Malignant pleural mesothelioma (MPM) is a rare but aggressive thoracic malignancy primarily linked to asbestos exposure. Despite advances in research, the prognosis of MPM remains poor, and there is a lack of efficient and precise prognostic assessment tools. Its insidious onset, limited treatment options, and high resistance to therapy contribute to poor clinical outcomes and underscore the urgent need for reliable prognostic biomarkers. To address this gap, we integrated multi-omics data to identify a prognostic gene set for MPM and subsequently developed a prognostic model aiming to improve the clinical outcomes of this disease. Transcriptomic data from TCGA were used to identify genes related to prognosis and tumor stage in malignant pleural mesothelioma (MPM), followed by construction of a prognostic model via LASSO-Cox regression and external validation using a GEO dataset. The model was then integrated with DNA methylation and SNP data for GO enrichment analysis. Chemotherapy drug IC50 data from MPM cell lines were correlated with model gene expression to evaluate associations between risk scores and drug sensitivity, which were further validated using TCGA clinical response data. Finally, the model’s impact on the immune microenvironment was assessed using single-cell RNA-seq data, and its predictive value for immunotherapy response was validated in an independent MPM immunotherapy cohort. Our novel prognostic model demonstrated consistent performance in both the training and validation cohorts. Patients with high-risk scores had poorer outcomes, with AUC values exceeding 0.8 and reaching 0.9 for 3 year survival prediction. The risk score accurately reflected biological processes such as tumor proliferation and metastasis in mesothelioma. It was also closely associated with clinical responses to radiotherapy and chemotherapy, with low-risk patients showing greater sensitivity and lower drug IC50 values. Additionally, the risk score correlated positively with tumor immune activity and was predictive of immunotherapy response. In conclusion, our prognostic model shows strong and reliable predictive power for patient survival and treatment response in malignant pleural mesothelioma. It not only reflects key tumor biological processes but also serves as a potential tool for guiding personalized therapy, including chemotherapy, radiotherapy, and immunotherapy.

Background: Microplastics are heterogeneous environmental contaminants whose transport, persistence, and potential biological interactions may depend on particle morphology, size, geometry, and polymer composition. Environmental monitoring studies commonly describe particle occurrence within individual matrices, whereas experimental studies frequently use standardized particles that may not adequately represent environmentally observed configurations. Objective: This study aimed to characterize microplastic morphology, dimensional characteristics, size distribution, and polymer composition in drinking-water and estuarine datasets and to identify environmentally observed particle profiles that may inform future musculoskeletal research. Methods: Two publicly available environmental datasets were analyzed separately. Overall morphology was summarized across 44,221 San Francisco Bay particle records with valid classifications. For matrix-level analyses, quality-assurance samples were excluded, and sample-level morphology proportions were compared using permutational multivariate analysis of variance based on Bray–Curtis dissimilarities with 9,999 permutations. Homogeneity of multivariate dispersion was assessed using PERMDISP. Particle length and aspect ratio were analyzed using generalized estimating equations accounting for particles clustered within samples and adjusted for environmental matrix. Drinking-water records were analyzed descriptively because of heterogeneity in observational units and analytical methods. Results: In the complete archived particle inventory, fibers were the most frequently recorded morphology (49.8%, n = 22,012), followed by fragments (40.3%, n = 17,804). Among 266 environmental or biological samples retained for inferential analysis after exclusion of quality-assurance samples, sample-level morphology profiles differed significantly among matrices (PERMANOVA pseudo- F = 34.14, R2 = 0.344, p < 0.001), although multivariate dispersion also differed (PERMDISP F = 94.31, p < 0.001). In cluster-adjusted models, fibers were approximately 2.51 times longer than fragments (95% CI, 2.35–2.68) and had 24.76 times their aspect ratio (95% CI, 22.70–27.01). Conclusion: Environmental microplastics occurred as heterogeneous combinations of fibers, fragments, particle sizes, and polymer categories. These environmentally observed combinations of morphology, dimensions, and material identity may provide empirically grounded parameters for selecting more representative particles in future cartilage-, synovium-, and bone-related experimental studies. However, this study does not measure human intake, internal exposure, tissue accumulation, biological responses, or musculoskeletal outcomes and therefore does not establish toxicological risk, clinical association, or causality.