Back

Identification of an epigenetically and phenotypically distinct peritumoral glioblastoma cell population linked to inferior patient outcome

Neves, I.; Lu, X.; Jokinen, V.; Latini, F.; Maturi, N.; Sundstrom, A.; Dang, Y.; Uppman, I.; Bergstrom, T.; Xing, P.; Ryttlefors, M.; Chen, X.; Swartling, F. J.; Uhrbom, L.

2026-01-17 cancer biology
10.64898/2026.01.16.699862 bioRxiv
Show abstract

Glioblastoma (GB) is an aggressive and therapy-resistant primary brain tumor with dismal prognosis. Lethality is in most patients caused by a local peritumoral (Edge) relapse near the resection cavity. This region is insufficiently studied and experimental models are scarce. We have analyzed matched tissue samples and cell cultures from bulk tumors and Edge regions of 11 GB patients. Genomic profiling displayed similar genetic alterations and subclonal distributions of matched bulk and Edge regions. Functional, phenotypic and combined single nucleus (sn) RNA-seq and ATAC-seq analyses showed distinct differences between bulk and Edge GB cells with similar shifts across patients. Sn multiomics uncovered a subset of Edge cells defined by unique chromatin accessibility signatures and elevated mesenchymal and immune gene expression. Integrated computational and functional investigations converged on microglia-derived Oncostatin-M as a driver of the Edge-unique phenotype, and high expression of Edge-related signatures correlated with inferior patient outcome in independent GB cohorts.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

1
Nature Communications
○ 5641 papers in training set
Top 3%
30.2%
Similar papers in this journal
2
Nature
○ 645 papers in training set
Top 3%
6.1%
Similar papers in this journal
3
Cancer Discovery
● 66 papers in training set
Top 0.5%
5.3%
Similar papers in this journal
4
Nature Genetics
○ 286 papers in training set
Top 1%
4.7%
Similar papers in this journal
5
Communications Biology
○ 993 papers in training set
Top 6%
3.2%
Similar papers in this journal
6
Nature Cancer
○ 39 papers in training set
Top 0.6%
3.1%
50% of probability mass above
7
npj Precision Oncology
○ 53 papers in training set
Top 0.5%
3.1%
8
Cell Reports
○ 1498 papers in training set
Top 13%
3.0%
9
Cancer Research
● 130 papers in training set
Top 1%
2.7%
10
Nature Cell Biology
○ 118 papers in training set
Top 1%
2.6%
11
Genome Medicine
○ 183 papers in training set
Top 2%
2.3%
12
eLife
● 5828 papers in training set
Top 44%
2.3%
13
Science Advances
● 1243 papers in training set
Top 21%
1.7%
14
Cell Reports Medicine
○ 153 papers in training set
Top 3%
1.5%
15
Genes & Development
● 90 papers in training set
Top 0.9%
1.5%
16
Genome Biology
○ 637 papers in training set
Top 6%
1.4%
17
Neuro-Oncology
◐ 36 papers in training set
Top 0.5%
1.1%
18
Cancer Cell
○ 42 papers in training set
Top 1%
1.1%
19
Journal of Clinical Investigation
● 179 papers in training set
Top 5%
1.0%
20
Cell Genomics
○ 172 papers in training set
Top 4%
1.0%
21
iScience
○ 1154 papers in training set
Top 38%
0.8%
22
The EMBO Journal
○ 309 papers in training set
Top 7%
0.8%
23
Nature Methods
○ 385 papers in training set
Top 7%
0.8%
24
Scientific Reports
○ 3612 papers in training set
Top 78%
0.6%
25
Proceedings of the National Academy of Sciences
● 2444 papers in training set
Top 46%
0.6%

"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.