Abstract
Lipids participate in many important biological functions through energy storage, membrane structure stabilization, signal transduction, and molecular recognition. Previous studies have shown that patients with esophageal squamous cell carcinoma (ESCC) have abnormal lipid metabolism. However, studies characterizing lipid metabolism in ESCC patients through lipidomics are limited. Plasma lipid profiles of 65 ESCC patients and 42 healthy controls (HC) were characterized by lipidomics-based ultraperformance liquid chromatography quadrupole time-of-flight mass spectrometry (UPLC-Q-TOF-MS). Single-factor and multi-factor statistical analysis were used to screen the differences in blood lipids between groups, and combined with component ratio analysis and receiver operating characteristic (ROC) curve diagnostic efficiency assessment, to reveal the potential mechanisms and biomarkers of ESCC. There were significant differences in lipid profiles between the ESCC and HC groups. Thirty-six differential lipids (11 up-regulated and 25 down-regulated) were selected based on the criteria of p < .05 and fold change > 1.3 or < 0.77. Glycerophospholipids were the major differential lipids, suggesting that these lipid metabolic pathways exhibit a significant imbalance that may contribute to the development of esophageal squamous cell carcinoma. Among them, the seven candidate biomarkers for esophageal squamous cell carcinoma with the highest diagnostic value are three phosphatidylserine (PS), three fatty acids (FA) and one phosphatidylcholine (PC).
Introduction
Esophageal carcinoma (EC) is a common gastrointestinal malignancy, globally ranked seventh and sixth in incidence and mortality, respectively (Citation1). Esophageal squamous cell carcinoma (ESCC) is the main subtype of EC, accounting for approximately 90% of EC cases in developing countries, while adenocarcinoma is instead mainly found in developed countries (Citation2, Citation3). Despite significant improvements in multidisciplinary treatments such as surgery, radiotherapy, and chemotherapy, prognosis remains poor due to the relatively advanced stage of ESCC cases diagnosed, with a probable five-year survival rate of only about 20% (Citation4). At present, the early screening and detection of ESCC are primarily achieved through gastrointestinal endoscopy and biopsy (Citation5, Citation6). However, these methods are invasive, expensive, and highly subjective (Citation7). Moreover, the sensitivity and specificity of traditional serum tumor markers are low, and have limited value in the early diagnosis of ESCC (Citation8). Thus, it is imperative to develop novel early diagnostic biomarkers of ESCC that are noninvasive and cost-effective, with high sensitivity and specificity.
Lipids are widely distributed in the body, represent the majority of the metabolites in the blood, and account for 50% of the weight of the cell membrane (Citation9). Lipids may be classified into eight categories (Citation10), namely fatty acyls (FA), glycerolipids (GL), glycerophospholipids (GP), sphingolipids (SP), sterol lipids (ST), prenol lipids (PR), saccharolipids (SL), and polyketides (PK). Studies have found that lipids have many important biological functions such as energy conversion, substance transport, signal transduction, cell development and differentiation, and regulation of cell death. The comprehensive study of the composition and role of lipids in cells, biofluids, tissues, and organs is known as lipidomics (Citation11). Lipids are both the main components of cell membranes and significant energy sources. These roles make lipids essential biomolecules that provide nutritional and structural support for cancer cell proliferation (Citation12). Dysregulation of lipid metabolism might be an early indicator of cancer start, and a growing number of studies have indicated that an imbalance in lipid metabolism is related with numerous types of malignancies (Citation13, Citation14). Advanced technologies, including chromatography and mass spectrometry, are used in lipidomics to obtain rapid results. It is used for the determination of lipid molecules in biological samples on a qualitative and quantitative basis and the evaluation of lipid alterations throughout the disease process, in order to identify abnormally expressed lipid molecules for diagnostic purposes (Citation15, Citation16).
