Multiple myeloma (MM) is an incurable malignant neoplasm hallmarked by a clonal expansion of plasma cells, the presence of a monoclonal protein in the serum and/or urine (M-spike), lytic bone lesions, and end organ damage. Clinical outcomes for patients with MM have improved greatly over the last decade as a result of the re-purposing of compounds such as thalidomide derivatives, as well as the development of novel chemotherapeutic agents including first and second generation proteasome inhibitors, bortezomib (Bz) and carfilzomib. Unfortunately, despite these improvements, the majority of patients relapse following treatment. While Bz, one of the most commonly used proteasome inhibitors, has been successfully incorporated into clinical practice, some MM patients have de novo resistance to Bz, and the majority of the remainder subsequently develop drug resistance following treatment. A significant gap in clinical care is the lack of a reliable clinical test that would predict which MM patients have or will subsequently develop Bz resistance. Thus, as Bz resistance remains a significant challenge, research efforts are needed to identify novel biomarkers of early Bz resistance, particularly when an early therapeutic intervention can be initiated. Recent advances in MM research indicate that genomic data can be extracted to identify novel biomarkers that can be utilized to select more effective, personalized treatment protocols for individual patients. Computationally integrating large patient databases with data from whole transcriptome profiling and laboratory-based models can potentially revolutionize our understanding of MM disease mechanisms. This systems-wide approach can provide rational therapeutic targets and novel biomarkers of risk and treatment response. In this review, we discuss the use of high-content datasets (predominantly gene expression profiling) to identify novel biomarkers of treatment response and resistance to Bz in MM.