Back

Establishment of human glioblastoma cell culture collection

Denis, H.; Merone, J.; Watters, V.; Fort, V.; Khelifi, G.; Lavoie, M.; Berthiaume, L.; Rondeau, F.; Beaumont, E.; Michaud, K.; Dastous, M.; Saikali, S.; Huot, M.-E.; Audet-Walsh, E.; Richer, M.; Hussein, S. M. I.

2025-09-10 cancer biology
10.1101/2025.09.05.674480 bioRxiv
Show abstract

Glioblastoma (GBM) is a highly aggressive primary brain cancer with poor prognosis (<15 months), highlighting the urgent need for more effective therapies. As current treatments are not effective, the need for a deeper understanding of the biology of GBM cells, including how they reprogram their metabolism to support their aberrant and uncontrolled growth, is critical. To this end, we established a collection of 41 human glioma cell lines derived from freshly resected tumour tissues from 99 patients. We characterized 12 of these cell lines by combining histologic, genetic, stem cell derivation and self-renewal, and metabolomic analyses. Histological and genetic profiles included IDH mutation status, Ki-67 proliferation index, ATRX status, mutant TP53 expression, chromosome 10q loss, EGFR amplification, and MGMT promoter methylation. Of these, only p53 mutation expression status showed weak segregation of the cell lines into 2 separate metabolic groups based on amino acid levels, but none showed an effect on stem cell derivation or self-renewal. Further characterization of these 12 cell lines revealed significant metabolic and phenotypic differences when comparing mesenchymal versus proneural gene expression subtyping. We show significant increases in TCA cycle metabolites in mesenchymal-like GBM cells and higher overall metabolic activity compared to proneural-like cells. These findings highlight the complexity of GBM and the need for personalized treatments that consider the metabolome of each subtype as a potential therapeutic avenue.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.