Skip to main content

ORIGINAL RESEARCH article

Front. Pharmacol.
Sec. Pharmacology of Anti-Cancer Drugs
Volume 16 - 2025 | doi: 10.3389/fphar.2025.1529525
This article is part of the Research Topic Decoding the Epigenetic Landscape: Elucidating Cancer Pathology and Identifying Novel Therapeutic Targets View all 13 articles

Development and Pan-cancer Validation of an Epigenetics-based Random Survival Forest Model for Prognosis Prediction and Drug Response in OS

Provisionally accepted
Chaoyi Yin Chaoyi Yin 1Kede Chi Kede Chi 2Zhiqing Chen Zhiqing Chen 1Shabin Zhuang Shabin Zhuang 1Yongsheng Ye Yongsheng Ye 1Binshan Zhang Binshan Zhang 1Cailiang Cai Cailiang Cai 1*
  • 1 Dongguan Hospital of Guangzhou University of Chinese Medicine, Dongguan, China
  • 2 Zhongshan Hospital of Traditional Chinese Medicine, Zhongshan, China

The final, formatted version of the article will be published soon.

    Background: Osteosarcoma (OS) exhibits significant epigenetic heterogeneity, yet its systematic characterization and clinical implications remain largely unexplored. Methods: We analyzed single-cell transcriptomes of five primary OS samples, identifying cell type-specific epigenetic features and their evolutionary trajectories. An epigenetics-based Random Survival Forest (RSF) model was constructed using 801 curated epigenetic factors and validated in multiple independent cohorts. Results: Our analysis revealed distinct epigenetic states in the OS microenvironment, with particular activity in OS cells and osteoclasts. The RSF model identified key predictive genes including OLFML2B, ACTB, and C1QB, and demonstrated broad applicability across multiple cancer types. Risk stratification analysis revealed distinct therapeutic response patterns, with low-risk groups showing enhanced sensitivity to traditional chemotherapy drugs while high-risk groups responded better to targeted therapies. Conclusion: Our epigenetics-based model demonstrates excellent prognostic accuracy (AUC>0.997 in internal validation, 0.832-0.929 in external cohorts) and provides a practical tool for treatment stratification. These findings establish a clinically applicable framework for personalized therapy selection in OS patients.

    Keywords: os, Epigenetic heterogeneity, single-cell RNA sequencing, Random survival forest, Prognostic model, drug sensitivity, Pan-cancer analysis

    Received: 17 Nov 2024; Accepted: 08 Jan 2025.

    Copyright: © 2025 Yin, Chi, Chen, Zhuang, Ye, Zhang and Cai. 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: Cailiang Cai, Dongguan Hospital of Guangzhou University of Chinese Medicine, Dongguan, China

    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.