Gene Expression Profiling of Inflammatory Breast Cancer

BACKGROUND: Inflammatory breast cancer (IBC) is a rare but aggressive form of breast cancer. Despite multimodality treatment, the long-term survival rate for patients with IBC has remained inferior at 50%. Until recently, IBC was understudied at the molecular level. Since 2004, new high-throughput molecular profiling technologies have been applied to clinical samples with the aim of identifying genes or pathways potentially involved in disease development that may represent new, clinically relevant targets. METHODS: The authors conducted gene expression profiling studies of IBC clinical samples and investigated issues that may be addressed in the future to allow the "omics" approach to reach its full potential in IBC. RESULTS: Starting in December 2004, 6 research groups compared the expression profiles of IBC samples and non-IBC samples. The series of samples were small (37 IBCs for the largest study) and heterogeneous (various tumor selection criteria and technologic platforms were used). The results indicated the feasibility of messenger RNA expression profiling from IBC biopsies, and they demonstrated the great transcriptional heterogeneity of IBC and the existence of molecular subtypes similar to non-IBC that more frequently were basal and positive for ERBB2. Supervised analyses demonstrated differences in gene expression levels between the IBC and non-IBC variable across studies with sometimes no or very subtle differences and, to date, no gene overlap across the reported signatures. No signature predictive of therapeutic response or clinical outcome has been reliably identified or validated. CONCLUSIONS: Because of the great heterogeneity of IBC, future studies will have to include larger series of IBC samples that are selected using homogeneous criteria. This calls for urgent international collaborations. Cancer 2010;116(11 suppl):2783-93. (C) 2070 American Cancer Society.