High-Resolution Genomic Profiling Reveals Association of Chromosomal Aberrations on 1q and 16p with Histologic and Genetic Subgroups of Invasive Breast Cancer

Daniel E. Stange, Bernhard Radlwimmer, Falk Schubert, Frank Traub, Andreas Pich, Grischa Toedt, Frank Mendrzyk, Ulrich Lehmann, Roland Eils, Hans Kreipe, Peter LichterClinical 2006 · DOI

Abstract

Molecular profiling using bacterial artificial chromosome arrays identified DNA copy number imbalances on 1q and 16p as significant classifiers of histologic and molecular subgroups.

Introduction

Invasive ductal carcinoma (IDC) and invasive lobular carcinoma (ILC) represent the two most prevalent forms of invasive breast cancer. While these subtypes exhibit distinct histopathological features and clinical behaviors—including patterns of metastasis and hormone receptor responsiveness—they have proven difficult to differentiate using standard mRNA-based gene expression profiling. This study investigates the hypothesis that DNA copy number imbalances provide a more stable and definitive "molecular portrait" of these cancers, identifying genetic drivers that distinguish IDC from ILC.

Materials and Methods

The study analyzed 40 invasive breast cancer samples (18 IDC, 21 ILC, and 1 mixed case) collected at the Medizinische Hochschule Hannover. Genomic profiling was conducted using matrix-comparative genomic hybridization (matrix-CGH) with Bacterial Artificial Chromosome (BAC) microarrays, offering a high resolution of approximately 0.5 Mb.

Data analysis integrated several bioinformatic approaches:

  • Support Vector Machines (SVM): Employed as a machine learning classifier to identify genomic fragments distinguishing IDC from ILC.
  • Significance Analysis of Microarrays (SAM): Used to pinpoint regions with statistically significant frequency differences.
  • Unsupervised Hierarchical Clustering: Applied to group tumors based on intrinsic genetic patterns.
  • Validation: Real-time quantitative PCR (qPCR) was used to correlate DNA gains with mRNA expression levels for specific candidate genes.

Results and Discussion

Identification of 1q and 16p as Histological Classifiers

The most significant genomic differences between ILC and IDC were localized to chromosome arms 1q and 16p. Gains in these regions were markedly more frequent in ILC cases. Specifically, gain of 1q24.2-q25.1 was observed in 76.2% of ILC versus 25.9% of IDC, while gain of 16p11.2-p13.12 was found in 42% of ILC compared to only 6.5% of IDC.

Comparison of genomic profiles for IDC and ILC (A, B) and the resulting frequency difference plot (C), highlighting significant gains on 1q and 16p in ILC.

Using a classifier comprising 733 genomic fragments, the researchers achieved a 65% accuracy in distinguishing the two histological types. The stability of these DNA copy number changes suggests they serve as reliable markers of the tumor's evolutionary trajectory, less prone to the "noise" often associated with transcriptomic data.

Candidate Gene Mapping and Expression

High-resolution mapping allowed for the identification of specific subregions and candidate genes likely driving tumor behavior. On chromosome 1q, genes such as FMO2 and PTGS2 were identified. On chromosome 16p, the genes BCL7C and FUS were found to be overexpressed by 5- to 10-fold in tumors harboring the corresponding DNA gains.

Detailed mapping of chromosomes 1 and 16, showing regions identified by SAM and SVM analysis and the location of candidate genes.

Molecular Subgrouping and Clinical Correlation

Unsupervised clustering revealed three distinct genetic groups that transcended simple histological classification:

  • Group I: Predominantly IDC, characterized by high-level MYC (8q24) gains.
  • Group II: Primarily low-grade ILC, defined by 1q gains and 16q losses.
  • Group III: High-instability tumors with widespread chromosomal aberrations.

These molecular subgroups showed significant correlation with clinical parameters, including tumor grade ($P = 0.004$) and estrogen receptor (ER) expression ($P = 0.005$).

Conclusion

This study demonstrates that high-resolution genomic profiling can uncover specific DNA copy number imbalances that distinguish breast cancer subtypes and correlate with clinical outcomes. While the sample size (n=40) is a limitation, the identification of 1q and 16p as key classifiers provides a foundation for future multi-omic integration. Combining DNA, mRNA, and protein data will be essential for developing personalized diagnostic and therapeutic strategies in breast cancer management.

Summary updated

Abstract

Molecular profiling using bacterial artificial chromosome arrays identified DNA copy number imbalances on 1q and 16p as significant classifiers of histologic and molecular subgroups.
BibTeX
@inproceedings{stange2006high,
  author = {Daniel E. Stange and Bernhard Radlwimmer and Falk Schubert and Frank Traub and Andreas Pich and Grischa Toedt and Frank Mendrzyk and Ulrich Lehmann and Roland Eils and Hans Kreipe and Peter Lichter},
  title = {{High-Resolution Genomic Profiling Reveals Association of Chromosomal Aberrations on 1q and 16p with Histologic and Genetic Subgroups of Invasive Breast Cancer}},
  booktitle = {Clinical},
  year = {2006},
  doi = {10.1158/1078-0432.ccr-05-1633},
}
Figure