Plasma markers for gastric cancer diagnosis: a metabolomics- and machine learning-based exploratory study
-
摘要:
目的 基于代谢组学和机器学习算法探究血浆代谢物对胃癌的诊断价值。 方法 分别收集20例胃癌患者(胃癌组)和20名健康自愿者(健康对照组)的血浆样本,用甲醇提取血浆样本中代谢物,对提取的代谢物进行液相色谱串联质谱分析;通过与 mzCloud、mzVault 和 Masslist 数据库比对,对分析后的代谢物进行注释;以变量权重值(variable importance of projection,VIP) > 1、P < 0.05,log2|Fold Change| > 1的标准筛选2组血浆差异代谢物;采用超几何检验将筛选出的差异代谢物富集至京都基因与基因组百科全书(kyoto encyclopedia of genes and genomes,KEGG)数据库进行富集分析,确定2组间的差异代谢通路;通过Boruta算法建立基于特征代谢物的诊断模型和绘制受试者工作特征曲线(receiver operating characteristic,ROC),确定胃癌血浆特征诊断性代谢物,并计算其在2组中的相对含量。 结果 从2组血浆提取物中共筛选出230个差异代谢物;富集到5条代谢差异通路,分别为苯丙氨酸代谢,苯丙氨酸、酪氨酸和色氨酸的生物合成,精氨酸的生物合成,组氨酸代谢,泛酸和辅酶A的生物合成;共确定9个重要的特征诊断性代谢物,分别为天门冬氨酸、苯乙胺、鸟氨酸、马尿酸、瓜氨酸、泛酰巯基乙胺、1 – 甲基组胺、酪氨酸及组氨酸。胃癌组血浆中9种特征诊断性代谢物相对含量均明显低于健康对照组(P < 0.05)。 结论 富集到的5条代谢差异通路可能参与了胃癌的发展过程,确定的9种特征诊断性代谢物可作为胃癌诊断的生物标志物。代谢组学与机器学习算法相结合有助于确定胃癌诊断标志物。 Abstract:Objective To investigate the significance of plasma metabolites for gastric cancer diagnosis based on metabolomics and machine learning algorithms. Methods Plasma samples were collected from 20 gastric cancer patients and 20 gender- and age-matched healthy volunteers (controls). After extracted with methanol, the metabolites in the plasma samples were analyzed with chromatography-mass spectrometry and annotated with mzCloud, mzVault and Masslist databases. The differential metabolites between the cases and controls were screened with the value of variable importance of projection (VIP: > 1, P < 0.05, log2|fold change| > 1); the identified differential metabolites were subjected to the Kyoto Encyclopedia of Genes and Genomes (KEGG) database enrichment using hypergeometric tests to detect differential metabolic pathways between the two groups. Specific metabolite-based diagnostic model and receiver operating characteristic (ROC) curve were established using Boruta algorithm to identify significant differential metabolites for gastric cancer diagnosis. Relative contents of the differential metabolites were also calculated. Results A total of 230 differential metabolites were screened out from the plasma extracts of the two groups and 5 differential metabolic pathways were identified, including phenylalanine metabolism, biosynthesis of phenylalanine/tyrosine/tryptophan, arginine biosynthesis, histidine metabolism, and pantothenic acid/coenzyme A biosynthesis. The identified 9 significant differential metabolites for gastric cancer diagnosis were L-aspartic acid, 2-phenylethylamine, ornithine, hippuric acid, citrulline, pantetheine, 1-methylhistamine, L-tyrosine, and L-histidine and the relative levels of all the 9 differential metabolites were significantly lower in the plasma from the cases than those from the controls (P < 0.05 for all). Conclusion The enriched five metabolic differential pathways may be involved in the development of gastric cancer, and the nine differential metabolites can be used as metabolic biomarkers for gastric cancer diagnosis. The combination of metabolomics and machine learning algorithms could help identify markers for gastric cancer diagnosis. -
Key words:
- metabolomics /
- machine learning /
- gastric cancer /
- diagnostic marker
-
表 1 9种特征诊断性代谢物的基本信息及ROC曲线分析结果
KEGG 编号 中文名称 英文名称 模式 精确分子质量 AUC 95 % CI C00049 天门冬氨酸 L- aspartic Acid 阴离子 133.0375 1.000 — C05332 苯乙胺 2-Phenylethylamine 阳离子 121.0891 0.880 0.775~1.000 C00077 鸟氨酸 Ornithine 阳离子 132.0898 0.835 0.704~0.966 C01586 马尿酸 Hippuric acid 阴离子 179.0582 0.825 0.695~0.955 C00327 瓜氨酸 Citrulline 阴离子 175.0957 0.785 0.640~0.930 C00831 泛酰巯基乙胺 Pantetheine 阳离子 278.1300 0.770 0.618~0.922 C05127 1 – 甲基组胺 1-Methylhistamine 阳离子 125.0953 0.767 0.617~0.918 C00082 酪氨酸 L-Tyrosine 阴离子 181.0739 0.762 0.611~0.914 C00135 组氨酸 L-Histidine 阴离子 155.0695 0.742 0.586~0.899 表 2 9种特征诊断性代谢物在胃癌组和健康对照组中归一化后的相对含量[M(P25,P75)]