Recent studies have confirmed the presence of abnormal lipid levels in ESCC; for example, glycerophospholipids were decreased in ESCC patients compared with healthy controls, especially phosphatidylcholines (PCs) and phosphoethanolamines (PEs) (Citation17, Citation18). However, there are limited studies on the overall metabolic levels of lipids in ESCC serum samples. Given the diverse and complex factors leading to disease progression, combined analysis of multiple biomarkers formed by synergistic changes in several lipid species may improve predictive accuracy. Thus, a comprehensive understanding of lipid dysregulation is required to screen for biomarkers with diagnostic performance.
In this study, we characterized plasma lipid metabolic profiles of healthy controls and ESCC using ultra-performance liquid chromatography quadrupole time-of-flight mass spectrometry (UPLC-Q-TOF-MS). The primary aim of this study was to identify signature lipid biomarkers of ESCC.
Materials and Methods
Clinical Sample Collection
A total of 65 patients diagnosed with ESCC and 42 healthy controls (HC) were recruited for serum lipidomic analysis at the First Affiliated Hospital of Wannan Medical College (Wuhu, Anhui province, China) between January 2021 and December 2022. The experimental group consisted of patients diagnosed with ESCC after endoscopic and histopathological examination. Staging of the tumor was based on the 8th Staging System of the American Joint Committee on Cancer (AJCC) (Citation19). Inclusion criteria for patients in the ESCC group: meeting the pathological diagnostic criteria for esophageal squamous cell carcinoma; no special treatment such as surgery, radiotherapy, chemotherapy, etc.; no history of metabolic diseases, liver and kidney diseases or other combined malignant tumors. Exclusion criteria for study subjects in the HC group: suffering from metabolic diseases such as thyroid diseases, hypertension, diabetes; combining with hepatitis, nephritis and other diseases that affect metabolism; taking drugs that may affect the body’s metabolism over a long period of time. Detailed characteristics of the study cohort were shown in . The Institutional Ethics Committee of the First Affiliated Hospital of Wannan Medical College approved this study, and all participants provided written informed consent.
Table 1. Information of clinical characteristics for study cohort.
Venous blood was collected after fasting, and the samples were stored in EDTA anticoagulation tubes. Within 1 h, the samples were centrifuged at 5,000 rpm for 10 min at 4 °C. The serum samples were immediately stored at −80 °C prior to analysis.
Sample Preparation for Lipidomic Analysis
All serum samples stored at −80 °C were thawed on ice and stored at 4 °C during the whole preparation process. Each sample was prepared by adding 40 µL to 300 µL of pre-cooled methanol (-20 °C), followed by the addition of 1 mL of methyl tert-butyl ether (MTBE) and 300 µL deionized water. The mixture was vortexed for 10 s after each addition, and oscillated at 1,200 rpm for 15 min after MTBE addition. The sample preparations were then equilibrated at 4 °C for 1 min before 8,000 rpm centrifugation at 4 °C for 10 min. At last, the upper organic phase was blown with nitrogen and frozen to dry, and 150 µL of acetonitrile/isopropanol/H2O (ACN/IPA/H2O) (65:30:5, v/v/v) mixture was added to re-dissolve the samples, vortexed thoroughly, passed through a 0.2 µm filter membrane and stored for subsequent analysis. The quality control (QC) samples were prepared by mixing equal volumes of serum from each sample included in the pending analysis. To monitor the stability and reproducibility of the LC-MS system, 11 QC samples were used, with the same pretreatment method used.
