Transcriptome Analysis Identifies Novel Prognostic Genes in Osteosarcoma

Area of Science:

• Oncology
• Genomics
• Bioinformatics

Background:

• Osteosarcoma (OS) is a highly aggressive bone cancer primarily affecting young adults.
• Predicting prognosis in OS patients is challenging due to small sample sizes and limited public datasets.
• Advancements in high-throughput technologies enable the identification of prognostic biomarkers for cancer survival.

Purpose of the Study:

• To identify novel prognostic gene expression biomarkers for osteosarcoma (OS) survival prediction.
• To develop and validate a robust prognostic risk model for OS patients using gene expression data.
• To explore the biological functions associated with different risk groups in OS.

Main Methods:

• Utilized RNA-sequencing data from 82 OS samples from the TARGET database.
• Employed random forest models to rank candidate prognostic genes.
• Developed an optimal multivariable Cox regression model with five key predictors: CD180, MYC, PROSER2, DNAI1, and FATE1.
• Created a scoring method for patient risk stratification and validated the model's performance.

Main Results:

• Identified a five-gene signature (CD180, MYC, PROSER2, DNAI1, FATE1) as an optimal prognostic model for OS.
• The developed risk model demonstrated robust performance in stratifying OS patients.
• Differentially expressed genes between risk groups were significantly associated with immune-related biological pathways and processes.

Conclusions:

• The identified five-gene signature serves as a novel prognostic biomarker for predicting OS patient survival.
• The developed risk model can aid in stratifying patients and guiding the development of molecularly targeted therapies.
• These findings contribute to improving patient outcomes and survival in osteosarcoma.