Cells

Pancreatic cancer is a highly intractable malignancy that necessitates personalized treatment strategies. Conventional patient-derived models, such as three-dimensional organoids, are often limited by intellectual property constraints and high costs. In this study, we developed an affordable adherent culture system for patient-derived pancreatic cancer cells using a proprietary medium and laminin-coated dishes. Primary cultures were successfully established from 28 patients with pancreatic ductal adenocarcinoma, exceeding a 90% success rate. Validation of eight samples confirmed maintenance of epithelial cell adhesion molecule expression and preservation of oncogenic KRAS mutations. Transcriptomic profiling revealed consistent upregulation of a six-gene signature (FAP, IGFBP5, PRRX1, SPARC, WNT5A, and ADAMTS12), which is associated with malignancy. In vitro drug sensitivity assays revealed interpatient heterogeneity with preliminary clinical associations. In conclusion, this simplified platform provides high-purity cancer cells and serves as a functional precision medicine tool. Beyond conventional chemotherapy, this platform has the potential to support applications ranging from biomarker validation and exploratory preclinical testing of novel therapeutics, including immune checkpoint inhibitors and antibody–drug conjugates. This optimization can lead to personalized therapeutic strategies for pancreatic cancer.

Human sepsis is a complex disease that manifests with a diverse range of phenotypes and inherent variability among individuals, making it hard to develop a comprehensive animal model. Despite this difficulty, numerous animal models have been developed that capture many key aspects of human sepsis. Though the animal models have contributed to the fundamental advances in understanding the pathogenesis of septic patients, the translational value of these models has been constantly questioned because many clinical trials of targeted therapies based on the advances in animal models have failed, highlighting the urgent need for developing new research models or refining previous animal models for sepsis research. In this review, we will summarize recent advances in new and emerging research models for sepsis, including human-based in vitro systems, highly tailored animal models, AI and digital models analyzing vast datasets to define patient subgroups and predict outcomes, and the FAMOUS framework ensuring that therapies are tested against the specific mechanism they are designed to target. We will discuss the strengths and limitations of these models, reflecting the clinical course of sepsis, and discuss the future directions in this subject area.

Liquid biopsy (LB) offers a minimally invasive approach to characterizing and monitoring glioblastoma (GB), a tumor marked by extensive heterogeneity, limited surgical accessibility and rapid molecular evolution. By analyzing circulating tumor-derived components such as circulating tumor DNA (ctDNA), extracellular vesicles (EVs), circulating RNA species and circulating tumor cells (CTC), LB provides dynamic molecular information that cannot be captured by neuroimaging or single-site tissue sampling. Cerebrospinal fluid (CSF) currently yields the highest sensitivity for detecting tumor-specific alterations, while plasma enables repeat monitoring despite lower biomarker abundance. EVs have gained particular prominence due to their ability to preserve DNA, RNA, and protein cargo that reflects key genomic changes, treatment resistance mechanisms, and immune evasion. Although advances are substantial, clinical implementation remains constrained by low analyte concentrations, methodological variability, limited standardization and the high cost of testing, which is rarely reimbursed by insurers. This review summarizes current evidence on circulating biomarkers in GB and highlights research priorities essential for integrating LB into future diagnostic and therapeutic workflows.