¹Service de Dermatologie, Hôpital Cochin, Laboratoire de Recherche en Dermatologie, Unité UPRES1833, Faculté de Médecine Paris V, Université René Descartes, Paris, France

Correspondence: Prof. Nicolas Dupin, Service de Dermatologie, Unité UPRES1833, Hôpital Cochin, 89 rue d'Assas, 75006 Paris, France. Email: nicolas.dupin@cch.ap-hop-paris.fr

Abstract

The level of Kaposi's sarcoma-associated herpesvirus (KSHV) cellular viremia in patients with Kaposi's sarcoma could be related to both the tumor burden and its progression. The origin of the spindle cell, the hallmark cell of Kaposi's sarcoma, is still debated. Two hypotheses may be formulated, favoring a lymphatic endothelial cell or an endothelial-cell precursor evolving into a lymphatic phenotype, both preferentially targeted by KSHV.

Kaposi's sarcoma (KS) is an opportunistic tumor that occurs in HIV-infected patients (AIDS-KS) and in the setting of solid-organ transplantation (iatrogenic KS) (Boshoff and Weiss, 2002). KS is also reported in patients of Mediterranean origin (classic KS) and sub-Saharan African origin (endemic KS), without any major immune impairment clearly identified. Kaposi's sarcoma-associated herpesvirus (KSHV) is a gammaherpesvirus that was discovered in a patient with AIDS-KS and was also named human herpesvirus 8. Using PCR techniques, several groups have detected KSHV in the vast majority of KS tumors, and it has been shown to be associated with all epidemiologic forms of KS (Chang et al., 1994; Boshoff and Weiss, 2002).

Morphologic studies based on immunohistochemistry or electron microscopy have shown that most of the cells in KS lesions are latently infected, although a small proportion of the cells of unknown origin may contain replicative virus that may have a key role in local tumor progression (Orenstein et al., 1997; Grundhoff and Ganem, 2004). The presence of KSHV DNA in the blood of AIDS–KS patients has been reported in more than 50% of cases (Whitby et al., 1995). More interestingly, it has been shown that KSHV cellular viremia is associated with both the development and the progression of KS in HIV-infected individuals (Whitby et al., 1995). These results suggest that the detection of KSHV in blood could predict the evolution of KS in the setting of HIV infection. KSHV is also associated with the multicentric form of Castleman's disease, a rare lymphoproliferative disorder that is associated with an excess risk of KS, as nearly 75% of AIDS patients with Castleman's disease have KS. KSHV viremia is clearly associated with clinical activity in the setting of Castleman's disease, and, unlike in KS, 10%–15% of the cells support lytic infection in Castleman's disease tumors (Oksenhendler et al., 2000).

The value of measuring KSHV viremia has also been reported in patients with iatrogenic KS, but not in patients without any apparent immune impairment, such as patients with classic or endemic KS. In this issue, the group led by Céleste Lebbé adds important insights on the value of KSHV viremia in both classic KS and endemic KS (Pellet et al., 2006). The study involved 70 patients with classic KS and 11 patients with endemic KS. Using quantitative PCR, the authors confirmed that KSHV cellular viremia reflects tumor burden and demonstrated that it can be correlated to tumor progression. These findings suggest that it may be useful to monitor KSHV cellular viremia in patients with KS but without HIV infection, such as patients with classic or endemic KS.

KSHV may be detected in blood, but there are few data on the status of the virus in peripheral blood mononuclear cells. One study showed that the virus may replicate in peripheral blood mononuclear cells while it is latent in KS lesions (Decker et al., 1996). Therefore, given the correlation between tumor progression and KSHV cellular viremia, it could be suggested that antiherpetic drugs such as cidofovir, which have been shown to decrease KSHV replication in vitro, may impede KS progression (Kedes and Ganem, 1997). However, the only study evaluating cidofovir activity in patients with KS showed no improvement at all (Little et al., 2003). What kind of information, then, can we expect to gain from cellular viremia in the setting of KS?

1. Estimation of tumor activity and tumor extent;
2. Improved accuracy of prognosis, which could assist in therapeutic decision making;
3. Measurement of efficacy or failure after starting a treatment.

Given the low level of cellular viremia, is there a place for monitoring of such loads in the management of classic and endemic KS in the future? Probably, but the low detection level of viremia observed in these patients must be improved by an increase in the sensitivity of detection in order to precisely assess the correlation between KSHV viral load and KS staging and progression in patients with classic or endemic KS.

