Khalique et al., pp 1579–1586 The evolution of primary cancers to metastases is not fully understood. One commonly accepted theory, linear clonal evolution, proposes that the accumulation of multiple genetic alterations, which offers a selective advantage and eventually a metastatic potential, in one subclone results in the sequential expansion of that clone within a primary cancer to a metastatic tumor. Alternative theories – parallel clonal expansion and same gene model – have also been proposed to explain metastatic development with variable supporting evidence. In this report, the authors investigate clonal evolution of metastases in ovarian cancer. To do this, they selected 22 patients with high-grade serous epithelial ovarian cancer. These patients had developed metastases at various sites and, in some, various regions were identified within their primary tumor. Using microsatellite analysis, Khalique et al., established that intratumor genetic heterogenecity was slightly higher, though not statistically different, in primary than in metastastic cancers. They analyzed the clonal relationship between different regions of the primary and metastatic tumors. Metastatic cells are highly likely to be related to the primary tumor. However, unlike linear clonal evolution, in 65% of patients the genetic profiles suggested that metastatic lesions can develop from both early and late stage clones within the primary tumor. The authors concede that the genetic markers they used in this study have no direct functional role in tumorigenesis. However, this study demonstrates that metastatic development in ovarian cancer is likely to have a more complicated pathway than linear clonal expansion. Schwinn et al., pp. 1594–1604 Recent data suggest that triggering of natural killer (NK) cell co-stimulatory molecule NKG2D on cytotoxic lymphocytes (CTLs) by ligands expressed on melanoma cells greatly enhances the killing capacity of the CTLs. Skin-resident NK cells express this co-stimulatory molecule NKG2D upon activation. MHC-negative melanoma cells, a common phenotype in patients, should be highly susceptible to NK attack but killing appears to be dependent on the presence – or not – of NKG2D ligands (NKG2DL) on the target cells. In this study Schwinn and colleagues wanted to unravel factors that influence NKG2DL expression in melanoma cells and the subsequent fate of these MHC-negative melanoma cell lines in the hands of NK cells. For this, the authors explored the role of IFN-γ on the expression of NKG2D ligands – MICA and ULBP2. A MHC class I negative melanoma cell line established from a patient was first confirmed to be susceptible to NK killing. The authors then showed that IFN-γ could downregulate NKG2DLs, MICA in particular, in melanoma cells in a dose- and time-dependent manner. This mechanism was confirmed to act at the mRNA level. The signaling molecule STAT-1 appears to be involved in this process as STAT1 silencing dampened the influence of IFN-γ on MICA expression. Most interestingly, IFN-γ could impair NK lysis of MHC-negative melanoma cells by about 50% and this effect was shown to be largely dependent on NKG2DL expression. . Lysis of MHC Class I negative melanoma cells - Mel249 abd Ma-Mel-86b - is inhibited by the presence of IFN-γ and an anti-NKG2D blocking antibody. This study reveals an intriguing effect of IFN-γ on NKG2D ligand expression in melanoma cells that results in reduced NK killing of these cells. The biological function of this finding is not clear; however, increasing NKG2D ligand expression on MHC class I negative tumor cells may be a potential route to enhance anti-tumor immunity. Moormann et al., pp. 1721–1726 Children in equatorial Africa are at high risk for endemic Burkitt lymphoma (eBL). This incidence is linked to Epstein-Barr virus (EPV) infection at a young age and holoendemic malaria exposure. Whether a lack of immune control of EBV, specifically Epstein-Barr nuclear antigen 1 (EBNA1)–which is the sole viral latent antigen expressed in BL tumors, is linked to tumor BL development is an open question. In this report, the authors investigate EBV-specific immune function in Kenyan children with eBL compared to healthy children who live in regions with divergent malaria exposure. To do this, Moormann et al., isolated PBMCs from eBL and healthy children and stimulated them with overlapping peptides of EBNA1 and measured the EBV-specific IFN-γ T-cell responses by ELISPOT. Only 16% of eBL children responded to IFN-γ as compared to over 66% of the healthy controls. On the other hand, eBL children could perfectly well respond to EBV epitopes not expressed in tumors (CD8+ T cell response) and a malaria antigen recognized by CD4+ T cells. In addition, eBL children were found to show a normal humoral response as gauged by IgG1 anti- EBNA1 levels. In short, eBL children appear to have a selective loss of EBNA1-specific IFN-g-secreting T cells. The protective effect that such T cells may elicit against eBL cannot be answered by this study, but certainly warrants further investigation.