Pediatric and adult forms of type I autoimmune hepatitis in argentina: Evidence for differential genetic predisposition



The aim of this study was to compare major histocompatibility complex (MHC) class II susceptibility to type 1 autoimmune hepatitis (AH) between children and adults of the same ethnic group. HLA-DRB1, HLA-DRB3, HLA-DQA1, and HLA-DQB1 gene subtypes were examined by high resolution oligonucleotide typing in 122 pediatric (PAH) and 84 adult (AAH) patients and in 208 controls. In children, HLA-DRB1*1301 was the primary susceptibility allele (66.4% patients vs. 10.6% controls, relative risk [RR] = 16.3, Pc < 10−24) whereas HLA-DRB1*1302, which differs from HLA-DRB1*1301 by only 1 amino acid, appeared to be protective. The exclusion of individuals with HLA-DRB1*1301 from control and pediatric patients allowed us to find a secondary association of PAH with HLA-DRB1*0301. Possession of HLA-DRB1*1301, however, was associated with a lower therapeutic response rate. Analysis of peptide binding pocket residues indicated that Tyr 10, Ser 11, Ser 13, and Val 86 in the class II β chain were present in 85% of patients compared with 37% of controls, suggesting that a high proportion of AH susceptibility is attributable to these residues (etiologic fraction [EF] = 76%). In contrast to the class II associations in children, AAH was associated with HLA-DRB1*0405 (RR = 10.4, Pc < .005) but not with HLA-DRB1*1301 or HLA-DRB1*0301. In addition, HLA-DR4 with the class I gene, HLA-A11, appeared synergistic in predisposing AAH patients to develop extra-hepatic autoimmune (AI) manifestations (odds ratio [OR] = 104.9, Pc < 10−4). Concomitant differences in autoantibody profiles were also observed in PAH versus AAH: smooth muscle antibodies (SMA) were most prevalent in PAH but antinuclear antibodies were most prevalent in AAH (P = .003). This study therefore reveals that different HLA-DRB1 allotypes confer susceptibility to AH in children and adults and raises the possibility that PAH and AAH may be triggered by different factors.