Abstract P1-16-01: High-Resolution Genomic Profiling of Male Breast Cancer Reveals Differences behind the Similarities with Female Breast Cancer

Abstract

Abstract Background: Male breast cancer (MBC) is a rare disease, accounting for <1% of all breast cancer cases. Although it is similar to female breast cancer (FBC) in many ways, distinct differences in incidence, age distribution, levels of hormone receptors, prognosis and survival have been reported. Due to the rarity of the disease, no large comprehensive studies of the etiology of the disease or randomized trials for optimizing patient management have been conducted. Thus, the recommendations for treating MBC are extrapolated from prior knowledge of FBC. The aim of this study was to characterize MBC by genomic imbalances using high-resolution genomic data and to compare it with FBC, and further to investigate whether this holds any prognostic information. Materials and Methods: Fifty-six fresh frozen MBC tumors were analyzed using tiling BAC arrays comprising ∼32,000 clones, making this the first study on a large series of MBC using array comparative genomic hybridization (aCGH) to date. An in-house set of 359 FBC samples was used as a reference data set. Results: Our data revealed a broad pattern of aberrations in MBC, confirming that MBC is a heterogeneous tumor type. Overall, MBC displayed more regions of genomic gain than FBC, but fewer losses; many aberrations were large, involving whole chromosome arms. The most common aberrations were similar to genomic aberrations documented in FBC. Fifteen loci that were affected by high-level amplifications were found in MBCs involving 15 tumors (27% of all MBCs), while in the reference FBC data set 41 loci, involving 139 samples (39% of all FBCs), were found. Using Genomic Identification of Significant Targets in Cancer (GISTIC), 39 significant genomic aberrations were found in MBC. In contrast, 133 GISTIC regions were found in the reference FBC data set. Further, three stable genomic subgroups of MBC were identified using three different approaches: clustering with the FBC GISTIC regions, clustering with the MBC GISTIC regions and centroid classification, respectively. The majority of the MBC tumors clustered in a group with striking similarities to a subset of luminal FBCs and contained all known BRCA2 mutated tumors. This group seemed to have the worst survival among all MBCs. Two HER2 amplified tumors were included, and they clustered together in a small group where 17q12 was gained in all samples and which was similar to the HER2 subgroup of FBC. The third subgroup was characterized by a very low fraction of genome altered and did not resemble any known FBC subgroup. Hence, this might be a new subgroup of breast cancer only occurring in males, and warrants further investigation. Discussion: In summary, genomic profiling of MBC revealed overall similarities with FBC, with certain striking exceptions: genomic gains were more common in MBC and often involved whole chromosome arms, while losses of genomic material were less common. MBC could further be classified into comprehensive subgroups based on genomic profiles, one of which appears remarkably different from any genomic subgroup defined in FBC. Taken together, these findings suggest the existence of heterogeneity among breast cancers diagnosed in males, with potential implications for clinical management. Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P1-16-01.