McMahon Alexandra N, Reis Isildinha M, Takita Cristiane, Wright Jean L, Hu Jennifer J
Department of Public Health Sciences, University of Miami Miller School of Medicine, Miami, FL 33136, USA.
Sylvester Comprehensive Cancer Center, University of Miami Miller School of Medicine, Miami, FL 33136, USA.
Cancers (Basel). 2025 Mar 5;17(5):891. doi: 10.3390/cancers17050891.
This study aims to explore metabolic biomarkers and pathways in breast cancer prognosis.
We performed a global post-radiotherapy (RT) urinary metabolomic analysis of 120 breast cancer patients: 60 progression-free (PF) patients as the reference and 60 with progressive disease (PD: recurrence, second primary, metastasis, or death). UPLC-MS/MS (Metabolon Inc.) identified 1742 biochemicals (1258 known and 484 unknown structures). Following normalization to osmolality, log transformation, and imputation of missing values, a Welch's two-sample -test was used to identify biochemicals and metabolic pathways that differed between PF and PD groups. Data analysis and visualization were performed with MetaboAnalyst.
Metabolic biomarkers and pathways that significantly differed between the PD and PF groups were the following: amino acid metabolism, including phenylalanine, tyrosine, and tryptophan biosynthesis (impact value (IV) = 1.00; = 0.0007); histidine metabolism (IV = 0.60; < 0.0001); and arginine and proline metabolism (IV = 0.70; = 0.0035). Metabolites of carbohydrate metabolism, including glucose ( = 0.0197), sedoheptulose ( = 0.0115), and carboxymethyl lysine ( = 0.0098), were elevated in patients with PD. Gamma-glutamyl amino acids, myo-inositol, and oxidative stress biomarkers, including 7-Hydroxyindole Sulfate and sulfate, were elevated in patients who died ( ≤ 0.05).
Amino acid metabolism emerged as a key pathway in breast cancer progression, while carbohydrate and oxidative stress metabolites also showed potential utility as biomarkers for breast cancer progression. These findings demonstrate applications of metabolomics in identifying metabolic biomarkers and pathways as potential targets for predicting breast cancer progression.
本研究旨在探索乳腺癌预后中的代谢生物标志物和代谢途径。
我们对120例乳腺癌患者进行了放疗后尿液代谢组学分析:60例无进展(PF)患者作为参照,60例疾病进展(PD:复发、第二原发性肿瘤、转移或死亡)患者。采用超高效液相色谱-串联质谱法(Metabolon公司)鉴定出1742种生化物质(1258种已知结构和484种未知结构)。在对渗透压进行归一化、对数转换和缺失值插补后,使用韦尔奇两样本t检验来鉴定PF组和PD组之间存在差异的生化物质和代谢途径。数据分析和可视化使用MetaboAnalyst软件完成。
PD组和PF组之间存在显著差异的代谢生物标志物和代谢途径如下:氨基酸代谢,包括苯丙氨酸、酪氨酸和色氨酸生物合成(影响值(IV)=1.00;P=0.0007);组氨酸代谢(IV=0.60;P<0.0001);精氨酸和脯氨酸代谢(IV=0.70;P=0.0035)。糖代谢的代谢产物,包括葡萄糖(P=0.0197)、景天庚酮糖(P=0.0115)和羧甲基赖氨酸(P=0.0098),在PD患者中升高。γ-谷氨酰氨基酸、肌醇以及氧化应激生物标志物,包括7-羟基吲哚硫酸盐和硫酸盐,在死亡患者中升高(P≤0.05)。
氨基酸代谢是乳腺癌进展中的关键途径,而碳水化合物和氧化应激代谢产物也显示出作为乳腺癌进展生物标志物的潜在效用。这些发现证明了代谢组学在识别代谢生物标志物和代谢途径作为预测乳腺癌进展潜在靶点方面的应用。