Pemphigus is a severe autoimmune disease characterized by circulating and bound in vivo pemphigus autoantibodies. It was revealed that the autoantibodies occur in healthy first-degree relatives of pemphigus patients; however, their significance is not fully elucidated. Thus, the aim of the study was to assess the frequency of circulating IgG pemphigus autoantibodies in the healthy relatives of pemphigus patients and of their ability to bind in vivo in the epidermis. We also analyzed IgG subclasses distribution, both in the serum-positive relatives and in the patients. Our study included 67 healthy relatives, 50 healthy normal controls and 33 patients (25 at an active stage of the disease, 8 in clinical remission). To detect circulating pemphigus antibodies we applied indirect immunofluorescence and anti-desmoglein ELISA. Monoclonal anti-human IgG1, IgG2, IgG3, IgG4 antibodies were used to assess subclass distribution. The frequency of circulating pemphigus autoantibodies in the relatives, detected by IIF (30/67) was statistically higher (P < 0.001) than in the control group (0/50). ELISA revealed anti-desmoglein 1 and/or 3 antibodies in 13 out of 67 relatives. Direct immunofluorescence performed in 25 out of 32 seropositive relatives did not show intercellular bound in vivo IgG and/or C3 in the epidermis in any cases. Circulating IgG2 subclass was observed in 60% of the examined relatives and IgG4 was detected in 23.3% of them. In the patients at an active stage of pemphigus IgG4 and IgG1 were the dominant subclasses (96 and 76% relatively) while in clinical remission antibodies predominantly belonged to the IgG2 (75%) and IgG4 (37.5%) subclass. The obtained results confirmed polyclonal production of pemphigus autoantibodies and their different distributions dependent on the disease activity. Statistical analysis showed that the frequency of IgG1 and IgG4 subclasses was significantly higher in the patients at an active stage of the disease when compared to the patients in clinical remission (P < 0.001) or with seropositive healthy relatives (P < 0.001). The relevance of the presence of IgG4 autoantibodies in the healthy relatives' sera requires further studies that focus on their potential pathogenicity.