ORIGINAL RESEARCH article
Front. Immunol.
Sec. Cancer Immunity and Immunotherapy
This article is part of the Research TopicFormation and Remodeling of Immunological Niches in Tumors: Organ-Specific Mechanisms and Inflammatory Parallels: Volume IIView all 19 articles
Liquid-liquid phase separation-driven molecular subtyping and prognostic modeling in colorectal cancer
Provisionally accepted
Hui Guan1
Chengzi Tian2
Jiawen Lin3Lihuan Zhang4
Run Shi5Wenjing Wang6Shuping Li7
Yuan Sui8Yanwen Lu4Tianjiao Cui3
Duo Chen4*- 1First Affiliated Hospital of Zhengzhou University, Zhengzhou, China
- 2The First People's Hospital of Yunnan Province, Kunming, China
- 3Sun Yat-Sen University, Guangzhou, China
- 4The First Affiliated Hospital of Zhengzhou University, Zhengzhou, China
- 5The First Affiliated Hospital With Nanjing Medical University, Nanjing, China
- 6Southern University of Science and Technology, Shenzhen, China
- 7The University of Oklahoma Health Sciences, Oklahoma City, United States
- 8Salk Institute for Biological Studies, La Jolla, United States
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
Background: Liquid–liquid phase separation (LLPS) orchestrates the spatiotemporal organization of biomolecular condensates and regulates numerous biological processes. However, the extent to which dysregulated LLPS facilitates the progression of colorectal cancer (CRC) has not been elucidated. Elucidating how LLPS influences CRC possibly offers valuable insights into diagnosis and therapeutic intervention. Methods: Differentially expressed genes (DEGs) were identified from 566 CRC samples and 19 normal controls in the GSE39582 dataset. LLPS-linked genes were collected from the DrLLPS database. Prognostically significant genes were identified via univariate Cox regression, least absolute shrinkage and selection operator regression, and stepwise akaike information criterion algorithm. The risk score was derived utilizing the LLPS-linked gene signature. Patient characteristics were evaluated concerning the computed risk scores. The biological and clnical distinctions across high-risk and low-risk cohorts were further investigated, leveraging the COAD, READ, and GSE17536 validation cohorts. The expression and spatial distribution of the five prognostic genes were examined via the GSE166555 dataset and spatial transcriptomics analysis. The hydroxyacyl-coenzyme A dehydrogenase (HADH) expression-related enrichment pathways were further analysed via WGCNA combined with Metascape. The expression and biological functions of HADH were verified in vitro. Results: A total of 430 LLPS-related DEGs were identified, from which five prognostic genes were selected to construct the LLPS-associated risk signature. Marked differences in gene expression profiles, overall prognosis, clinicopathological attributes, somatic mutations, signaling pathway activity, tumor microenvironment composition, and drug sensitivity were noted across the high-risk and low-risk populations. Furthermore, the
Keywords: colorectal cancer, Liquid-liquid phase separation, prognosis, Spatial transcriptomics, Tumor immune microenvironment
Received: 08 Nov 2025; Accepted: 11 Dec 2025.
Copyright: © 2025 Guan, Tian, Lin, Zhang, Shi, Wang, Li, Sui, Lu, Cui and Chen. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence: Duo Chen
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.