Patient-derived glioblastoma cultures preserve respiration phenotypes during ex vivo maintenance and show sex-associated differences in migration.

Glioblastoma (GBM) is a highly heterogeneous, invasive brain tumor with profound metabolic plasticity. Patient-derived cultures provide valuable preclinical models, yet the stability and clinical relevance of their metabolic phenotypes during ex vivo maintenance remain unclear. Here, we performed longitudinal functional profiling of mitochondrial respiration, substrate utilization, and migration in primary patient-derived GBM cultures at one and five weeks ex vivo using Seahorse Bioanalyzer assays and wound-healing assays, complemented by exploratory integration with routine clinical parameters (MGMT promoter methylation, p53 expression, sex, and age). Early profiling identified two mitochondrial subgroups distinguished by spare and maximal respiratory capacity, with MGMT promoter methylation observed exclusively in the high-respiration cluster. Over time, most cultures retained their initial phenotype, whereas a subset transitioned from low to high respiration, indicating dynamic adaptation under standardized conditions. Extended culture was accompanied by patient-specific shifts in fuel utilization, including increased glutamine dependency and reduced inter-sample variability in glucose and fatty-acid parameters. Exploratory analyses suggested an association between higher age and lower fatty-acid oxidation capacity and indicated constrained fatty-acid dependency in p53-positive cultures. Migration assays revealed marked inter-sample heterogeneity and a 2-6 times faster gap closure in male-derived than female-derived cultures. Notably, early migratory behavior showed no strong association with OCR-defined metabolic state, and early fuel-flex parameters did not recapitulate OCR cluster structure. Together, these data provide a standardized longitudinal functional benchmark showing that patient-derived GBM cultures preserve intrinsic metabolic heterogeneity while undergoing time-dependent adaptation, and that migration represents an additional, partially independent functional axis. These findings are hypothesis-generating and support stratified, time-sensitive follow-up studies of metabolic and invasive phenotypes in GBM.