Pan-Cancer Metabolic Signature Predicts Co-Dependency on Glutaminase and De Novo Glutathione Synthesis Linked to a High-Mesenchymal Cell State

The enzyme glutaminase (GLS1) is currently in clinical trials for oncology, yet there are no clear diagnostic criteria to identify responders. The evaluation of 25 basal breast lines expressing GLS1, predominantly through its splice isoform GAC, demonstrated that only GLS1-dependent basal B lines required it for maintaining de novo glutathione synthesis in addition to mitochondrial bioenergetics. Drug sensitivity profiling of 407 tumor lines with GLS1 and gamma-glutamylcysteine synthetase (GCS) inhibitors revealed a high degree of co-dependency on both enzymes across indications, suggesting that redox balance is a key function of GLS1 in tumors. To leverage these findings, we derived a pan-cancer metabolic signature predictive of GLS1/GCS co-dependency and validated it in vivo using four lung patient-derived xenograft models, revealing the additional requirement for expression of GAC above a threshold (log₂RPKM + 1 ≥ 4.5, where RPKM is reads per kilobase per million mapped reads). Analysis of the pan-TCGA dataset with our signature identified multiple indications, including mesenchymal tumors, as putative responders to GLS1 inhibitors. Daemen et al. address the current unmet need for diagnosing responders to glutaminase (GLS1) cancer therapy. They identify co-dependency of GLS1 and gamma-glutamylcysteine synthetase (GCS) inhibitors, suggesting that redox balance is a key function of GLS1 in tumors, and validate a predictive pan-cancer metabolic signature for GLS1/GCS co-dependency in vivo.