The other important point raised by Pellet et al. (2006) concerns the nature of the cells infected by KSHV in blood. Previous reports have demonstrated that B lymphocytes are infected preferentially, but there is still controversy about infection of T lymphocytes and monocytes (Ambroziak et al., 1995). In this study, Pellet et al. confirmed that B lymphocytes and, to a lesser extent, monocytes may be infected, and they demonstrated that 50% of patients have circulating endothelial cells infected by KSHV. This very important finding is directly linked with the pathogenesis of KS. KS is a complex angiogenic tumor characterized by both the presence of ectatic vessels and the proliferation of spindle-shaped cells, which are the hallmark cells of this neoplasm. Although the origin of the spindle cells is still undetermined, it was shown that they are a heterogeneous population of cells expressing markers for smooth muscle cells, macrophages, and dendritic cells. However, most of the spindle cells in KS lesions express endothelial markers, including CD31 and CD34, and recent studies suggest that spindle cells are probably of endothelial origin, although the lymphatic or vascular nature of these cells is still debated, and some recent works suggest that they may be more of lymphatic origin. Indeed, when KSHV infects microvascular endothelial cells, it may drive the cells to lymphatic reprogramming with induction of the main lymphatic lineage-specific genes, including PROX1, which play a major role in lymphatic development (Hong et al., 2005). Moreover, one study reports that the virion envelope-associated glycoprotein B can activate vascular endothelial growth factor receptor-3 on the surface of microvascular endothelial cells and that vascular endothelial growth factor receptor-3 expression enhances KSHV infection (Zhang et al., 2005). Because spindle cells express lymphatic endothelial markers such as vascular endothelial growth factor receptor-3, lyve-1, and podoplanin and are also the main cells harboring KSHV in KS lesions, it may be hypothesized that lymphatic endothelial cells are the favored targets of infection. This is also illustrated by the fact that in KS lesions, KSHV is not present in resident blood vessels despite the production of KSHV virions in such tumors (Dupin et al., 1999). However, another hypothesis that has been addressed is that KSHV may infect endothelial-cell precursors and drive these cells toward a lymphatic endothelial-cell phenotype. Because neoangiogenic cells associated with malignant tumors may express some of the lymphatic markers, the specificity of these observations should be interpreted with caution (Jussila et al., 1998).

Finally, Pellet et al. (2006) demonstrated the absence of an impairment of immune response, as they did not find any significant decrease in circulating IFN--producing T cells in patients with KS. Moreover, the detection of IFN--producing T cells was not associated with negative KSHV viremia; this suggests that IFN--deficient activity is not implicated in the control of KSHV production in KSHV-infected patients. So why do subjects with no specific immunosuppression develop KS? It may be due to a KSHV-specific immune defect that would allow the virus to escape immune-cell control, as some studies have shown that patients with KS lack specific immune cellular response, in contrast to KSHV-infected patients without KS. This study suggests that monitoring KSHV viremia may be useful in patients with classic and endemic KS as long as these patients harbor no apparent immune impairment. The predictive value of KSHV detection in blood for KS development is, however, not clear, unlike that of KSHV detection in HIV-infected patients. Other factors such as genotype and specific cellular immune dysfunction may be associated with KS development and progression in patients with non-HIV-related KS.

However, many questions about the origin of KSHV-infected circulating endothelial cells still arise (Figure 1). Do these cells come from KS lesions recirculating in blood? Or do they originate from medullar precursors and thereafter seed in the skin and promote development of KS? Why do patients with KS have an increased number of circulating endothelial cells? Cause or consequence of KS lesion: The question is still unanswered and deserves to be explored by further studies. What is the next step? Probably to demonstrate that the circulating endothelial cells infected by KSHV and the spindle cells found in KS tumors share common features — especially the lymphatic markers, as spindle cells may be derived from lymphatic endothelial precursors.

Figure 1.

Hypothesis of KSHV-infected circulating endothelial cells in Kaposi's sarcoma development. (1) Blood compartment with Kaposi's sarcoma-associated herpesvirus (KSHV)-infected circulating endothelial cells (brown diamonds). (2) Infected circulating endothelial cells may seed in the skin (this property may be acquired secondary to KSHV infection). (3) Infected circulating endothelial cells may interfere with lymphatic endothelium cells (blue ovals). The resulting mixed population of infected and uninfected cells may participate in the formation of ectatic vessels (initial stage of Kaposi's sarcoma). (4) KSHV may confer a growth advantage to infected cells, promoting their expansion (nodular stage of Kaposi's sarcoma). (5) Infected cells may shed into the blood compartment, contributing to an increase in KSHV viremia and leading to development of new lesions.

Full figure and legend (31K)