【目的】

自身免疫病（Autoimmune diseases, AIDs）患者罹患恶性肿瘤（Cancer, CA）的风险显著升高，代谢重编程是二者共有的核心特征。然而，在两者共病状态下的代谢特征及其潜在机制仍缺乏系统研究。本研究旨在基于血清靶向代谢组学技术，系统描绘AID合并肿瘤（AID-CA）患者的代谢特征并探索潜在的生物标志物。

【方法】

本研究共纳入111名受试者，包括47名AID患者、49名AID-CA患者和15名健康对照。其中，AID组包括类风湿关节炎（Rheumatoid arthritis, RA）16例、系统性红斑狼疮（Systemic lupus erythematosus, SLE）9例、干燥综合征（Sjögren’s syndrome, SS）7例、特发性炎症性肌病（Idiopathic inflammatory myopathies, IIM）10例及系统性硬化症（Systemic sclerosis, SSc�5例；AID-CA组包括RA-CA 16例、SLE-CA 8例、SS-CA 7例、IIM-CA 12例、SSc-CA 6例。所有受试者血清样本均采用液相色谱-串联质谱平台（LC-MS/MS）进行靶向代谢物检测。利用主成分分析（PCA）和正交偏最小二乘判别分析（OPLS-DA）评估各组间代谢谱差异，通过变量投影重要性（Variable Importance in Projection, VIP）值和p值联合筛选差异代谢物。随后基于KEGG数据库进行代谢通路富集分析，以揭示相关代谢物的潜在生物学机制，并通过ROC曲线评估其判别效能。

【结果】

1.五个AID-CA亚组中,与相应的AID比较，共筛选出显著差异代谢物127种，涵盖氨基酸、核酸及脂类等多个代谢类别。其中，RA-CA组差异代谢物最多（56个），IIM-CA组最少（10个）。交集分析发现26种代谢物在至少两个亚组中同时差异表达，尤其以RA组与其他亚组之间交集最多。其中N-乙酰基-S-(3-羟基丙基)半胱氨酸在RA-CA、SLE-CA和SS-CA三个亚组中一致上调。

2. 通路富集和拓扑分析显示，氨基酸代谢在所有亚组中均为核心改变通路，尤其是丙氨酸-天冬氨酸-谷氨酸代谢和甘氨酸-丝氨酸-苏氨酸代谢通路，在多个亚组中均具有较高的影响力。同时，嘧啶代谢、甘油磷脂代谢、氨酰tRNA合成等通路也展现出较高显著性与网络连接性。

3. ROC曲线分析揭示了多种具有较强判别能力（AUC>0.8）的代谢物，其中氨基酸及有机酸类代谢物在多个亚组中反复出现，提示其具有潜在的共性的生物标志物价值；同时，不同亚组亦呈现出独特的代谢物谱。

【结论】

AID-CA亚组的代谢组学分析各亚组分别筛选出多种显著差异代谢物，主要涉及氨基酸、核酸及脂类。部分代谢物在多个亚组中一致性改变，提示可成为潜在的共病代谢标志物。AID-CA患者普遍存在以氨基酸代谢为核心的代谢重构，伴随嘧啶代谢、脂类代谢、抗氧化及神经调控相关通路的协同变化，呈现高度网络连接性和系统性调控特征。ROC分析进一步验证了多种代谢物的良好鉴别能力，表明AID-CA状态下存在具有共性基础又具亚组特异性的代谢重编程模式，为疾病机制理解及生物标志物筛选提供了依据。

Objective Patients with autoimmune diseases (AIDs) have a significantly increased risk of developing cancer. Metabolic reprogramming is a shared hallmark of both conditions. However, the metabolic characteristics and underlying mechanisms in the comorbid state remain largely unexplored. This study aims to systematically characterize the metabolic features of patients with AIDs complicated with cancer (AID-CA) using targeted serum metabolomics and to identify potential metabolic biomarkers. Methods A total of 111 participants were enrolled in this study, including 47 patients with autoimmune diseases (AID), 49 patients with autoimmune diseases complicated by cancer (AID-CA), and 15 healthy controls. The AID group consisted of 16 patients with rheumatoid arthritis (RA), 9 with systemic lupus erythematosus (SLE), 7 with Sjögren’s syndrome (SS), 10 with idiopathic inflammatory myopathies (IIM), and 5 with systemic sclerosis (SSc). The AID-CA group included 16 RA-CA, 8 SLE-CA, 7 SS-CA, 12 IIM-CA, and 6 SSc-CA patients. Serum samples from all participants were analyzed using a targeted metabolomics platform based on liquid chromatography–tandem mass spectrometry (LC-MS/MS). Principal component analysis (PCA) and orthogonal partial least squares discriminant analysis (OPLS-DA) were applied to assess metabolic profile differences among groups. Differential metabolites were identified based on variable importance in projection (VIP) values and p-values. Subsequently, pathway enrichment analysis was performed using the KEGG database to elucidate potential biological mechanisms, and receiver operating characteristic (ROC) curve analysis was conducted to evaluate the discriminatory performance of candidate metabolites. Results 1. A total of 127 significant differential metabolites were identified between the five AID-CA subgroups compared with AIDs, involving various metabolic classes such as amino acids, nucleotides, and lipids. Among them, the RA-CA subgroup exhibited the highest number of differential metabolites (n = 56), while the IIM-CA subgroup had the fewest (n = 10). Intersection analysis revealed 26 metabolites that were differentially expressed in at least two subgroups, with the RA group showing the greatest overlap with other subgroups. Notably, N-acetyl-S-(3-hydroxypropyl)cysteine was consistently upregulated in the RA-CA, SLE-CA, and SS-CA subgroups. 2. Pathway enrichment and topology analysis indicated that amino acid metabolism was the central altered pathway in all subgroups, particularly the alanine–aspartate–glutamate metabolism and glycine–serine–threonine metabolism pathways, which exhibited high impact scores across multiple subgroups. In addition, pyrimidine metabolism, glycerophospholipid metabolism, and aminoacyl-tRNA biosynthesis also demonstrated significant enrichment and strong network connectivity. 3. Receiver operating characteristic (ROC) curve analysis identified multiple metabolites with strong discriminative power (AUC > 0.8). Among them, amino acids and organic acids frequently appeared across subgroups, suggesting their potential as common metabolic biomarkers. Meanwhile, distinct metabolic profiles were also observed in individual subgroups. Conclusion Metabolomic analysis of AID-CA subgroups identified multiple significantly altered metabolites, mainly involving amino acids, nucleotides, and lipids. Some metabolites showed consistent changes across subgroups, suggesting potential shared biomarkers. AID-CA patients exhibited metabolic reprogramming centered on amino acid metabolism, along with changes in pyrimidine metabolism, lipid metabolism, antioxidant, and neuroregulatory pathways, indicating systemic metabolic alterations. ROC analysis confirmed the good discriminative performance of several metabolites, supporting the presence of both common and subgroup-specific metabolic features in AID-CA. These findings provide a basis for understanding disease mechanisms and exploring metabolic biomarkers.