UPLC-Q-TOF-MS Based Lipidomic Analysis
An ultra-performance liquid chromatography (UPLC) system coupled with a quadrupole time-of-flight tandem mass spectrometry (Q-TOF-MS) platform (Agilent, USA) was used for lipidomics analysis. The lipid extract separation was performed by An Agilent ZORBAX Eclipse Plus C18 column (100 × 2.1 mm, 1.8 µm, Agilent, USA). Mobile phases A and B, used in both positive and negative ion modes, were ACN/H2O (60:40, v/v) and ACN/IPA (10:90, v/v), respectively, both containing 10 mM ammonium formate. The column temperature and flow rate were set to 65 °C and 0.60 mL/min, respectively. Each sample was injected with 1 µL. Lipid profiling analysis data were obtained in the full-scan mode in both positive (+) and negative (−) ion modes. The detailed settings of the source parameters were as follows: m/z scan range, 100–1500; drying gas flow rate, 10 L/min; drying gas temperature, 350 °C; atomizing gas pressure, 30 psi; capillary voltage, 3,500 V; skimmer voltage, 65 V; octupole RF voltage, 750 V; capillary outlet voltage, 150 V. The internal standard ions of the reference solution in the positive and negative modes were m/z 121.0509, 922.0098, and m/z 112.9856, 1,033.9881, respectively.
Data Processing and Statistical Analyses
The raw data were processed by peak alignment, deconvolution, and normalization by sum, and the LIPID MAPS database was used to identify lipids according to their extract mass, retention time, and MS/MS fragment information. The SIMCA-P software (version 14.1, Umetrics, Umea, Sweden) was used for multivariate analysis. The overall stability of the data and system were visualized using unsupervised principal component analysis (PCA), while supervised orthogonal partial least squares discriminant analysis (OPLS-DA) was used to obtain a better understanding of the variables responsible for classification, and variable importance in projection (VIP) was obtained to identify significant lipids. The differences in lipid composition between the two groups were characterized using Student’s t-test with the Benjamini-Hochberg-based false discovery rate (FDR). Subsequently, lipids with VIP > 1, adjusted p (FDR) < .05, and |log2(fold change)| > 0.38 were selected as significant differential lipids. A heat map was drawn to assess the relative abundance of differential lipid molecules, and correlation analysis (Pearson) of the selected lipids was performed using R software (version 3.6.3). The MetaboAnalyst 5.0 (https://www.metaboanalyst.ca/), Cytoscape software (version 3.9.1), and the Kyoto Encyclopedia of Genes and Genomes (KEGG) database were used for metabolic pathway analysis of the selected lipids. Binary logistic regression analysis and receiver operating characteristic (ROC) curves were constructed to evaluate the diagnostic performance of combinational markers.
Results
Differential Lipid Profiles between ESCC and HC
39 lipid species in the positive ion mode and 102 lipid species in the negative ion mode were obtained from 107 serum samples. The detected lipid species were classified into four categories: 115 glycerophospholipids (GP, 81.56%), 15 sphingolipids (SP, 10.64%), 9 fatty acids (FA, 6.38%), and 2 glycerolipids (GL, 1.42%) ().
PCA and OPLS-DA models, the common multivariate statistical methods used in omics study, were utilized to evaluate the diferences between groups regarding lipid metabolism of ESCC and HC groups. As shown in , the QC samples were gathered closely in the center of the PCA score plot, indicating the analysis system with excellent robustness and reproducibility during the batch analysis process. The OPLS-DA model clearly distinguished between ESCC and HC groups on the basis of the lipid dataset, and with a good fitting and predictive performance (R2X = 0.747, R2Y = 0.992, Q2Y = 0.977), which indicated remarkable differences between groups and obvious dysregulation in lipid metabolism of ESCC relative to HC group (). Moreover, a 200 times permutation test was performed to verify the reliability and applicability of the OPLS-DA model for data analysis, the permutation test demonstrated that the model was not over-fitted to the intercepts of the R2Y and Q2Y values (0.0, 0.399 and 0.0, −0.673, respectively), which provided proof that OPLS-DA model was rational and not overfitting for the data analysis ().
Figure 1. Multivariate statistical analysis of differential lipid features between ESCC and HC groups using PCA and OPLS-DA models. (A) Pie chart of lipid categories and counts. GP: glycerophospholipid; SP: sphingolipid; FA: fatty acid; GL: glycerolipid. (B) The PCA analysis of all subjects and quality controls. (C) The score plot of OPLS-DA model for ESCC and HC groups. (D) The Overftting test (200 times) for the OPLS-DA model.
