Enrichment of Rare Variants of Hemophagocytic Lymphohistiocytosis Genes in Systemic Juvenile Idiopathic Arthritis

Abstract

Objective: Our objective was to evaluate whether there is an enrichment of rare variants in familial hemophagocytic lymphohistiocytosis (HLH)-associated genes among patients with systemic juvenile idiopathic arthritis (sJIA) with or without macrophage activation syndrome (MAS).

Methods: Targeted sequencing of HLH genes (LYST, PRF1, RAB27A, STX11, STXBP2, UNC13D) was performed in patients with sJIA from an established cohort. Sequence data from control participants were obtained in silico (database of Genotypes and Phenotypes: phs000280.v8.p2). Rare variant association testing (RVT) was performed with sequence kernel association test package. Significance was defined as P < 0.05 after 100,000 permutations.

Results: Sequencing data from 524 sJIA cases were jointly called and harmonized with exome-derived target data from 3,000 controls. Quality control operations produced a set of 480 cases and 2,924 ancestrally matched control participants. RVT of cases and controls revealed a significant association with rare protein-altering variants (minor allele frequency [MAF] < 0.01) of STXBP2 (P = 0.020) and ultrarare variants (MAF < 0.001) of STXBP2 (P = 0.006) and UNC13D (P = 0.046). A subanalysis of 32 cases with known MAS and 90 without revealed a significant difference in the distribution of rare UNC13D variants (P = 0.0047) between the groups. Additionally, patients with sJIA more often carried two or more HLH variants than did controls (P = 0.007), driven largely by digenic combinations involving LYST.

Conclusion: We identified an enrichment of rare HLH variants in patients with sJIA compared with controls, driven by STXBP2 and UNC13D. Biallelic variation in HLH genes was associated with sJIA, driven by LYST. Only UNC13D displayed enrichment in patients with MAS. This suggests that HLH variants may contribute to the pathophysiology of sJIA, even without MAS.

Figure 1.. Aggregated allele frequencies from crude counts of rare qualifying variants in the sJIA cohort compared to controls.

Aggregated alternate allele frequencies were calculated as the total count of alternate alleles in all the variants in the gene, divided by the mean number of alleles identified in all the variants in the gene (to account for missing genotypes). Rare qualifying variants defined as minor allele frequency <0.01, exonic, non-synonymous, passing quality filters, present in the canonical transcript. Burden test showed significant difference in all HLH genes as a panel (p = 0.014) and SKAT showed significant association of STXBP2 (p= 0.020).

Figure 2.. Distribution of individuals with one or more rare, protein-altering HLH variants.

A bar plot displays the proportion of patients with either 1 or ≥2 rare (MAF < 0.01), protein-altering variants on canonical HLH gene transcripts among sJIA patients with a history of MAS (MAS+), sJIA patients with no history of MAS (MAS-), the full sJIA cohort and the control collection. MAS, macrophage activation syndrome. MAF, minor allele frequency. HLH, hemophagocytic lymphohistiocytosis.