特征诊断性代谢物 胃癌组(n = 20) 健康对照组(n = 20) W P 值 天门冬氨酸 – 0.755(– 0.851,– 0.631) 0.432(0.308,0.523) 400 < 0.001 苯乙胺 – 0.917(– 1.022,– 0.819) – 0.259(– 0.366,– 0.165) 352 < 0.001 鸟氨酸 0.560(0.478,0.635) 0.729(0.655,0.837) 334 < 0.001 马尿酸 0.121(– 0.207,0.272) 0.539(0.330,0.798) 330 < 0.001 瓜氨酸 0.065(– 0.002,0.200) 0.258(0.128,0.377) 314 < 0.01 泛酰巯基乙胺 – 1.046(– 1.242,– 0.832) – 0.501(– 0.823,– 0.284) 308 < 0.01 1 – 甲基组胺 – 1.320(– 1.480,– 1.134) – 0.904(– 1.186.– 0.660) 307 < 0.01 酪氨酸 0.320(0.264,0.428) 0.501(0.360,0.592) 305 < 0.01 组氨酸 0.121(0.038,0.169) 0.192(0.139,0.246) 297 < 0.01 -
[1] Smyth EC, Nilsson M, Grabsch HI, et al. Gastric cancer[J]. The Lancet, 2020, 396(10251): 635 – 648. doi: 10.1016/S0140-6736(20)31288-5 [2] Nannini G, Meoni G, Amedei A, et al. Metabolomics profile in gastrointestinal cancers: update and future perspectives[J]. World Journal of Gastroenterology, 2020, 26(20): 2514 – 2532. doi: 10.3748/wjg.v26.i20.2514 [3] Dettmer K, Aronov PA, Hammock BD. Mass spectrometry-based metabolomics[J]. Mass Spectrometry Reviews, 2007, 26(1): 51 – 78. doi: 10.1002/mas.20108 [4] Corona G, Cannizzaro R, Miolo G, et al. Use of metabolomics as a complementary omic approach to implement risk criteria for first-degree relatives of gastric cancer patients[J]. International Journal of Molecular Sciences, 2018, 19(3): 750. doi: 10.3390/ijms19030750 [5] Ritchie SA, Ahiahonu PW, Jayasinghe D, et al. Reduced levels of hydroxylated, polyunsaturated ultra long-chain fatty acids in the serum of colorectal cancer patients: implications for early screening and detection[J]. BMC Medicine, 2010, 8: 13. doi: 10.1186/1741-7015-8-13 [6] Jordan MI, Mitchell TM. Machine learning: trends, perspectives, and prospects[J]. Science, 2015, 349(6245): 255 – 260. doi: 10.1126/science.aaa8415 [7] Amin MB, Edge SB, Greene FL, et al. AJCC cancer staging manual[M]. 8th ed. Cham: Springer, 2017: 203 – 220. [8] 国家卫生健康委员会. 胃癌诊疗规范(2018年版)[J]. 中华消化病与影像杂志(电子版), 2019, 9(3): 118 – 144. [9] Chen ZJ, Huang XC, Gao Y, et al. Plasma-metabolite-based machine learning is a promising diagnostic approach for esoph-ageal squamous cell carcinoma investigation[J]. Journal of Phar-maceutical Analysis, 2021, 11(4): 505 – 514. doi: 10.1016/j.jpha.2020.11.009 [10] Lê S, Josse J, Husson F. FactoMine R: an R package for multivariate analysis[J]. Journal of Statistical Software, 2008, 25(1): 1 – 18. [11] Thévenot EA, Roux A, Xu Y, et al. Analysis of the human adult urinary metabolome variations with age, body mass index, and gender by implementing a comprehensive workflow for univariate and OPLS statistical analyses[J]. Journal of Proteome Research, 2015, 14(8): 3322 – 3335. doi: 10.1021/acs.jproteome.5b00354 [12] Sung H, Ferlay J, Siegel RL, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J]. CA: A Cancer Journal for Clinicians, 2021, 71(3): 209 – 249. doi: 10.3322/caac.21660 [13] Hasim A, Ma H, Mamtimin B, et al. Revealing the metabonomic variation of EC using 1H-NMR spectroscopy and its association with the clinicopathological characteristics[J]. Molecular Biology Reports, 2012, 39(9): 8955 – 8964. doi: 10.1007/s11033-012-1764-z [14] Miyagi Y, Higashiyama M, Gochi A, et al. Plasma free amino acid profiling of five types of cancer patients and its application for early detection[J]. PLoS One, 2011, 6(9): e24143. doi: 10.1371/journal.pone.0024143 [15] Wiggins T, Kumar S, Markar SR, et al. Tyrosine, phenylalanine, and tryptophan in gastroesophageal malignancy: a systematic review[J]. Cancer Epidemiology, Biomarkers and