Tumor necrosis factor (TNF) alleles have been associated with systemic sclerosis (SSc); however, these alleles may be in linkage with other genes. Allograft inflammatory factor-1 (AIF-1) is a newly identified gene on the short arm of chromosome 6 in the class III region of the human leukocyte antigen. It appears to be involved in inflammation and was originally identified in rat cardiac allografts undergoing rejection. AIF-1 has several sequence variations (single nucleotide polymorphisms, SNPs), three of which result in nonsynonymous changes in amino acid coding. We analyzed the linkage of five TNFA and five AIF-1 SNPs by polymerase chain reaction in 239 Caucasian individuals. The TNFA-1031T/T genotype was found to be associated with SSc (P < 0.0001) and both the DcSSc (diffuse subset of SSc) and the LcSSc (limited subset of SSc) subsets (P= 0.0004 and P= 0.0009, respectively) and the TNFA-237G/G genotype was found to be associated with all SSc (P= 0.0003) and with the DcSSc and LcSSc subsets (P= 0.01 and P= 0.005, respectively). Furthermore, the TNFA-857C/T genotype was associated with LcSSc (P= 0.0003) and TNFA-307A/A genotype associated with DcSSc (P= 0.028). In AIF-1, RS2269475 exon 4A allele, which generates a nonsynonymous change (tryptophan to arginine), was significantly associated in patients with SSc (P= 0.0009) and was associated with those patients who had DcSSc (P= 0.002). A strong linkage disequilibrium was observed between the AIF-1 alleles, A allele of RS2269475 and the A allele of RS4711274 (P < 0.0001), and linkage was observed between AIF-1 and TNFA alleles. Here, we report a novel and significant association of a nonsynonymous change within the AIF-1 with SSc and identified the linkage with TNFA alleles within 50 kb of this gene. Our study lends support that TNFA may be an important inflammatory modulator in SSc and may play a significant role with AIF-1 in disease pathogenesis.

To compare the characteristics of patients with systemic sclerosis who died within 2 years of diagnosis to those who died after 2 years of diagnosis. A retrospective chart review of all medical records of deceased systemic sclerosis (SSc) patients who had been followed at our institution from 1985 to 2007 was performed. We identified 87 deceased SSc patients within this period. From the 87 deceased individuals, 20 had died within 2 years after they were diagnosed, and 67 had died after 2 years of their diagnosis. Patients who died within 2 years of diagnosis were more likely to be anticentromere antibody negative when compared to the patients who died after 2 years (17/20 vs. 48/67, P = 0.03). The time from the first appearance of non-Raynaud's symptoms to diagnosis was significantly shorter in the group who died within 2 years than in the group who died after 2 years of diagnosis (11.8 ± 10.3 vs. 60.7 ± 64.9 months, P = 0.002). According to the Medsger severity score, there was more severe muscle (0.82 ± 1.13 vs. 1.8 ± 1.28, P = 0.0014) and heart (0.86 ± 1.37 vs. 2.1 ± 1.71, P = 0.0013) involvement at the initial evaluation in patients who died before 2 years of diagnosis when compared to the group of patients who died after 2 years of diagnosis. The time from the first symptoms to treatment initiation was significantly shorter in patients who died early (9.43 ± 6.3 vs. 38.3 ± 54.4 months, P = 0.05). The interval between treatment initiation and death was also significantly shorter (15.1 ± 9.48 vs. 60.7 ± 49.7 months, P = 0.001), reflecting greater severity of disease. Patients who died within the first 2 years of SSc diagnosis were typically anticentromere negative and had significant muscle and cardiac involvement. The time from the first appearance of non-Raynaud phenomenon symptoms to death was much shorter in the patients who died within 2 years of diagnosis, suggesting a very fulminant form of systemic sclerosis.