Screening and Identification of Diagnostic Lipid Biomarkers for ESCC
The difference analysis in serum lipid profiles between the ESCC and HC groups were performed using univariate and multivariate statistical methods. The potential differential lipid metabolites between the ESCC and HC groups were identified by selecting lipids with significant parameters (VIP > 1, FDR < 0.05, and |log2(fold change)| > 0.38), a total of 36 differential lipids were annotated, 11 of which were up-regulated and 25 of which were down-regulated (). Detailed characteristics of the 36 lipid species are summarized in . A clustering heatmap was utilized to display the expression of 36 annotated differential lipids between ESCC and HC, in TNM stages I, II, III, and IV (). A clear distinction was observed between the ESCC and HC samples, with HC samples having higher enrichment for most lipids, including phosphatidylserine (PSs) and triglycerides (TGs). Taken together, PSs and TGs are considered to be the main influencing factor that contributed to the ESCC formation.
Figure 2. Screening and identification of lipid species by lipidomics analysis. (A) Venn diagram of VIP (variable importance in projection), FDR (false discovery rate) and FC (fold change). (B) Volcano plot of lipid species with adjusted p value and fold change. (C) Heat map of the selected lipid species between ESCC and HC groups. Level distribution of differential lipids between ESCC and HC groups. Clustering heat map was draw using R software by data transforming with log10. The majority of differential lipids in the ESCC group showed a tendency of significant decrease compared to HC group.
Table 2. Identification and selection of 36 differential lipids.
Performance Evaluation of Potential Lipid Biomarkers of ESCC
The diagnostic performance of 36 differential lipids between ESCC and HC was evaluated by ROC analysis using MetaboAnalyst 5.0, which could maximize the area under the curve (AUC) as calculated by the trapezoidal method to select the most suitable cutoff point. Generally, the AUC values of the differential lipids ranged from 0.678 to 0.883, and most of them had comparatively low AUC values (). By combining with AUC ≥ 0.800 as selected criteria, we obtained 7 differential lipids with good diagnostic performance for ESCC (), mainly including 3 PSs, 3 FAs lipids, and 1 PC. To construct a more effective diagnostic model, logistic regression analysis was used to calculate the predicted probability distributions of the lipid species with an AUC > 0.8 among the ESCC and HC groups. The AUC value of the combined panel was 0.944 (). All seven lipids exhibited significant downregulation in ESCC patients, and showed significant differences between TNM stages I and II, and between stages III and IV ().
Figure 3. (A-G) ROC curves for the top 7 differential lipids of highest diagnostic value. PC: phosphatidylcholine; PS: phosphatidylserine. (H) ROC curve for the combination diagnostic model of seven lipids. AUC: Area Under Curve. CI: Confidence Interval.
Figure 4. Box plots of the relative abundance of the 7 lipids with highest AUC value between ESCC (TNM I&II, III&IV) and HC groups. The levels of differential lipids were displayed with mean ± SEM. *: p < .05; **: p < .01; ***: p < .001; NS: no significance. PC: phosphatidylcholine; PS: phosphatidylserine.
Correlation and Pathway Analysis of the Differential Lipids
To measure the proximity between differential lipids and further understand the interrelationship between lipid molecules during changes of biological state, a correlation analysis between the relative abundance of the 36 identified lipids in the plasma of patients with ESCC is illustrated in . Positive and negative correlations differed among the various lipids, phosphatidylcholines (PC), sphingomyelin (SM), and phosphatidylethanolamine (PE) were negatively correlated with PS and TG, while the remaining lipids were slightly positively correlated.
Figure 5. Correlation and metabolic pathway analysis. (A) Heat map of the correlation strength of the lipids. **: p < .01; ***: p < .001. (B) The pathway impact plot of 36 differential lipids. (C) Network of differential lipids metabolic pathways, the red-labeled represented the differential lipids, while the metabolites associated with the differential lipids were marked in blue.
