Related Experiment Video
Updated: May 15, 2026
10:27
Testing Targeted Therapies in Cancer using Structural DNA Alteration Analysis and Patient-Derived Xenografts
Published on: July 25, 2020
Predictive modeling using a somatic mutational profile in ovarian high grade serous carcinoma
1Samsung Cancer Research Institute, Seoul, Korea.
Plos One
|January 18, 2013
Summary
This study developed a predictive model for ovarian cancer survival using whole exome somatic mutations. The model accurately predicts overall survival and progression-free survival, offering a new strategy for clinical application of genomic data.
Area of Science:
- Genomics
- Oncology
- Bioinformatics
Background:
- High-throughput sequencing reveals numerous somatic mutations in various cancers.
- Ovarian high-grade serous carcinomas (Ov-HGSCs) present a complex mutational landscape.
Purpose of the Study:
- To develop a predictive model for survival in Ov-HGSCs using whole exome somatic mutational profiles.
- To assess the model's accuracy in predicting overall survival (OS) and progression-free survival (PFS).
Main Methods:
- Utilized The Cancer Genome Atlas (TCGA) data for 311 Ov-HGSC patients for OS and 259 for PFS.
- Analyzed mutations in 509 genes and employed leave-one-out cross-validation for model validation.
- Generated cross-validated Kaplan-Meier and time-dependent ROC curves to evaluate predictive accuracy (AUC).
Main Results:
- The predictive model showed significant differences in OS (median 28.1 vs. 61.5 months) and PFS (median 10.9 vs. 22.3 months) between high- and low-risk groups (p<0.001).
- Achieved cross-validated AUC values of 0.807 for OS and 0.747 for PFS at 36 months.
- Gene-based models without clinical covariates demonstrated high predictive efficacy.
Conclusions:
- A novel predictive model was designed using somatic mutation profiles from high-throughput genomic sequencing in Ov-HGSC.
- This approach may offer a new strategy for integrating genomic data into clinical practice for ovarian cancer management.
