Protein expression profiling identifies a prognostic model for ovarian cancer

Area of Science:

• Oncology
• Genomics
• Proteomics

Background:

• Ovarian cancer presents significant morbidity and mortality, impacting female health.
• Reliable prognostic models are crucial for monitoring disease progression and improving patient outcomes.

Purpose of the Study:

• To develop and validate a predictive risk model for ovarian cancer survival.
• To identify key proteins associated with overall survival in ovarian cancer patients.

Main Methods:

• Utilized TCPA and TCGA databases for protein expression and survival data.
• Employed univariate and multiple Cox regression analyses to screen proteins and construct a risk model.
• Assessed the model's predictive power and prognostic implications through validation, co-expression, and enrichment analyses.

Main Results:

• Identified 20 proteins significantly associated with ovarian cancer patient survival (p < 0.01).
• A six-protein model (GSK3α/β, HSP70, MEK1, MTOR, BAD, NDRG1) was constructed, demonstrating significant prognostic value.
• The high-risk group showed unfavorable overall and disease-specific survival (p < 0.001).
• The risk model outperformed age, tumor grade, and stage in predictive accuracy (AUC = 0.789).

Conclusions:

• The developed six-protein risk model effectively predicts ovarian cancer patient survival.
• This model serves as an independent prognostic factor and can aid in ovarian cancer disease management.