By conducting an analysis of the metabolic pathways of different lipids, the results showed that these lipids are mainly involved in seven metabolic pathways, of which glycerophospholipid metabolism, glycerolipid metabolism and sphingolipid metabolism in cancer were considered the most significant (p < .01) (). The Cytoscape software (Metscape app) was used to construct a network of differential lipid metabolism pathways ().
Discussion
Lipidomics is an emerging technology that integrates the analyses of the end products of lipid metabolism to reveal internal changes within the whole organism, and it has been widely applied in many cancer studies. Previous researches have demonstrated that lipid metabolism plays a crucial role in tumor development and metastasis, and several types of lipids and their derivatives have been reported to be potential biomarkers for a variety of cancers (Citation20–22). To date, there have been few reports of lipid profiles in ESCC that were used to identify novel biomarkers. Our non-targeted UPLC-Q-TOF/MS results showed that 36 individual lipid molecules were significantly differentially expressed between ESCC patients and healthy controls. Seven lipids with AUC values ranging from 0.801 to 0.883 demonstrated excellent diagnostic ability, and the significant differences in expression levels across tumor growth stages suggests that these lipids may have an impact on tumor disease progression. In addition, we explored a panel of ESCC prognosis-associated lipids.
Glycerophospholipids are the most prevalent and abundant phospholipids in the body, and include phosphatidic acid (PA), phosphatidylserine (PS), phosphatidylethanolamine (PE), phosphatidylcholine (PC), phosphatidylinositol (PI), phosphatidylglycerol (PG), and cardiolipin (CL). They are important biomolecules that contribute to the formation of cell membrane structure and participate in many bioregulatory processes. PC species are the most abundant glycerophospholipid in eukaryotic cell membranes and are associated with tumor cell proliferation and signal transduction (Citation23). Previous studies have shown that PC metabolism is disturbed in a variety of tumors, and that some PCs have great potential as tumor biomarkers (Citation24). A metabolomics study revealed that the serum PC levels were downregulated in ESCC patients compared to healthy subjects (Citation17), and another study showed both increased and decreased PC levels in ESCC (Citation25). In our study, three PC homologs were identified, and the relative abundance of PC(20:5/15:0) and PC(22:6/13:0) was higher in patients with ESCC than in healthy controls, whereas PC(14:1/20:4) level was decreased in ESCC patients. Generally, these results are consistent with those of previous studies. Serum PC levels are regulated by phospholipase A2 (PLA2), which breaks down PC into fatty acids and lysophospholipase. It has been found that PLA2 is highly active and overexpressed in ovarian and breast cancers, which may contribute to the downregulation of serum PC levels in patients (Citation26, Citation27). In addition, Saito et al. discovered that patients with renal clear cell carcinoma had lower levels of the CHPTCitation1 gene, which is associated with PC biosynthesis, compared to healthy controls (Citation28). This may contribute to the downregulation of PC levels in cancer patients. Overall, the results of this study suggest that PC metabolism is dysregulated in the serum of ESCC patients, and we speculate that this may be related to PLA2 activity or abnormal expression of genes related to PC synthesis in the patients, however, its specific pathogenesis still needs to be verified by further studies.
Phosphatidylethanolamine is an important lipid component present in the inner mitochondrial membrane and involved in regulatory processes such as cell survival, signal transduction, and apoptosis. PE metabolism has been confirmed to be abnormal in several cancers (Citation29). In this study, the levels of PE(P-16:0/20:4), PE(18:0/16:0), PE(22:6/19:0), and PE(22:1/17:0) were significantly increased in ESCC, indicating abnormal plasma phospholipid levels and energy metabolism dysfunction in ESCC patients. Concordant with our findings, Zhu et al. also found that the levels of serum PEs tended to be higher in patients with ESCC than in healthy subjects (Citation30). In addition, in a previous study on non-small cell lung cancer (NSCLC), Chen et al. found that PE expression levels were significantly increased in patients with early stage NSCLC (Citation31). Aberrant PE metabolism has also been reported in other cancers, such as hepatocellular carcinoma and cervical and colorectal cancers (Citation32–34).
