OBJECTIVE Metabolic profiling is a valuable tool to characterize tumor biology but remains largely unexplored in neuroendocrine tumors (NETs). Our aim was to comprehensively assess the metabolomic profile of NETs and identify novel prognostic biomarkers and dysregulated molecular pathways. DESIGN AND METHODS Multiplatform untargeted metabolomic profiling (GC-MS, CE-MS and LC-MS) was performed in plasma from 77 patients with G1-2 extra-pancreatic NETs enrolled in the AXINET trial (NCT01744249) (study cohort) and from 68 non-cancer individuals (control). The prognostic value of each differential metabolite (N=155) in NET patients (p<0.05) was analyzed by univariate and multivariate analysis adjusted for multiple testing and other confounding factors. Related pathways were explored by Metabolite Set Enrichment Analysis (MSEA) and Metabolite Pathway Analysis (MPA). RESULTS 34 metabolites were significantly associated with progression-free survival (PFS) (N=16) and/or overall survival (OS) (N=27).13 metabolites remained significant independent prognostic factors in multivariate analysis, 3 of them with a significant impact on both PFS and OS. Unsupervised clustering of these 3 metabolites stratified patients in 3 distinct prognostic groups (1-year PFS of 71.1%, 47.7% and 15.4% (p=0.012); 5-year OS of 69.7%, 32.5% and 27.7% (p=0.003), respectively). MSEA and MPA of the 13-metablolite signature identified methionine, porphyrin and tryptophan metabolism as the three most relevant dysregulated pathways associated with the prognosis of NETs. CONCLUSIONS We identified a metabolomic signature that improves prognostic stratification of NET patients beyond classical prognostic factors for clinical decisions. The enriched metabolic pathways identified reveal novel tumor vulnerabilities that may foster the development of new therapeutic strategies for these patients.

中文翻译：

---

神经内分泌肿瘤（NET）的代谢组学特征表明，蛋氨酸、卟啉和色氨酸代谢是与患者生存相关的关键失调途径。

目的 代谢谱是表征肿瘤生物学特征的重要工具，但在神经内分泌肿瘤 (NET) 中仍未得到充分探索。我们的目标是全面评估 NET 的代谢组学特征并确定新的预后生物标志物和失调的分子途径。设计和方法 对参加 AXINET 试验 (NCT01744249)（研究队列）的 77 名 G1-2 胰外 NET 患者和 68 名患者的血浆进行多平台非靶向代谢组分析（GC-MS、CE-MS 和 LC-MS）。非癌症个体（对照）。通过针对多重测试和其他混杂因素进行调整的单变量和多变量分析，分析了 NET 患者中每种差异代谢物 (N=155) 的预后价值 (p<0.05)。通过代谢物集富集分析（MSEA）和代谢物途径分析（MPA）探索相关途径。结果 34 种代谢物与无进展生存期 (PFS) (N=16) 和/或总生存期 (OS) (N=27) 显着相关。在多变量分析中，13 种代谢物仍然是显着的独立预后因素，其中 3 种具有显着相关性。对 PFS 和 OS 都有影响。这 3 种代谢物的无监督聚类将患者分为 3 个不同的预后组（1 年 PFS 分别为 71.1%、47.7% 和 15.4% (p=0.012)；5 年 OS 分别为 69.7%、32.5% 和 27.7% (p=0.003) ）， 分别）。13 代谢物特征的 MSEA 和 MPA 确定甲硫氨酸、卟啉和色氨酸代谢是与 NET 预后相关的三个最相关的失调途径。结论 我们确定了一个代谢组学特征，该特征可以改善 NET 患者的预后分层，超越临床决策的经典预后因素。所发现的丰富的代谢途径揭示了新的肿瘤脆弱性，可能促进为这些患者开发新的治疗策略。