Computer-assisted HER2 Scoring in Breast Cancer
Author Information
Author(s): Bonnie H. Hall, Monica Ianos-Irimie, Parisa Javidian, Wenjin Chen, Shridar Ganesan, David J. Foran
Primary Institution: UMDNJ
Hypothesis
Can a computer-assisted diagnostic algorithm improve the accuracy of HER2 scoring in breast cancer specimens compared to manual scoring?
Conclusion
The study demonstrated that a computer-assisted diagnostic algorithm can more accurately score HER2 status in breast cancer specimens than standard manual evaluation.
Supporting Evidence
- The computer-assisted method showed a greater Area Under the Curve (AUC) in ROC analysis than manual scoring.
- Incorporating positive controls into the analysis improved diagnostic accuracy.
- The computer-assisted approach reduced the false positive rate in equivocal cases.
Takeaway
This study shows that using a computer program to help score breast cancer tests can make the results more accurate, especially for tricky cases.
Methodology
The study compared computer-assisted analysis of HER2 IHC with manual scoring and FISH results on 99 breast cancer cases, using ROC analysis for evaluation.
Potential Biases
Potential bias in manual scoring due to subjective interpretation by pathologists.
Limitations
The study was retrospective and limited to a specific set of cases, which may not represent all breast cancer patients.
Participant Demographics
The study included breast cancer cases diagnosed between January 2005 and March 2007.
Statistical Information
P-Value
p<0.05
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website