Development Of Methods To Perform Next Generation Sequencing (NGS)-Based Genomics Studies On Multiple Myeloma Clinical Samples

NGS is a powerful tool for interrogating genomic alterations in cancer cells from the scale of single nucleotides up to whole chromosomes. While methods for performing NGS analysis on solid tumors have been widely reported in the literature, fewer methods are available for hematological malignancies, which are often collected and stored in different conditions. Here we report our efforts to overcome the unique challenges of performing a whole exome sequencing (WES) study on hundreds of multiple myeloma (MM) patient samples.

As part of Onyx-sponsored clinical trials of the proteasome inhibitor, carfilzomib, bone marrow aspirates were collected from patients and subjected to bead-based CD138 positive selection to enrich tumor cells. The yield of CD138+ cells after enrichment varied greatly from 200,000 to >10,000,000, depending primarily on the disease state and the volume and quality of aspirate received. CD138+ cells were banked in a variety of formats that differed in their suitability for downstream DNA and RNA analysis, including as pellets or re-suspended in reagents such as TRIzol®, RNAlater and Buffer RLT. At the time of abstract submission we have performed extensive analyses of samples stored in TRIzol.

TRIzol is renowned for its excellence in preserving RNA integrity while providing a means to simultaneously extract DNA and protein. However, the standard DNA extraction method yields material that is not of sufficient quality for DNA analyses such as WES. We attempted to remedy this problem through three approaches: (1) modifying extraction conditions to improve the quality of the DNA, (2) rescuing the extracted DNA with double-stranding steps prior to library preparation, and (3) applying newer, low input and single-strand library construction methods. Our efforts to modify the extraction conditions were unsuccessful, but by combining DNA rescue and newer library construction methods, we could ultimately generate high quality exomes (as judged by comparison to exomes generated from matched, pelleted samples) from the whole range of clinical samples in our biobank. We believe these methods create new opportunities for the community to interrogate, with the latest NGS approaches, vast numbers of clinical samples banked in the era of gene-expression profiling (GEP).