Is Multiple Myeloma an Infectious Disease?

ABSTRACT & COMMENTARY

Synopsis: Bone marrow dendritic cells in all 15 patients with multiple myeloma and in 25% of patients with monoclonal gammopathy of uncertain significance (MGUS) were found to contain a DNA sequence from Kaposi sarcoma-associated herpesvirus. This virus expresses an interleukin-6-like protein which stimulates the growth of myeloma cells. Indeed, expression of viral IL-6 was also detected in samples from patients with multiple myeloma.

Source: Rettig MB, Berenson J. Science 1997;276: 1851-1854.

Multiple myeloma is the 12th most common malignancy in patients age 65 years or older,1 and the incidence may be on the rise. Although there have been reports linking myeloma to radiation exposure, this cannot account for the vast majority of cases, and to date, no clear causative agent has been identified. Recently, data from several centers in Los Angeles suggest that a virus, identified only a few years ago as being central to an AIDS-related neoplasm, may play an important role in the development of myeloma.

In 1994, Chang and colleagues reported that a novel, herpesvirus-like DNA sequence could be isolated in more than 90% of cases of epidemic Kaposi’s sarcoma, suggesting that a new member of the herpesvirus family had been identified.2 Subsequently, the genome for this virus (now referred to as Kaposi’s sarcoma-associated herpesvirus [KSHV] or human herpesvirus-8 [HHV-8]) has been characterized, and seroconversion against this virus has been shown to precede the development of Kaposi’s sarcoma. Moreover, the viral sequence has been identified in tumor specimens from patients with non-epidemic Kaposi’s sarcoma (i.e., those who are not HIV-infected).3

Recently, Chang’s laboratory reported that some of the proteins expressed by KSHV mimic human cytokines, notably interleukin-6 and macrophage inhibitory protein (MIP)-1.4 The viral interleukin-6 (vIL-6), structurally similar to human IL-6, was biologically active in assays using IL-6 dependent cell lines. Cytokines that bind to the interleukin-6 receptor have been shown to stimulate proliferation of Kaposi’s sarcoma, as well as certain multiple myeloma cell lines. If the KSHV sequence was found in one of these malignancies, might it be present in the other? Could multiple myeloma have an infectious basis?

Intrigued by this hypothesis, Rettig and Berenson sought to detect vIL-6 in the bone marrow of patients with multiple myeloma. Using polymerase chain reaction (PCR), they attempted to amplify a segment of the KSHV genome from fresh myeloma cells, but they found none in 23 bone marrow samples. This indicated that myeloma cells themselves were not infected with KSHV. Undaunted, they hypothesized that the virus could be affecting a marrow stromal cell. In the bone marrow microenvironment, stromal cells can be a major source of IL-6. These cells normally remain adherent to the marrow architecture and are present in relatively small numbers among the mononuclear cells obtained from bone marrow aspirate material.

Using cell culture techniques, bone marrow stromal cells from myeloma patients were isolated and DNA from them selectively amplified, and once again PCR was used to detect the KSHV genomic segment. This time, all 15 samples contained the appropriately sized amplified DNA segment indicating the presence of the KSHV genome. Furthermore, KSHV was detected in two of eight stromal cell cultures from patients with MGUS (monoclonal gammopathy of undetermined significance).

To confirm specificity, bone marrow material was obtained from 10 normal individuals and from 16 patients with other malignancies, including lymphoproliferative disorders, acute myelogenous leukemia, and adenocarcinoma metastatic to bone. In none of these cases was the KSHV DNA sequence detected. Southern blotting, using a known KSHV probe, confirmed the presence of the target sequence in the myeloma patients, and its absence from the non-myeloma samples. In situ hybridization demonstrated labeled KSHV DNA probe in the nucleus and cytoplasm of marrow stromal cells but not in the myeloma cells. Finally, treating the material with enzymes that selectively degrade DNA and RNA abolished the binding.

Demonstrating that the KSHV DNA sequence is present in the bone marrow of myeloma patients is only half the story. Are these infected cells expressing the vIL-6 gene? To answer this question, Rettig and Berenson next used vIL-6-specific primers to perform RT-PCR (reverse transcriptase PCR) to amplify the vIL-6 mRNA transcripts. As suspected, the vIL-6 mRNA was detected in all three samples obtained from patients with multiple myeloma, but in neither of the two samples from normal individuals.

Finally, the authors determined that the bone marrow stromal cells that contained the KSHV DNA sequences were dendritic cells as demonstrated by positive staining with antibodies to CD68, fascin, CD83 and vimentin, and the absence of staining with antibodies to CD31, CD34, lysozyme, and CD1a.