Phosphatidylserine comprises a minor percentage of the total phospholipid volume, but it has an important impact on the mediation of apoptosis and the activation enzyme synthesis and metabolism (Citation34). A recent report on serum lipids in patients with chronic obstructive pulmonary disease (COPD) showed that they had increased serum PS levels compared with healthy controls (Citation35). Interestingly, all 16 PSs were significantly downregulated in patients with ESCC in this study, and several PSs were found to be highly associated with the development of ESCC, revealing potential clinical significance to diagnosis. However, to the best of our knowledge, there are few reports on changes in plasma PS metabolism in ESCC.
FAs are intermediates necessary for the maintenance of the cell membrane structure and function, energy storage, and cell signaling. The acylation of FAs plays an important role in the lipid metabolism pathway, and the elucidation of FA structure and content has attracted widespread attention as an integral part of the comprehensive understanding of lipid metabolism in organisms. Previous studies have suggested that key enzymes involved in fatty acid synthesis and catabolism may be closely linked to tumor development (Citation36). Fatty acid synthase (FAS) promotes breast cancer formation by affecting epidermal growth factor receptor expression, promotes metastasis in hepatocellular carcinoma, induces the epithelial-mesenchymal transition in ovarian cancer. Degradation of COX-2 inhibits the growth of lung, breast, and colorectal cancer cells as well as metastasis of ovarian cancer cells and promotes cell proliferation (Citation37–40). In the present study, the serum levels of six FAs (DHA, 12E, 16E-octadecadienoic acid, (R)-2-hydroxyhexadecanoic acid, 11S-hydroxy-hexadecanoic acid, 11E-hexadecenyl acetate and decyl octanoate) were reduced in patients with ESCC. On the other hand, we speculate that dietary difficulties in ESCC patients were responsible for the decrease in FAs. In addition, the levels of two GLs, TG(12:0/18:2/22:6) and TG(14:1/14:1/22:6), also showed a declining trend. Although there are no detailed reports on the metabolism of serum FAs and GLs in ESCC, we observed unique labeling patterns for these two lipid groups in patients with ESCC, suggesting that they may play essential roles in ESCC development.
ROC curves were generated to assess the diagnostic efficacy of candidate lipid biomarkers, and the predictive power of a single lipid species was unsatisfactory for ESCC diagnosis. Fortunately, previous studies have reported that a multi-metabolite model can provide better diagnostic efficacy. Therefore, we conducted a binary logistic regression analysis to establish a multilipid model. A lipid biomarker panel containing PC(14:1/20:4), DHA, PS(18:1/22:0), 12E, 16E-octadecadienoic acid, (R)-2-Hydroxyhexadecanoic acid, PS(18:3/19:0), and PS(15:1/20:0) had the highest AUC values. These results indicate that the combinational lipid biomarker could be useful in the diagnosis of ESCC. It should be emphasized that further validation with a large number of samples is necessary before it could be considered for clinical applications.
However, this study has several limitations that should be addressed. First, patients pathologically diagnosed with adenocarcinoma or esophagitis were not included in the experimental cohort. Second, owing to the limited sample size, further validation efforts using a large independent dataset are required. Third, the study solely focused on lipids. A multi-omics research platform involving lipidomics, metabolomics, and proteomics could be established for the targeted quantification of diverse serum markers to develop a more comprehensive and reliable diagnostic model.
Conclusions
In conclusion, the patterns of lipid levels in patients with ESCC differ a lot from that of healthy subjects based on the plasma lipidomics analysis in the present study. Our findings suggest that the defined several novel lipid biomarkers will contribute to the diagnosis of ESCC, as well as can provide a more efficient treatment strategy for patients with ESCC.
