Integrated analysis of independent gene expression microarray datasets improves the predictability of breast cancer outcome
2007

Improving Breast Cancer Outcome Predictions with Gene Expression Data

Sample size: 78 publication 10 minutes Evidence: moderate

Author Information

Author(s): Zhang Zhe, Chen Dechang, Fenstermacher David A

Primary Institution: University of North Carolina, Chapel Hill

Hypothesis

Can integrated analysis of multiple gene expression datasets improve the predictability of breast cancer outcomes?

Conclusion

The study confirmed that integrating multiple datasets for gene expression profiles leads to better predictions of breast cancer recurrence.

Supporting Evidence

  • The integrated profiling of two independent datasets led to higher classification accuracy (71.11% vs. 70.59%).
  • Cell cycle, especially M phase mitosis, was significantly overrepresented by the 60-gene profile obtained from integrated analysis.
  • The profiles significantly differentiated poor and good prognosis in a third patient cohort.

Takeaway

Scientists looked at gene data from many breast cancer patients to make better predictions about who might get sick again after treatment.

Methodology

The study used integrated analysis of two independent datasets and validated the results with a third dataset, employing random re-sampling and cross-validation.

Potential Biases

The results may be influenced by the clinical characteristics of the patient samples, such as ER and lymph node status.

Limitations

The study's findings may be confounded by the clinical backgrounds of the patient cohorts used.

Participant Demographics

The study involved breast cancer patients classified into good and poor prognosis groups based on recurrence within three years.

Statistical Information

P-Value

0.003

Confidence Interval

95% CI not specified

Statistical Significance

p < 0.0001

Digital Object Identifier (DOI)

10.1186/1471-2164-8-331

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication