The major histocompatibility complex (MHC) on chromosome 6, differs from the
IL10 genomic region, with several examples of long range LD. Although the blocklike microstructure of this region is similar (
Fig. 1), longer range LD arises in a subset of MHC haplotypes because of linkage between segments of strong LD. The
TNFA and
LTA genes are tandemly arranged in the MHC class III region—only 1.2 kb separates the polyadenylation site of
LTA and the TSS of
TNFA. Strong LD has been demonstrated between
TNF coding and promoter SNPs, as far upstream as the
LTA+
252 SNP, and 8 kb downstream to the
TNFd microsatellite locus, in intron 4 of
LST1.
61 Both loci are linked to extended haplotypes across the human MHC region and to the transcriptional activity of
TNFA.
37,38 Some variability in the results of these studies may be attributable again to differences in experimental method and the chance representation of different haplotypes in this region.
55,56 Nevertheless, the
TNFA−
308A polymorphism associates with the
LTA+
252G polymorphism in several combined
HLA-TNF-LT haplotypes
62 ; and these haplotypes have been linked to TNF production levels and numerous immunopathologic diseases by several independent studies.
63–65 Furthermore, the
TNFd4 allele is linked to high TNFα production by leukocytes in vitro,
38 and the
TNFd4b allele shows strong LD with
TNF−
238A.
48 This may explain why, in this study, carriers of
LTA+
252G,
TNFhtSNP1A(−308A) and
TNFhtSNP2A(−238A), associated with increased TNFα production, have a greater chance of developing uveitis, and
TNFd4 carriers were more frequently patients with nonremitting ocular inflammation or those requiring higher levels of immunosuppression. We did not demonstrate any associations between
TNF polymorphisms and visual outcome. While a census date was chosen to make a final assessment of clinical course that was consistent across patient groups, the duration of disease at this date did vary between patients. Furthermore, both measures were likely confounded by ongoing disease activity and/or the incidence of cataract and other complications. Further analysis of visual outcome in uveitis would be desirable, correlated with the development of complications and disease duration. Such an analysis would necessitate a larger prospective study with appropriate statistical power.