Abstract Breast cancer is a heterogeneous disease as reflected by histopathology, molecular alterations and clinical behavior. In order to relate cellular and sub-cellular features to clinical parameters and outcome, substantial effort has been exerted towards identifying tumor groups with distinct molecular features. The gene expression based classification that we proposed ten years ago identified five different subgroups with difference in survival; two Luminal-cell related groups (Luminal A and luminal B), a myoepithelial-cell related group (Basal-like), a HER2 enriched, and a Normal-like group. The Basal-like and Luminal carcinomas have different etiologies and for most purposes may be considered as distinct diseases. This is also reflected in the genomic portraits defined by aCGH (array Comparative Genomic Hybridization), and it seems evident that the history of the molecular subgroups is written in the DNA alterations. We have developed two platform independent algorithms to explore genomic architectural distortion using aCGH data to measure whole arm gains and losses (WAAI) and complex rearrangements (CAAI). By applying CAAI and WAAI to data from 595 breast cancer patients we were able to separate the cases into eight subgroups with different distribution of genomic distortion. Within each subgroup data from expression analyses, sequencing and ploidy indicated that progression occurs along separate paths into more complex genotypes. Several of the CGH based subgroups were enriched for either luminal related or basal related tumors, and showed distinct patterns of genomic alterations. In particular, basal related tumors were separated into genomic classes with distinct patterns of alterations emphasizing genomic heterogeneity also in this expression subclass. Histological grade had prognostic impact only in the Luminal related groups while the complexity identified by CAAI had an overall independent prognostic power. These results emphasizes the relationship between structural genomic alterations, expression subtypes and clinical behavior, and provides a score of genomic complexity as a new tool for prognostication in breast cancer. Using SNP arrays and a novel bioinformatic approach, ASCAT (Allele-Specific Copy number Analysis of Tumors), we have performed genome wide allele-specific copy number analysis of 112 breast cancer genomes. We were able to accurately dissect the allele-specific copy number in each tumor, simultaneously estimating and adjusting for both tumor ploidy and non-aberrant cell admixture, allowing calculation of “Tumor Profiles” (genome-wide allele-specific copy-number profiles) from which gains, losses, copy-number-neutral events and LOH could be determined. The tumors had on average a non-aberrant cell admixture of 49%, and 45% of them were aneuploid (>2.7n). By aggregation of Tumor Profiles across the sample set a genome-wide view of LOH and copy-number-neutral events was obtained. The Tumor Profiles also revealed differences in aberrant tumor cell fraction, ploidy, gains, losses, LOH and copy-number-neutral events between the five expression subtypes. Basal-like breast carcinomas have a significantly higher frequency of LOH compared to other subtypes, and their Tumor Profiles show large-scale loss of genomic material during tumor development, followed by a whole-genome duplication, resulting in near-triploid genomes. From the Tumor Profiles we could also construct a genome-wide map of allelic skewness, indicating loci where one allele is preferentially lost while the other allele is preferentially gained. We hypothesize that these alternative alleles have a different influence on breast carcinoma development. Citation Information: Clin Cancer Res 2010;16(7 Suppl):PL1-1