Prevention, 2015, 24(1): 32 – 38. doi: 10.1158/1055-9965.EPI-14-0980 [16] González-Domínguez R, Jáuregui O, Queipo-Ortuño MI, et al. Characterization of the human exposome by a comprehensive and quantitative large-scale multianalyte metabolomics platform[J]. Analytical Chemistry, 2020, 92(20): 13767 – 13775. doi: 10.1021/acs.analchem.0c02008 [17] Holt MC, Assar Z, Zavareh RB, et al. Biochemical characterization and structure-based mutational analysis provide insight into the binding and mechanism of action of novel aspartate aminotrans-ferase inhibitors[J]. Biochemistry, 2018, 57(47): 6604 – 6614. doi: 10.1021/acs.biochem.8b00914 [18] Frolkis A, Knox C, Lim E, et al. SMPDB: the small molecule pathway database[J]. Nucleic Acids Research, 2010, 38(S1): D480 – D487. [19] Jewison T, Su YL, Disfany FM, et al. SMPDB 2.0: big improve- ments to the small molecule pathway database[J]. Nucleic Acids Research, 2014, 42(D1): D478 – D484. doi: 10.1093/nar/gkt1067 [20] Muthukumaran S, Jaidev J, Umashankar V, et al. Ornithine and its role in metabolic diseases: an appraisal[J]. Biomedicine and Pharmacotherapy, 2017, 86: 185 – 194. doi: 10.1016/j.biopha.2016.12.024 [21] Lee H, Juncosa JI, Silverman RB. Ornithine aminotransferase versus GABA aminotransferase: implications for the design of new anticancer drugs[J]. Medicinal Research Reviews, 2015, 35(2): 286 – 305. doi: 10.1002/med.21328 [22] Saxena S, Achyuth BS, Murthy TPK, et al. Structural and functional analysis of disease-associated mutations in GOT1 gene: an in silico study[J]. Computers in Biology and Medicine, 2021, 136: 104695. doi: 10.1016/j.compbiomed.2021.104695 [23] Bednarz-Misa I, Fleszar MG, Fortuna P, et al. Altered l-arginine metabolic pathways in gastric cancer: potential therapeutic targets and biomarkers[J]. Biomolecules, 2021, 11(8): 1086. doi: 10.3390/biom11081086 [24] Ouaknine Krief J, Helly de Tauriers P, Dumenil C, et al. Role of antibiotic use, plasma citrulline and blood microbiome in advanced non-small cell lung cancer patients treated with nivolumab[J]. Journal for Immunotherapy of Cancer, 2019, 7(1): 176. doi: 10.1186/s40425-019-0658-1 [25] Lee KM, Lee H, Han D, et al. Combined the SMAC mimetic and BCL2 inhibitor sensitizes neoadjuvant chemotherapy by targeting necrosome complexes in tyrosine aminoacyl-tRNA synthase-positive breast cancer[J]. Breast Cancer Research, 2020, 22(1): 130. doi: 10.1186/s13058-020-01367-7 [26] Zhang C, Lin XT, Zhao Q, et al. YARS as an oncogenic protein that promotes gastric cancer progression through activating PI3K-Akt signaling[J]. Journal of Cancer Research and Clinical Oncology, 2020, 146(2): 329 – 342. doi: 10.1007/s00432-019-03115-7 [27] Petrova B, Kanarek N. Potential benefits and pitfalls of histidine supplementation for cancer therapy enhancement[J]. The Journal of Nutrition, 2020, 150(S1): 2580S – 2587S. [28] Lee W, Um J, Hwang B, et al. Assessing the progression of gastric cancer via profiling of histamine, histidine, and bile acids in gastric juice using LC-MS/MS[J]. The Journal of Steroid Biochemistry and Molecular Biology, 2020, 197: 105539. doi: 10.1016/j.jsbmb.2019.105539 [29] Novita Sari I, Setiawan T, Seock Kim K, et al. Metabolism and function of polyamines in cancer progression[J]. Cancer Letters, 2021, 519: 91 – 104. doi: 10.1016/j.canlet.2021.06.020 [30] Casero RA Jr, Murray Stewart T, Pegg AE. Polyamine metabolism and cancer: treatments, challenges and opportunities[J]. Nature Reviews Cancer, 2018, 18(11): 681 – 695. doi: 10.1038/s41568-018-0050-3 [31] Kursa MB, Rudnicki WR. Feature selection with the boruta package[J]. Journal of Statistical Software, 2010, 36(11): 1 – 13.