Immunosurveillance and survivin-specific T-cell immunity in children with high-risk neuroblastoma.

PURPOSE: Tumor immunosurveillance influences oncogenesis and tumor growth, but it remains controversial whether clinical failure of immunosurveillance is a result of lymphocyte dysfunction or tumor escape. In this study, our goal was to characterize the physiology of tumor immunosurveillance in children with high-risk neuroblastoma (HR-NBL). PATIENTS AND METHODS: Immunohistopathologic studies were carried out on 26 tumor samples from a cohort of HR-NBL patients diagnosed at Children's Hospital of Philadelphia for the 2-year period from May 2003 to May 2005. Blood from nine HLA-A2+ patients in this cohort was analyzed for T cells specific for the antiapoptotic protein survivin. RESULTS: Survivin protein was expressed by 26 of 26 tumors. In HLA-A2+ patients, circulating cytotoxic T lymphocytes (CTLs) specific for survivin were detected by peptide/major histocompatibility complex tetramer analysis in the blood of eight of nine children with HR-NBL at the time of diagnosis. Rather than being selectively rendered anergic in vivo, circulating survivin-specific CTLs were highly functional as shown by cytotoxicity and interferon gamma enzyme-linked immunospot assays in six of nine patients. Survivin-specific CD107a mobilization by T cells was found in five of five patients. By immunohistochemistry, tumor-infiltrating T cells were few or absent in 26 of 26 tumors. CONCLUSION: Children with HR-NBL harbor robust cellular immune responses to the universal tumor antigen survivin at the time of diagnosis, but intratumoral T cells are strikingly rare, suggesting a failure of cellular immunosurveillance. Efforts to develop novel therapies that increase T-cell trafficking into tumor nests are warranted.

Multiple initial culture conditions enhance the establishment of cell lines from primary ovarian cancer specimens.

To increase the efficiency of stable cell line establishment from primary ovarian cancer specimens, we simultaneously initiated cultures under multiple conditions, varying extracellular matrices and the inclusion of supplements (e.g., serum or serum albumin), while minimizing exposure to xenogeneic antigens (e.g., fetal calf serum). Primary cultures were initiated from 30 specimens; cell lines were established from 10 of these for a success rate of 33%. In some instances, multiple cell lines were established from the same specimen. Five lines were characterized extensively with respect to growth properties, antigen expression, and genomic alterations. Although these lines are all low-passage, marked heterogeneity was observed, even between lines derived from the same specimen. The culture approach outlined herein will facilitate generation of reagents useful for many aspects of ovarian cancer biology.