Integrative Analysis of the Genomic Landscape Underlying Multiple Myeloma at Diagnosis: An Mmrf Commpass Analysis

Multiple myeloma is a hematological malignancy characterized by an expansion of clonal plasma cells in the bone marrow. Our understanding of the pathogenesis of this disease has improved dramatically with the development of whole genome analysis technologies. However, to date no study has comprehensively analyzed a large cohort of myeloma patients. The Multiple Myeloma Research Foundation (MMRF) CoMMpass trial (NCT145429) is a longitudinal study of 1143 patients with newly-diagnosed multiple myeloma patients from clinical sites in the United States, Canada, Spain, and Italy. To be enrolled on the study each patient must be diagnosed with myeloma and must receive a treatment regimen within 30 days after providing a bone marrow aspirate with a minimum of 1% plasma cells and a recovery of >250,000 CD138 positive plasma cells. The initial treatment regimen is required to contain a proteasome inhibitor, immunomodulatory agent, or both. Clinical parameters are collected at study enrollment and every three months through the eight-year observation period. Each baseline and progression tumor specimen is characterized using Whole Genome Sequencing (WGS), Whole Exome Sequencing (WES), and RNA sequencing (RNAseq).

This analysis includes all data available as of January 1, 2017 from interim analysis 11, which represents the first data freeze with the complete molecular analysis of all baseline specimens. This dataset includes 1143 enrolled patients of whom 1003 are molecularly characterized with 982 having some level of baseline characterization. As part of the baseline characterization we released WGS on 890, WES on 926, and RNAseq on 714 patients, respectively. In total 591 patients had complete characterization of the provided baseline bone marrow specimen.

The median follow-up of the cohort now exceeds 2 years which has identified a median PFS of 36 months for the cohort. The median OS has still not been reached. Interestingly, there is a significant difference in PFS and OS, with males performing worse than females, p<0.001 and p<0.001, respectively, that is largely associated with dramatic differences for patients in ISS stage I.

We identified a median of 153 non-immunoglobulin related mutations per patient, a median of 29 structural events, and a median of 133 copy number segments per tumor. To identify a set of significantly mutated genes we applied a consensus-based approach resampling 80% of the cohort 1000 times, which identified 60 distinct genes mutated in at least 1% of the cohort. Many of these genes are already implicated in myeloma but many are novel including; BMP2K, PABPC1, PANK3, PTPN11, and RPS3A that all have mutations at recurrent amino acid positions, characteristic of an oncogene. We performed a similar consensus clustering approach on the copy number profiles and gene expression profiles, identifying 14 and 12 distinct subtypes respectively. To generate an integrative view of the genetics underlying myeloma we identified potential loss-of-function and gain-of-function genes using an integrated approach leveraging somatic copy number alterations, copy neutral loss-of-heterozygosity, mutations, structural abnormalities, gene expression alterations, and constitutional inherited gene defects. This approach identified numerous genes known to be involved in myeloma like CCND1, NRAS, TRAF3, and IRF4. But unlike other methods it found numerous novel genes missed by other strategies like DPYD and PSPC1. This also highlighted the critical importance of chromosome 13 and X loss, each with 15 genes with complete bi-allelic loss in multiple patients. Moreover, the complete loss of DPYD identifies a subset of patients that may have dramatic responses to 5-FU through an Achilles heel based mechanism. Overall, this represents the most comprehensive analysis of the genetic underpinnings of myeloma.