Lack of serologic association between human herpesvirus-8 infection and multiple myeloma and monoclonal gammopathies of undetermined significance.

An interesting and provocative study by Rettig et al. (1) has raised the possibility that a virus—human herpesvirus 8 (HHV-8), also known as Kaposi’s sarcoma-associated herpesvirus—may be involved in multiple myeloma etiopathogenesis. HHV-8, a new member of the herpesvirus family, has been associated to Kaposi’s sarcoma, body cavitybased lymphoma, and multicentric Castleman’s disease. More than 80% of Kaposi’s sarcoma patients have detectable antibodies to HHV-8; the general U.S. population has a frequency of 4%– 25%, depending on method of analysis, geographic distribution, age, and sex. Rettig et al. (1) reported that HHV-8 DNA sequences were detected in bone marrow dendritic cells of patients with multiple myeloma (15 of 15 cases), as well as of patients with monoclonal gammopathies of undetermined significance (MGUS) (two of eight cases). In contrast, no HHV-8 DNA was detected in bone marrow from 26 control subjects. Furthermore, the HHV-8 sequences appeared to be transcriptionally active because a viral homologue of human interleukin 6 (vIL-6) was detected in cell cultures prepared from patients with multiple myeloma (three of three cases). Detection of HHV-8 DNA in myeloma samples has been confirmed by Brousset et al. (2). However, contrasting results were obtained by other researchers (3–4). To address the issue of association between HHV-8 infection and multiple myeloma, we performed a serologic analysis of blood samples prepared from 36 patients with multiple myeloma and 19 patients affected by MGUS. Twenty-five healthy blood donors and 10 patients with acquired immune deficiency syndrome (AIDS)– Kaposi’s sarcoma were also tested as control subjects. Coded sera from experimental and control groups were screened with use of an immunofluorescence-based assay for the presence of antibodies reacting with HHV-8 latency-associated antigen and lyticassociated antigens on phorbol esterinduced BCBL-1 cells. Sera were also assayed for anticomplement immunofluorescence to detect antibodies directed against Epstein-Barr virus nuclear antigens and by enzyme-linked immunosorbent assay to detect antibodies against human cytomegalovirus. The slides were interpreted independently by at least two of us (R.S., A.A., and A.F.), who were blinded to the clinical source of each sample. The results are presented in Table 1. Only three out of 36 myeloma patients developed antibodies against HHV-8 latency-associated antigens, but not against lytic-associated antigens. None of the serum samples from MGUS patients had antibodies against either latent-associated or lyticassociated antigens. Analysis of antibody reactions to Epstein-Barr virus nuclear antigens and cytomegalovirus showed no evidence of reduced antibody response in either group of patients affected by multiple myeloma or MGUS, compared to control subjects. AIDS–Kaposi’s sarcoma patients were positive for either latent-associated antigens (five of 10 cases) or lyticassociated antigens (seven of 10 cases); four sera yielded positive reactions for both antigens. In addition, one of the healthy control subjects was positive for both latent-associated antigens and lytic-associated antigens. In summary, our results agree with previous studies (3–7) showing no serologic association between HHV-8 infection and multiple myeloma and MGUS. The low prevalence of HHV-8 antigens detected in patients with multiple myeloma and those with MGUS makes it difficult to envision the possibility that HHV-8 infection is present at a high rate in these groups of patients, unless the viral infection is silent or, as suggested by Rettig et al. (1), that the detection of vIL-6 transcripts ‘‘in vitro,’’ is limited to a restricted pattern of expression, yet to be determined. Further studies, both at the molecular and serologic levels, are needed to establish the potential role of HHV-8 in the etiopathogenesis of multiple myeloma.