Search

Fifty-four (54) unrelated patients with Mediterranean Kaposi's sarcoma (MKS) and 8 patients members of 4 unrelated families with familial MKS were serotyped for HLA-A,B and DR antigens. The diagnosis was histologically confirmed and all patients were negative for anti-HIV antibodies. An increased frequency of HLA-B18 (44.4% vs 14.2% in the controls, p < 0.001, RR = 4.8) and HLA-DR5 (57.6% vs 37.2% in the controls, p < 0.025, RR = 2.29) was observed in the group of patients with MKS. Seven (7) of the 8 family members with FMKS possessed HLA-DR5, and the affected members in the 3 families shared a common haplotype which included HLA-DR5. These findings support the hypothesis that genetic factors linked to HLA-DR5 antigen may contribute to the pathogenesis of MKS.

Organ-specific cardiac antibodies are serological markers of autoimmunity in dilated cardiomyopathy (DCM). HLA-DR4 and possibly DR5 are immunogenetic markers of susceptibility in DCM, but it is not known whether they are associated with autoantibody production. We studied the frequency of HLA-DR antigens and the presence of organ-specific cardiac antibodies in 80 DCM Caucasian patients from Northern Italy. HLA-DR typing was performed by serology; 289 healthy blood donors from the same region were tested as controls. HLA-DR frequencies in DCM were also compared with VIII International Workshop control data for Italy. Cardiac antibodies were detected by indirect immunofluorescence on human heart. Skeletal muscle was used to identify cross-reacting antibodies. The prevalence of cardiac antibodies in DCM was: organ-specific 34% and skeletal muscle cross-reactive 30%. The previously reported positive association between DCM and HLA-DR4 was confirmed using either the controls from the same region (21.25% vs 10.73% p = 0.02, relative risk = 2.30) or from all of Italy (21.25% vs 12.3%, p = 0.03). HLA-DR5 frequency was slightly but not significantly higher in DCM than in controls from the same region (46.25% vs 31.49% p = 0.02, relative risk of 1.87, p corrected = NS) or from all of Italy (46.25% vs 35.8% p = NS). HLA-DR3 frequency was lower in DCM than in controls from the same region (12.50% vs 29.41% p = 0.003, relative risk of 0.36, p corrected = 0.03). This negative association was not confirmed using the control data from the whole of Italy (12.50% vs 16.5% p = NS).(ABSTRACT TRUNCATED AT 250 WORDS)

In a 5-year-study of HLA-phenotypes in 411 HIV-1-infected individuals, a progressive decrease of the formerly elevated frequency of HLA-DR5 has been observed. HLA-DR3 seems to have a protective effect. These results are discussed with respect to the mimicry-hypothesis of HLA and disease associations. Further preliminary results indicate that HIV-associated Kaposi's sarcoma could be associated with HLA-A28, and therefore might be a different etiologic entity than HIV-infection alone.