Introduction: Recently, we have developed reverse phase protein microarrays for the analysis of cellular signaling pathway activation in human biopsy material from breast, ovarian, lung, colorectal, prostate cancer and liver cancers. This approach can quantify protein activation status using antibodies specific for post-translational modifications such as phosphorylation or cleavage. Methods: We now extend the use of this technology to FACS sorted CD138 positive plasma cells obtained from bone marrow aspirates of 72 patients with active multiple myeloma, prior to therapy. Reverse phase microarrays were constructed by printing lysates in dilution curves on nitrocellulose-coated glass slides which were probed with a comprehensive panel of 67 kinase substrates. Stained slides were analyzed with MicroVigene™ software. Results: Molecular network analysis using unsupervised clustering techniques reveals a striking degree of patient-specific signaling activation. This is clearly dominated by two major groups that were discriminated based on high or low levels of phosphorylation amongst all endpoints analyzed. Within these dominant groups, subgroups were characterized by EGF receptor activation, pro-survival pathway activation and pro-mitogenesis activation. Some patients appeared to have hyper-phosphorylation of multiple pathways. More refined analysis of correlations using defined genetic markers of outcome and aggressiveness, such as GEPRisk70 scoring, indicated the possibility of AKT and mTOR pathway involvement. Assessment of the correlation of pathway activation with clinical outcome is ongoing. Conclusions: To our knowledge, this analysis represents the largest number of specific phosphorylation signaling endpoints quantitatively measured in a single study. Such post-translational analysis cannot be achieved from gene expression profiling alone. This phosphoproteomic analysis reveals striking variation in the signaling repertoire of myeloma in different patients. Such signaling heterogeneity could underpin the dramatic differences in clinical response rates seen for this disease, thereby highlighting the need for more substantive stratification for targeted therapy.

[Proc Amer Assoc Cancer Res, Volume 47, 2006]