Introduction: Tumor migration and invasion play key roles in the pathogenesis and prognosis of glioblastoma, the most common form of primary brain cancer in adults. This study aimed to develop a physics-based framework connecting to transcriptomic signature to predict patient-specific glioblastoma cell migration.
Methods: We applied physics-based modeling of a motor-clutch model, termed cell migration simulator (CMS), to parameterize the migration behaviors of glioblastoma cells and define physical biomarkers on a patient-by-patient basis.
Results: We reduced the eleven-dimensional parameter space of the CMS into three principal physical parameters that governed glioblastoma migration, including 1) motor number (defined by myosin II activity within the glioblastoma cell), 2) clutch number (defined by the number of contact points between the glioblastoma cell and the extracellular matrix), and 3) F-actin polymerization rate. Patient-derived (xenograft) (PD(X)) cell lines across the spectrum of mesenchymal (MES), proneural (PN), and classical (CL) subtypes (N=13 PDX’s) exhibited optimal motility and traction force on stiffnesses around 9.3kPa. When modeled under CMS parameterization, all 13 glioblastoma lines consistently displayed balanced motor/clutch ratios to enable effective migration, with MES cells exhibiting higher actin polymerization rates and motility. The CMS also predicted differential sensitivity to cytoskeletal drugs between patients. Finally, we identified an 18-gene signature that predicted the mechanics of glioblastoma cell migration and clinical glioblastoma survival in The Cancer Genome Atlas.
Conclusion : Our results provide a quantitative, physics-based framework for parameterizing glioblastoma migration and a gene signature that captures these parameters in the clinical setting. The findings are relevant to personalizing oncologic care and therapeutic advancement for glioblastoma patients.
.jpg)
