Haplotype analysis identifies functional elements in monoclonal gammopathy of unknown significance

Abstract

Genome-wide association studies (GWASs) based on common single nucleotide polymorphisms (SNPs) have identified several loci associated with the risk of monoclonal gammopathy of unknown significance (MGUS), a precursor condition for multiple myeloma (MM). We hypothesized that analyzing haplotypes might be more useful than analyzing individual SNPs, as it could identify functional chromosomal units that collectively contribute to MGUS risk. To test this hypothesis, we used data from our previous GWAS on 992 MGUS cases and 2910 controls from three European populations. We identified 23 haplotypes that were associated with the risk of MGUS at the genome-wide significance level (p < 5 × 10^-8) and showed consistent results among all three populations. In 10 genomic regions, strong promoter, enhancer and regulatory element-related histone marks and their connections to target genes as well as genome segmentation data supported the importance of these regions in MGUS susceptibility. Several associated haplotypes affected pathways important for MM cell survival such as ubiquitin-proteasome system (RNF186, OTUD3), PI3K/AKT/mTOR (HINT3), innate immunity (SEC14L1, ZBP1), cell death regulation (BID) and NOTCH signaling (RBPJ). These pathways are important current therapeutic targets for MM, which may highlight the advantage of the haplotype approach homing to functional units.

Fig. 1. Meta-analysis of the haplotypes from the three populations, originating from the Czech Republic (CZE), Germany (GER) and Sweden (SWE), represented by the joint SNP rs11368313 (in red).

Other SNPs common in all three populations are shown in green. For each population, the chromosome start and end position and the allele of the haplotype are shown.

Fig. 2. Haplotype regions and Forest plots from the five most interesting haplotypes associated with the risk of MGUS.

Haplotype regions are shown using UCSC Genome browser’s GRCh37/hg19 assembly and annotation tracks from ENCODE and GeneHancer. Forest plots show the overlapping haplotypes, represented by a joint SNP, of the three study populations from the Czech Republic (CZE), Germany (GER) and Sweden (SWE). For each population the odds ratio and the corresponding 95% confidence interval (CI) are shown as well as the summary estimate of the meta-analysis. A Chromosome 1, rs11368313, B Chromosome 6, rs10658790, C Chromosome 17, rs10163481, D Chromosome 20, rs111797554, E Chromosome 22, rs1045588.