Automated Brightfield Double In Situ Hybridization for HER2 in Breast Cancer
Author Information
Author(s): Nitta Hiroaki, Hauss-Wegrzyniak Beatrice, Lehrkamp Megan, Murillo Adrian E, Gaire Fabien, Farrell Michael, Walk Eric, Penault-Llorca Frederique, Kurosumi Masafumi, Dietel Manfred, Wang Lin, Loftus Margaret, Pettay James, Tubbs Raymond R, Grogan Thomas M
Primary Institution: Ventana Medical Systems, Inc
Hypothesis
The study aimed to develop an automated brightfield double in situ hybridization (BDISH) application for HER2 gene and chromosome 17 centromere detection in breast carcinomas.
Conclusion
The automated BDISH application for HER2 and CEN 17 targets was successfully developed and may replace manual two-color HER2 FISH methods.
Supporting Evidence
- The BDISH assay demonstrated a high consensus concordance with FISH results (98.9%).
- The sensitivity of the BDISH assay was 96.3% and specificity was 100%.
- Individual concordance rates among observers ranged from 97.8% to 100%.
- BDISH allows simultaneous analysis of two DNA targets within tissue morphology.
- BDISH technology can be archived for long-term storage and review.
- BDISH can be performed without specialized fluorescence microscopy.
- The study highlights the potential for BDISH to improve breast cancer patient care.
- Discordant cases were linked to tumor cell population heterogeneity.
Takeaway
Researchers created a new test that can quickly check for a specific gene in breast cancer, which could help doctors choose better treatments for their patients.
Methodology
The BDISH assay was developed using a specific DNA probe and evaluated on 94 breast cancer tissue samples, comparing results with manual FISH.
Potential Biases
Discordant cases showed tumor cell population heterogeneity, which could influence results.
Limitations
Non-consecutive tissue sections were used for FISH and BDISH analyses, which may affect concordance.
Participant Demographics
Breast cancer cases from the Cleveland Clinic Foundation and the Cleveland Clinic Lerner College of Medicine.
Statistical Information
P-Value
0.9736
Confidence Interval
95% CI = 0.9222 – 1.0000
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website