科研之友
微信
新浪微博
Facebook
分享链接
摘要
Background. Improved prediction of neuroblastoma (NB) behavior is needed to detect treatment-refractory disease and may allow further reduction in therapy for some patients. In this regard, serum metabolomic analysis has proven utility in several cancer types. We hypothesize that serum metabolomic analysis will correlate with risk-group classification for patients with NB, and sensitively detect NB in murine xenograft models. Procedure. A pilot study was done on Children's Oncology Group (COG) tumor bank sera from 10 patients (five high-, five low-risk). An institutional pilot study was carried out on five patients comparing sera obtained during active versus minimal disease (complete response/very good partial response; CR/VGPR). Xenograft. Flank tumors were established in Nu/Nu mice by injection of NB cell lines (IMR-32, SH-EP, SK-N-AS). Serum for comparison was drawn pre-injection, at 1 week after injection when there was no visible tumor, and again once tumors were grossly visible. Comparisons were also made between tumor bearing mouse serum and supernatants from NB cell lines. Metabolomic analysis. Samples were analyzed by nuclear magnetic resonance and/or gas chromatography-mass spectrometry. Multivariate data analysis was conducted using SIMCA-P (Umetrics). Results. Serum metabolomic analysis differentiated high-and low-risk patients as well as active disease from CR/VGPR. Differences were in nitrogen, amino acid, and carbohydrate metabolism, as well as ketosis. The serum metabolomic signature in murine xenograft models sensitively detected NB cells and correlated with disease burden. Similar metabolic changes attributable to NB were noted in both human and murine serum. Conclusions. Serum metabolomic analysis can distinguish several characteristics of NB. A larger analysis of COG banked sera is warranted.
- 出版日期2016-2
