Gene Signatures Predicting Metastasis in Breast Cancer
Author Information
Author(s): Yu Jack X, Sieuwerts Anieta M, Zhang Yi, Martens John WM, Smid Marcel, Klijn Jan GM, Wang Yixin, Foekens John A
Primary Institution: Erasmus MC, Josephine Nefkens Institute
Hypothesis
Different gene signatures can predict the development of metastases in breast cancer patients.
Conclusion
Pathway-derived gene signatures can effectively predict the prognosis of breast cancer patients based on their biological processes.
Supporting Evidence
- The study identified 76-gene signatures that predicted distant metastasis.
- The average AUC for ER-positive patients was 0.70, indicating good predictive power.
- The study found that different gene signatures share similar biological pathways.
Takeaway
Scientists studied genes in breast cancer to see which ones help predict if the cancer will spread. They found that certain groups of genes can give clues about how serious the cancer might be.
Methodology
The study used 500 re-sampled training sets to identify gene signatures and mapped them to biological pathways.
Potential Biases
Potential bias due to the selection of training sets and the specific patient population.
Limitations
The study focused only on node-negative breast cancer and may not apply to other types.
Participant Demographics
The cohort included 344 breast cancer patients, with 221 ER-positive and 123 ER-negative cases, mean age 53 years.
Statistical Information
P-Value
3.06E-7
Confidence Interval
95% CI: 0.61–0.77
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website