Authors contributions
QW. W and DQ. L designed and directed the study. Y. Z and LQ. J and participated in sample collection. L. L. and ZC. X and participate in data compilation. TW. S participated in data analysis and writing the manuscript. Final draft read and approved by all authors.
Acknowledgements
We thank all participants and the researchers and collaborators involved in this study.
Disclosure statement
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Data availability statement
The analyzed data sets generated during the study are available from the corresponding author on reasonable request.
Additional information
Funding
References
- Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2018;68(6):394–424. doi: 10.3322/caac.21492.
- Hou H, Meng Z, Zhao X, Ding G, Sun M, Wang W, Wang Y. Survival of esophageal cancer in China: a pooled analysis on hospital-based studies from 2000 to 2018. Front Oncol. 2019;9:548. doi: 10.3389/fonc.2019.00548.
- Napier KJ, Scheerer M, Misra S. Esophageal cancer: a review of epidemiology, pathogenesis, staging workup and treatment modalities. World J Gastrointest Oncol. 2014;6(5):112–20. doi: 10.4251/wjgo.v6.i5.112.
- Mannath J, Ragunath K. Role of endoscopy in early oesophageal cancer. Nat Rev Gastroenterol Hepatol. 2016;13(12):720–30. doi: 10.1038/nrgastro.2016.148.
- Chen XX, Zhong Q, Liu Y, Yan SM, Chen ZH, Jin SZ, Xia TL, Li RY, Zhou AJ, Su Z, et al. Genomic comparison of esophageal squamous cell carcinoma and its precursor lesions by multi-region whole-exome sequencing. Nat Commun. 2017;8(1):524. doi: 10.1038/s41467-017-00650-0.
- Li F, Li X, Guo C, Xu R, Li F, Pan Y, Liu M, Liu Z, Shi C, Wang H, et al. Estimation of cost for endoscopic screening for esophageal cancer in a high-risk population in Rural China: results from a population-level randomized controlled trial. Pharmacoeconomics. 2019;37(6):819–27. doi: 10.1007/s40273-019-00766-9.
- Yang Y, Huang X, Zhou L, Deng T, Ning T, Liu R, Zhang L, Bai M, Zhang H, Li H, et al. Clinical use of tumor biomarkers in prediction for prognosis and chemotherapeutic effect in esophageal squamous cell carcinoma. BMC Cancer. 2019;19(1):526. doi: 10.1186/s12885-019-5755-5.
- Armitage EG, Southam AD. Monitoring cancer prognosis, diagnosis and treatment efficacy using metabolomics and lipidomics. Metabolomics. 2016;12(9):146. doi: 10.1007/s11306-016-1093-7.
- Nguyen A, Rudge SA, Zhang Q, Wakelam MJ. Using lipidomics analysis to determine signalling and metabolic changes in cells. Curr Opin Biotechnol. 2017;43:96–103. doi: 10.1016/j.copbio.2016.10.003.
- Yang K, Han X. Lipidomics: techniques, applications, and outcomes related to biomedical sciences. Trends Biochem Sci. 2016;41(11):954–69. doi: 10.1016/j.tibs.2016.08.010.
- Snaebjornsson MT, Janaki-Raman S, Schulze A. Greasing the wheels of the cancer machine: the role of lipid metabolism in cancer. Cell Metab. 2020;31(1):62–76. doi: 10.1016/j.cmet.2019.11.010.
- Costello E. A metabolomics-based biomarker signature discriminates pancreatic cancer from chronic pancreatitis. Gut. 2018;67(1):2–3. doi: 10.1136/gutjnl-2016-313665.
- Perrotti F, Rosa C, Cicalini I, Sacchetta P, Del Boccio P, Genovesi D, Pieragostino D. Advances in lipidomics for cancer biomarkers discovery. Int J Mol Sci. 2016;17(12):1992. doi: 10.3390/ijms17121992.
