Absence of a Correlation Between Kaposi’s Sarcoma–Associated Herpesvirus (KSHV/HHV-8) and Multiple Myeloma

To the Editor:

Kaposi’s sarcoma–associated herpesvirus (KSHV/HHV-8) has been found within tumor tissue in Kaposi’s sarcoma, Castleman’s disease, and primary effusion lymphoma (PEL). KSHV has been shown to bear several genes with functional homology to cellular genes, including interleukin-6 (IL-6).1 Human IL-6 may stimulate myeloma cell growth, and reports by Rettig et al2,3 and Said et al4 have recently shown that KSHV could be isolated from cultured bone dendritic marrow cells from patients with multiple myeloma (MM), Waldenstrom’s macroglobulinemia, or primary amyloidosis, but not from normal individuals or patients with other hematological diseases. To investigate this association we looked for antibodies to KSHV in sera, and for KSHV genome in bone marrow samples from patients with MM and other diseases.

One hundred eighty individuals were enrolled (Table1): 99 patients with advanced-stage MM, 2 patients with monoclonal gammopathy of undetermined significance (MGUS), 14 patients with non-Hodgkin’s lymphoma, 19 patients with breast cancer, 34 normal individuals, 6 Epstein-Barr virus (EBV) IgG/IgM seropositive individuals, 2 human immunodeficiency virus (HIV)-positive patients with Kaposi’s sarcoma, and 4 HIV-positive patients without Kaposi’s sarcoma. Patient groups (>10 points) were matched for age. Serum probes of myeloma, lymphoma, and breast cancer patients were taken between 1 and 3 weeks before peripheral blood stem cell transplantation and frozen. Serum was used in an immunofluorescent assay to detect antibodies to lytic-phase proteins expressed in KS-1 cells, a PEL-derived EBV⁻ cell line that supports KSHV replication after stimulation. Sera from patients with Kaposi’s sarcoma taken as a positive control stained positive for KSHV. To exclude false-positive results due to the known homology between KSHV and EBV antigens, we tested sera from 6 patients with EBV reactivation. These sera showed no KSHV seropositivity. Sera from HIV patients without Kaposi’s sarcoma also showed negative results. Finally, polyvalent IgG (Endobulin; Baxter-Immuno, Heidelberg, Germany), administered to 72 of 99 patients with MM, was tested to exclude false-positive staining. We found no difference in seroprevalence of KSHV between MM, NHL, breast cancer patients, or normal individuals (Table 1).

In view of the low HHV-8 seroprevalence in MM we looked at the remission status at the time of sampling. Complete remission was found in 20% of cases, arguing against an inhibitory effect of paraprotein. In addition, concurrent tests for EBV and CMV IgG antibodies showed no significant difference between control and MM patients (90%/91% EBV IgG⁺, 67%/75% CMV IgG⁺).

From 43 patients with advanced-stage MM we tested fresh bone marrow aspirate samples for KSHV genome. The probes were analyzed for the presence of the virus genome by a nested polymerase chain reaction (PCR) that amplifies the KS330 233 sequence of KSHV. In only one patient could this sequence be detected, and this patient was also seropositive for KSHV. Dendritic cells cultivated from bone marrow from 25 of these patients were also negative for KSHV genome (manuscript submitted).

Our results confirm the French and United Kingdom5,6results that report no increased seroprevalence for KSHV in MM compared with healthy controls or other disease groups, and contradict that of Alsina et al.7 In addition we failed to find evidence by nested PCR of the presence of KSHV genome in the majority of MM patients or controls, either in bone marrow aspirates or dendritic cells cultured from those aspirates. It is difficult to reconcile our results with the PCR findings reported by Rettig et al, unless one assumes that exceedingly low levels of infected dendritic cells are capable of inducing antigen-specific B-cell tolerance within the bone marrow. It has been shown in studies with transgenic mice that antigen located in the bone marrow leads to clonal deletion or anergy of antigen-specific B cells⁸; however, such a mechanism has not yet been described for KSHV. To investigate this possibility we are currently following the KSHV antibody status of seropositive MM patients and controls in longitudinal studies.