Article Text
Abstract
Background MS is an autoimmune disease of complex aetiology characterised by variable age related pathology including an early inflammatory process reflected by clinical relapses and later neurodegeneration associated with slow accumulation of fixed disability. However, the relationship between these stages remains poorly understood. Recent advances in genetic analysis have improved our understanding of the pathophysiological processes but an understanding of the contribution of genetic factors to disease phenotype remain elusive, and although family studies have demonstrated convincing evidence of a genetic influence on disease phenotype, no such effect has been demonstrated in superficial analysis of disease associate SNPs in large GWAS. This suggests that any effects on disease expression are likely to be complex and multigenic. In this study we have examined the effect of interactions between genes on phenotypic aspects of MS.
Methods DNA from 1003 MS patients was genotyped for the 29 SNPs known to be most strongly associated with MS. Association with age at onset was tested using an extremes of outcome model by comparing paediatric and late onset disease. Associations with primary progressive MS (PPMS) was tested using chi–squared testing. Effects on time to secondary progression (SPMS) were analysed using Cox proportional hazards regression, corrected for age at onset and gender. Genotypes were tested individually and in combination to assess epistatic effects. Analyses were performed in PLINK v1.07_64 and R v2.14.1 and corrected for multiple testing using permutation analysis.
Results 703 patients (69.9%) were female, and mean follow up was 12.0 years (SD 10.2). Mean age at onset was 30.5 years (SD 12.1). 486 patients had relapsing remitting MS (RRMS), 413 patients had SPMS, and 95 patients (9.4%) had PPMS. 63 patients had onset <18yr (paediatric–onset (POMS)) and 54 patients had onset >50yr (late–onset (LOMS)).
Epistatic effects were found for associations with POMS: CD40 interacted with HLA–DQA2 (OR 0.18, p=0.0068), as well as with CXCR5 (OR 4.55, p=0.0033) and with IRF8 (OR 3.87, p=0.0050). CD40 was independently associated with PPMS (OR 1.54, 95% CI 1.03–2.3), in addition to an epistatic interaction with HLA–DRB1*1501–HLA–DQA1 (OR 0.33, p=0.0090). There was a further epistatic interaction between KIF21B and IL12 (OR 0.42, p=0.0074). KIF21B was significantly associated with time to SPMS when corrected for IL12 (HR 0.69 (0.50–0.95), p=0.021). Significant interactions were also found between IL7R and RGS1 (HR 0.64 (0.43–0.97), p=0.033), HLA–DQA1 and CD58 (HR 0.52 (0.28–0.99), p=0.045), and IL7R and CLECL1 (HR 1.37 (1.04–1.81), p=0.025).
Conclusions Our study begins to explore possible epistatic effects between genes in MS employing extremes of outcome, which increases power and may be able to detect effects that conventional models are underpowered for. Our findings suggest that interactions between antigen presenting cell activation and other components of the inflammatory cascade predispose patients to early–onset disease, and indicate a possible role for KIF21B in the onset of disease progression. We have set out a methodology for exploration of epistatic effects on MS phenotype which requires confirmation in replication studies.
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
Statistics from Altmetric.com
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE
- PARKINSON'S DISEASE
- STROKE