A Genome-Wide Assessment of Inherited Genetic Variants and the Risk of Langerhans Cell Histiocytosis

Introduction: Langerhans cell histiocytosis (LCH) is a disease characterized by inflammatory lesions including pathologic CD207+ dendritic cells. Clinically, LCH is highly variable ranging from single lesions to highly aggressive, disseminated disease involving multiple organs and requiring intensive chemotherapy. Recent data support a model of pathogenesis in which activating somatic mutations in MAPK pathway genes arise in myeloid DC precursors. However, little is known about genetic susceptibility to this condition. Therefore, we conducted a genome-wide association study to characterize the role of inherited genetic variants on disease risk.

Methods: We utilized a case-parent trio approach, which is immune to the effects of population stratification bias. Specifically, this allows for the inclusion of individuals regardless of genetic ancestry. LCH case-parent trios (n=134) were recruited from Texas Children's Cancer Center. Genotyping was performed using the Illumina Omni-5 Quad BeadChip. Genetic ancestry was determined using the bioinformatics algorithm STRUCTURE. To inform this algorithm, a set of 12,898 autosomal ancestry informative markers specifically identified to infer population substructure was extracted from the study trios. Estimated genomic ancestral proportions were then used to classify each study participant as either of European ancestry, Amerindian ancestry, or of African ancestry. For the association analysis, we focused on the role of common variants (i.e., minor allele frequency ≥5%). The association analysis was conducted utilizing the PREMIM-EMIM algorithm, an established, multinomial log-likelihood approach for assessing case-parent trios GWAS data. This method allows for the inclusion of "incomplete" trios (e.g., mother-case duos). We applied a genome-wide statistical significance cutoff of p<1.0x10^-5.

Results: In this GWAS, LCH cases were predominantly male (54%), and based on the genetic ancestry analysis, 60% were of European ancestry, 36% Amerindian ancestry, and 4% African ancestry. Among the 343 individuals included in the analysis, 1,672,105 SNPs autosomal SNPs were assessed and an overview of the results is displayed using a Manhattan plot. We identified five potential inherited genomic regions associated with LCH susceptibility. The strongest associations between inherited SNPs and childhood LCH were seen in SMAD6 on chromosome 15 (p-value = 2.38x10^-7) and in ECE1 on chromosome 1 (p-value = 2.15x10^-6).

Conclusions: In this genome-wide assessment of the role of inherited genetic variation on the risk of LCH, we identified SNPs with significant effects in genes implicated in diverse pathways including embryogenesis and cellular division. Among the proteins encoded by the regions identified, SMAD6and ECE1 have both been reported to impact ERK activation, a critical feature of LCH pathogenesis. These findings support potential for inherited genetic variants to influence risk of developing LCH.