COMMENTARY

Although there is effective therapy for localized solitary or extramedullary myeloma, multiple myeloma is essentially an incurable condition, except perhaps for the very small fraction of patients under age 55 who undergo successful allogeneic bone marrow transplantation. The median survival is approximately 30 months. Thus, any novel insight into the pathogenesis of this disease is certainly welcome, as it may lead to new avenues for attack.

Previous data have suggested that autocrine stimulation of IL-6 signaling pathways are pivotal to the proliferation of myeloma cells. As early as 1993, Hata and colleagues showed that fresh, purified myeloma cells expressed IL-6.5 It would be logical to expect that if a virus was involved in the pathogenesis of this disease, it would target the plasma cell. Thus, the absence of KSHV sequences in the myeloma cells might have stopped some investigators dead in their tracks. However, the IL-6 theory of pathogenesis of myeloma has been more resilient. It has survived data that demonstrate IL-6 production in only a small fraction of cases of myeloma and data that perhaps as few as one of the dozen or so authentic myeloma cell lines actually produce and respond to IL-6. These results have been bypassed by the hypothesis that paracrine IL-6 production from marrow stromal cells is the source of the stimulation. The paracrine hypothesis has survived data showing that in only a very small fraction of cases do myeloma cells proliferate in response to IL-6 stimulation. And so it is with the infectious hypothesis. These investigators tell us that KSHV has been detected in marrow antigen presenting cells, as has v-IL-6. The IL-6 hypothesis still lives. Needless to say, we are skeptical. Why don’t myeloma patients have an increased incidence of Kaposi’s sarcoma? Do dendritic cells in other sites of myeloma patients show evidence of KSHV infection? What determines the restricted host cell type in myeloma vs. Kaposi’s sarcoma? Myeloma patients develop an immune deficiency that might make them susceptible to an opportunistic virus after they get the disease, but are they susceptible to it before they develop myeloma? Why has the virus not been isolated from myeloma patients and propagated in dendritic cells in vitro? Perhaps these are unreasonable questions.

Rettig and Berenson also suggest that the finding that two of eight patients with MGUS were found to contain KSHV sequences may explain why about 25% of patients with MGUS progress to overt myeloma. Unfortunately for the infection theory, large series of patients carefully followed for up to 30 years suggest that MGUS progresses to myeloma in only about 11% of patients. Perhaps 11% is not different than the 25% infection rate obtained in the small sample reported here, but the case for KSHV could be more strongly made by following KSHV-positive MGUS patients and finding an increased rate of progression. Even if this were found, the fact that the MGUS stage existed in such patients would argue that there are other important steps in the pathogenesis of the disease. MGUS is not myeloma.

If KSHV is involved at all in myeloma, it is likely to be as a component of a multiple hit pathway. Evidence is emerging that the CD40 pathway may play a role in myeloma. Pammer and colleagues demonstrated that expression of CD40, a member of the tumor necrosis factor superfamily of cytokine receptors, was found to be upregulated in tumor cells and endothelial cells in Kaposi’s sarcoma.6 Both myeloma cells and bone marrow stromal cells express CD40, and stimulation of CD40 signalling results in autocrine IL-6 secretion in some myeloma cells.7 Thus, it is interesting to speculate that KSHV-infected bone marrow stromal cells might exhibit upregulated CD40 ligand expression, which would enhance CD40 signalling in the tumor cells if there were stromal cell/myeloma cell interactions. Does KSHV infection upregulate CD40 ligand expression in dendritic cells?

The present study is touted as a paradigm shift in how we view myeloma. However, the discovery of a viral association in a neoplasm may not immediately translate to novel and effective therapy. For example, the discovery of Epstein-Barr viral sequences in Hodgkin’s disease has not yet changed our management. However, in the case of myeloma, the development of IL-6 antagonists and receptor blockers and the development of CD40 ligand agonists and antagonists, all of which are at hand, may permit the clinical dissection of the role of these cytokine signalling pathways in the disease. Our prediction is that the dependence of authentic myeloma cells on ongoing cytokine stimulation is likely to be almost nil, but we hope we are wrong.

References

1. Miller BA, et al. SEER Cancer Statistics Review: 1973-1990. National Cancer Institute NIH Publication No. 93-2789, 1993. National Institutes of Health, Bethesda, MD.

2. Chang Y, et al. Science 1994;266:1865-1869.

3. Moore PS, Chang Y. N Engl J Med 1995;332:1181-1185.

4. Moore PS, et al. Science 1996;274:1739-1744.

5. Hata H, et al. Blood 1993;81:3357-3364.

6. Pammer J, et al. Am J Pathol 1996;148:1387-1396.

7. Westendorf JJ, et al. J Immunol 1994;152:117-128.