Gene expression profiling between normal B cells/plasma cells and multiple myeloma cells revealed four distinct subgroups of multiple myeloma plasma cells that have significant correlation with clinical characteristics known to be associated with poor prognosis. Diagnosis for multiple myeloma (and possibly monoclonal gammopathy of undetermined significance) based on differential expression of 14 genes, as well as prognostics for the four subgroups of multiple myeloma based on the expression of 24 genes were also established. Gene expression profiling also allows placing multiple myeloma into a developmental schema parallel to that of normal plasma cell differentiation. The development of a gene expression- or developmental stage-based classification system for multiple myeloma would lead to rational design of more accurate and sensitive diagnostics, prognostics and tumor-specific therapies for multiple myeloma.
1. A method for gene expression-based classification for multiple myeloma, comprising: extracting total RNA of plasma cells obtained from individuals with newly diagnosed multiple myeloma and control group of healthy individuals: isolating total RNA from said plasma cells; processing said total RNA for hybridization to a DNA microarray; hybridizing said total RNA to a DNA microarray; and applying hierarchical clustering analysis to the data obtained from said hybridization, wherein said clustering analysis will classify said individuals with multiple myeloma into one of the four distinct subgroups of multiple myeloma.