EventsThe 5th International Electronic Conference on Cancers
Published
This submission belongs to the session S7. Novel Methods and Technologies for Research and Treatment of the event The 5th International Electronic Conference on Cancers
Published date
05 Jun, 2026
Academic Editor
author-avatarAndrew A. Gumbs
Citation
Ekaterina Lindunen, Margarita Pustovalova, Sergey Leonov, Yulia Merkher, Bioinformatic Identification of ACTN4 as a Candidate Master Regulator of Local Invasion in Glioblastoma , in Proceedings of The 5th International Electronic Conference on Cancers, 10 June–12 June 2026, MDPI: Basel, Switzerland
Share
Email
Facebook
Twitter
LinkedIn

Bioinformatic Identification of ACTN4 as a Candidate Master Regulator of Local Invasion in Glioblastoma

image
image
1. Institute of Future Biophysics, Moscow Center for Advanced Studies, 123592 Moscow, Russia, Russia
2. Department of Biomedical Engineering, Technion - Israel Institute of Technology, Haifa, 3200003, Israel
Abstract

Introduction

Glioblastoma is the most aggressive brain tumor. The main reason for its lethality is the tumor's broad invasion into brain tissue, which is influenced by tumor cell mechanics. Successful invasion necessitates dynamic cytoskeletal remodeling, altered cell adhesion, and interaction with neural structures. We hypothesized that previously defined, focused panels of mechanobiological markers capturing these functions would elucidate key drivers of glioblastoma invasiveness.

Materials and Methods

mRNA expression data were obtained from cBioPortal (TCGA, PanCancer Atlas) for 592 primary glioblastoma and 514 low-grade gliomas samples. We assessed 11 genes related to actin dynamics, cell movement, adhesion, and mechanotransduction: ACTN4, PFN1, CFL1, FSCN1, ANXA2, ANXA6, MYH9, GSN, TAGLN, ECM1, and EZR. Differential expressions between glioblastoma and low-grade glioma were assessed using the non-parametric Mann–Whitney U-test. The diagnostic power of individual genes and the composite panel score was measured using the Area Under the ROC Curve (AUC).

Results

Glioblastoma demonstrated significant dysregulation of 4 out of 11 mechanobiological genes compared to low-grade gliomas (p < 0.01): FCN1 (immune mediator, ECM), TAGLN (actin-binding protein, contractility), ACTN4 (actin crosslinking, migration), and MYH9 (non-muscle myosin II, tension generation). The composite score of the panel showed a connection to tumor grade (AUC=0.67), while the score from 4 dysregulated genes was notably better (AUC=0.73). ACTN4 emerged as the top diagnostic biomarker (AUC=0.83) in the single-gene analysis, which points to the significance of actin cytoskeleton regulation in brain invasion.

Conclusions

The four-gene mechanobiological signature effectively embodies the vital traits associated with glioma aggressiveness. ACTN4 has emerged as a promising master diagnostic biomarker and a potential key molecular driver behind local invasion in glioblastoma. To establish its significance, validating its function in vitro and in vivo glioblastoma invasion models is crucial.

Keywords
Glioblastoma
ACTN4
Mechanobiology
Actin cytoskeleton
Gene markers
TCGA
Evaluating Agentic AI for Autonomous Hypothesis Generation in Clinical Datasets
IN SILICO PHARMACOKINETIC PROFILING AND MOLECULAR DOCKING OF MARGOLONONE AGAINST THE ONCOGENIC IDH1-R132H MUTANT