Glioblastoma (GBM) is among the most aggressive brain cancers, driven by genetic diversity and resistance to therapy. Mitochondrial metabolism-and in particular the electron transport chain (ETC)-has emerged as both a key weakness and a source of variable drug response. To investigate this, we integrated constraint-based metabolic modeling (CBM), high-resolution drug profiling, and genomic sequencing across three GBM cell models: LN229, U87MG, and neurospheres (NSP). Modeling predicted distinct ETC vulnerabilities, which were confirmed experimentally using inhibitors against Complexes I-V. Sensitivity to rotenone varied sharply: NSP cells were most vulnerable (IC50 = 0.007 mu M), LN229 showed intermediate sensitivity (0.021 mu M), and U87MG remained highly resistant (1.816 mu M). Across inhibitors, LN229 consistently showed steep dose-response slopes, U87MG maintained flat curves, and NSP displayed selective weaknesses. By incorporating slope (m) and Instantaneous Inhibitory Potential (IIP), median-effect analysis captured dynamic drug-response behaviour's that IC50 values alone overlooked. Genomic sequencing revealed striking differences in mutational burden: U87MG and NSP carried 354 and 307 single nucleotide polymorphisms (SNPs), respectively, compared with 141 in LN229. Several non-synonymous mutations were directly linked to altered drug sensitivity, including L194S, Y50 N, and L46V in LN229; S456L, A466T, and Y629F in U87MG; and the NSP-specific R159Q. Notably, mutations near catalytic sites correlated with changes in slope and IIP, providing mechanistic insight into therapeutic response. Together, these results show how genetic variation reshapes ETC function and drug sensitivity in GBM, offering a predictive framework for mutation-informed, personalized therapy.

Foreign