Houston Methodist Cancer Center, Houston, Texas, USA.
Center for Precision and Environmental Health, Baylor College of Medicine, Houston, Texas, USA.
Cancer Med. 2023 Jan;12(1):584-596. doi: 10.1002/cam4.4933. Epub 2022 Jun 8.
Non-small cell lung cancer (NSCLC) comprises the majority (~85%) of all lung tumors, with lung adenocarcinoma (LUAD) and squamous cell carcinoma (LUSC) being the most frequently diagnosed histological subtypes. Multi-modal omics profiling has been carried out in NSCLC, but no studies have yet reported a unique metabolite-related gene signature and altered metabolic pathways associated with LUAD and LUSC.
We integrated transcriptomics and metabolomics to analyze 30 human lung tumors and adjacent noncancerous tissues. Differential co-expression was used to identify modules of metabolites that were altered between normal and tumor.
We identified unique metabolite-related gene signatures specific for LUAD and LUSC and key pathways aberrantly regulated at both transcriptional and metabolic levels. Differential co-expression analysis revealed that loss of coherence between metabolites in tumors is a major characteristic in both LUAD and LUSC. We identified one metabolic onco-module gained in LUAD, characterized by nine metabolites and 57 metabolic genes. Multi-omics integrative analysis revealed a 28 metabolic gene signature associated with poor survival in LUAD, with six metabolite-related genes as individual prognostic markers.
We demonstrated the clinical utility of this integrated metabolic gene signature in LUAD by using it to guide repurposing of AZD-6482, a PI3Kβ inhibitor which significantly inhibited three genes from the 28-gene signature. Overall, we have integrated metabolomics and transcriptomics analyses to show that LUAD and LUSC have distinct profiles, inferred gene signatures with prognostic value for patient survival, and identified therapeutic targets and repurposed drugs for potential use in NSCLC treatment.
非小细胞肺癌（NSCLC）占所有肺癌的大部分（约 85%），其中肺腺癌（LUAD）和鳞状细胞癌（LUSC）是最常见的组织学亚型。多模态组学分析已在 NSCLC 中进行，但尚无研究报道与 LUAD 和 LUSC 相关的独特代谢物相关基因特征和改变的代谢途径。
我们整合了转录组学和代谢组学来分析 30 个人类肺肿瘤和相邻的非癌组织。差异共表达用于鉴定正常组织和肿瘤之间改变的代谢物模块。
我们确定了 LUAD 和 LUSC 特有的独特代谢物相关基因特征，以及在转录和代谢水平上均失调的关键途径。差异共表达分析表明，肿瘤中代谢物之间的一致性丧失是 LUAD 和 LUSC 的主要特征。我们确定了一个在 LUAD 中获得的代谢癌模块，其特征是由九个代谢物和 57 个代谢基因组成。多组学整合分析揭示了与 LUAD 不良生存相关的 28 个代谢基因特征，其中有六个与代谢物相关的基因作为个体预后标志物。
我们通过使用它来指导 AZD-6482 的重新定位，证明了这种综合代谢基因特征在 LUAD 中的临床应用，AZD-6482 是一种 PI3Kβ 抑制剂，可显著抑制 28 个基因特征中的三个基因。总体而言，我们整合了代谢组学和转录组学分析，表明 LUAD 和 LUSC 具有不同的特征，推断出具有患者生存预后价值的基因特征，并确定了治疗靶点和可能用于 NSCLC 治疗的重新定位药物。
