell autophagy shows a two-way effect in the pathogenesis and development of cancer. In the early stage of cancer, autophagy (the process of cell degradation of cell components) has the function of tumor suppression by degrading the wrongly coded cell components. In the later stage of cancer, autophagy has a protective effect on tumor cells, allowing cells to survive, invade and be resistant to treatment after exposure to initial drug treatment. The signaling pathway that regulates autophagy is strictly regulated. Inhibition measures include Akt (protein kinase B) and mTOR (mammalian target of rapamycin) pathways involving inhibiting autophagy, and the AMPK/mTOR pathway makes autophagy possible. The aberration of Beclin1 (loss of expression or mutation) is an important regulator of autophagy, which is associated with malignant tumors, including breast cancer and ovarian cancer. Autophagy not only regulates cell degradation but also regulates the progression of tumors through the synergy of immune regulation, cell metabolism, tumor microenvironment remodeling or lysosomal activity. The priority or targeting of regulatory networks involving autophagy - especially the AMPK/mTOR pathway or Beclin1 signal is an emerging strategy for low-toxic biomarker therapy. However, they are worth considering whether the context comes from the tumor stage and/or histology. In summary, this paper evaluates the two-way effect of autophagy on tumor progression and treatment, including its role in the tumor microenvironment, and provides new insights for targeted therapy and biomarker treatment regulation.

Show more

The treatment of ocular cancer faces unique challenges. The complex physiological structures of the eye, such as the corneal barrier and the blood-eye barrier, restrict the efficiency of drug delivery. Traditional chemotherapy drugs, due to their low solubility and poor stability, not only have poor therapeutic effects but also are prone to cause side effects such as ocular irritation. Polymer-based delivery systems, with their controllable chemical structure, good biocompatibility and targeted delivery capabilities, have become key carriers for optimizing the delivery effect of drugs for ocular cancer. The current commonly used machine learning algorithms are difficult to meet the requirements of the refined assessment of the ocular cancer delivery system. Therefore, this paper proposes the LSTM-Adaboost classification algorithm. Firstly, violin graph analysis and correlation analysis are carried out, and then multiple machine learning algorithms are used for testing. The results show that the core evaluation indicators of this algorithm are generally superior to those of ExtraTrees, decision tree, GBDT, Random Forest, CatBoost, AdaBoost and XGBoost algorithms. Its accuracy rate and recall rate both reach 89.3%, and the precision rate and F1 value are both 89.2%. Compared with the suboptimal ExtraTrees algorithm, all indicators were 87.4%, increasing by 1.9, 1.9, 1.8, and 1.8 percentage points respectively. Compared with the decision tree algorithm, which has an accuracy rate and recall rate of 78%, an precision rate of 80%, and an F1 value of 78.7%, and the GBDT algorithm, which has an accuracy rate and recall rate of 81.1%, an precision rate of 81.7%, and an F1 value of 81.4%, its advantages are more significant. The AUC indicator is 94.9%, which is slightly lower than the 95.1% of the ExtraTrees algorithm, but higher than other algorithms. Overall, it still leads. This algorithm provides a reliable method for the refined evaluation of ocular cancer delivery systems and is of great significance for promoting the optimization of drug delivery technologies for ocular cancer treatment.

Show more

Chronic obstructive pulmonary disease (COPD) is recognized as one of the three foremost causes of mortality among populations throughout the world. For patients diagnosed with acute exacerbation of COPD (AECOPD), three core clinical features are commonly observed: bronchial spasm, mucus accumulation-induced obstruction, and insufficient compensatory function of the lungs. In clinical practice, airway management measures must abide by a core principle – prioritizing non-invasive over invasive interventions while placing respiratory muscle protection at the core – a standard developed to reduce airway irritation and alleviate related tissue injury. Current clinical practice guidelines support evidence-based pharmacological treatment for AECOPD, yet traditional single-modal intervention methods only bring about temporary symptom relief. These conventional approaches cannot fully address the disease's underlying pathological mechanisms, a limitation that often leads to gradual respiratory function decline and marked impairment of patients' overall quality of life. This review collates cutting-edge clinical studies focusing on multimodal airway management for AECOPD, aiming to break the cycle of over-reliance on single therapies by implementing standardized patient assessments and stratified interventions. Beyond this primary aim, the review also seeks to provide practical clinical care guidance for frontline clinicians, optimize patients' short- and long-term prognosis, and clarify key directions for future research on personalized treatment strategies.

Show more

Drug-induced liver injury (DILI) in adults aged 18–60 years remains understudied despite its clinical heterogeneity and rising incidence. This study aimed to characterize the epidemiology, risk factors, and drug-specific profiles of DILI in this demographic. Utilizing data from the FDA Adverse Event Reporting System (FAERS) (2007–2024), we analyzed 17,464 DILI cases. Four disproportionality methods—Reporting Odds Ratio (ROR), Proportional Reporting Ratio (PRR), Bayesian Confidence Propagation Neural Network (BCPNN), and Multi-item Gamma Poisson Shrinker (MGPS)—were employed to identify high-risk drugs. Time-to-onset (TTO) and gender-specific risks were assessed. Seventeen drugs exhibited significant DILI signals. Rifampicin (ROR=19.33, 95%CI: 16.41–22.77), amoxicillin/clavulanic acid (ROR=16.39, 95%CI: 14.48–18.56), and paracetamol (ROR=10.3,95%CI: 9.61–11.04) showed the strongest associations. Non-steroidal anti-inflammatory drugs (NSAIDs) had the shortest median time-to-onset (7 days). Gender subgroup analysis revealed sex-biased hepatotoxicity, with females disproportionately affected by immunosuppressants and males by antibiotics. This large-scale real-world analysis identifies NSAIDs, antibiotics, and immunosuppressants as critical hepatotoxic threats in younger adults. The findings advocate for targeted hepatic monitoring and updated drug labeling to reflect class-specific latency patterns. Gender-tailored risk mitigation strategies are warranted to address sex-based disparities in DILI susceptibility.

Show more