Analytical Approaches for the BOAC SNP Panel Association with Progression Free Survival in Myeloma

The Bank On A Cure (BOAC) has established DNA banks from multiple cooperative and institutional clinical trials, and platforms for examining the association of genetic variations (SNPs) with disease risk and outcomes in myeloma. We have previously described the development and content of a novel custom SNP panel that contains 3,404 SNPs in 983 genes, representing cellular functions and pathways that may influence disease response, toxicities, complications, and survival. Although survival certainly varies according to tumor heterogeneity (ie. chromosomal abnormalities, gene expression variations) germline variations that influence the microenvironment, drug distribution, drug transport and metabolism, may also have an association with event free survival outcomes. To explore SNP associations with progression free survival (PFS) we compared the BOAC SNP profiles of short term PFS (less than 1 year, n=70) versus long term PFS (greater than 3 years, n=73) in two phase III clinical trials (ECOG E9487 and SWOG S9321). A variety of analytical approaches was undertaken including univariate rank ordering, recursive partitioning, and support vector machine learning tools (SVM). Each of these approaches has advantages and limitations in dealing with type I false positive errors as well as sensitivity and specificity. We included subset validation approaches and randomization of classes to address how robust and predictive different approaches were. From our analysis we conclude germline genomic variations do have an impact on progression free survival, with a subset of SNPs from the panel reaching 76% predictive association and hazard ratios of PFS of 9.6 (CI 4.5, 20.5), p<0.001, using SVM analysis. Based on univariate approaches, we find the most significant variations associated with PFS differences were genes that could be functionally categorized as pharmacologic. The presentation will focus on the analytical approaches, and refinements necessary to assure predictive value compared to random associations. Notwithstanding the clear importance of tumor cell variations in genetic deregulation, we conclude that various functions within the bone marrow and drug response likely interplay as a complex influence on disease progression, response, and survival. This suggests combining gene expression profiles of the tumors with germline SNP profiles may provide more accurate prognosis. These combined analytical approaches are currently being developed with BOAC data bases, and examples will be discussed.