- Han X. Lipidomics for studying metabolism. Nat Rev Endocrinol. 2016;12(11):668–79. doi: 10.1038/nrendo.2016.98.
- Hu C, van der Heijden R, Wang M, van der Greef J, Hankemeier T, Xu G. Analytical strategies in lipidomics and applications in disease biomarker discovery. J Chromatogr B Analyt Technol Biomed Life Sci. 2009;877(26):2836–46. doi: 10.1016/j.jchromb.2009.01.038.
- Mir SA, Rajagopalan P, Jain AP, Khan AA, Datta KK, Mohan SV, Lateef SS, Sahasrabuddhe N, Somani BL, Keshava Prasad TS, et al. LC-MS-based serum metabolomic analysis reveals dysregulation of phosphatidylcholines in esophageal squamous cell carcinoma. J Proteomics. 2015;127(Pt A):96–102. doi: 10.1016/j.jprot.2015.05.013.
- Ma W, Wang S, Zhang T, Zhang EY, Zhou L, Hu C, Yu JJ, Xu G. Activation of choline kinase drives aberrant choline metabolism in esophageal squamous cell carcinomas. J Pharm Biomed Anal. 2018;155:148–56. doi: 10.1016/j.jpba.2018.03.062.
- Rice TW, Ishwaran H, Ferguson MK, Blackstone EH, Goldstraw P. Cancer of the esophagus and esophagogastric junction: an eighth edition staging primer. J Thorac Oncol. 2017;12(1):36–42. doi: 10.1016/j.jtho.2016.10.016.
- Long J, Zhang CJ, Zhu N, Du K, Yin YF, Tan X, Liao DF, Qin L. Lipid metabolism and carcinogenesis, cancer development. Am J Cancer Res. 2018;8(5):778–91.
- Yan F, Zhao H, Zeng Y. Lipidomics: a promising cancer biomarker. Clin Transl Med. 2018;7(1):21. doi: 10.1186/s40169-018-0199-0.
- Mistry DA, French PW. Circulating phospholipids as biomarkers of breast cancer: a review. Breast Cancer. 2016;10:191–6. doi: 10.4137/BCBCR.S40693.
- Ridgway ND. The role of phosphatidylcholine and choline metabolites to cell proliferation and survival. Crit Rev Biochem Mol Biol. 2013;48(1):20–38. doi: 10.3109/10409238.2012.735643.
- Cífková E, Holčapek M, Lísa M, Vrána D, Gatěk J, Melichar B. Determination of lipidomic differences between human breast cancer and surrounding normal tissues using HILIC-HPLC/ESI-MS and multivariate data analysis. Anal Bioanal Chem. 2015;407(3):991–1002. doi: 10.1007/s00216-014-8272-z.
- Ide Y, Waki M, Hayasaka T, Nishio T, Morita Y, Tanaka H, Sasaki T, Koizumi K, Matsunuma R, Hosokawa Y, et al. Human breast cancer tissues contain abundant phosphatidylcholine(36 : 1) with high stearoyl-CoA desaturase-1 expression. PLOS One. 2013;8(4):e61204. doi: 10.1371/journal.pone.0061204.
- Cheng M, Bhujwalla ZM, Glunde K. Targeting phospholipid metabolism in cancer. Front Oncol. 2016;6:266. doi: 10.3389/fonc.2016.00266.
- Zhang Y, Liu Y, Li L, Wei J, Xiong S, Zhao Z. High resolution mass spectrometry coupled with multivariate data analysis revealing plasma lipidomic alteration in ovarian cancer in Asian women. Talanta. 2016;150:88–96. doi:10.1016/j.talanta.2015.12.021.
- Chen X, Chen H, Dai M, Ai J, Li Y, Mahon B, Dai S, Deng Y. Plasma lipidomics profiling identified lipid biomarkers in distinguishing early-stage breast cancer from benign lesions. Oncotarget. 2016;7(24):36622–31. doi:10.18632/oncotarget.9124.
- Saito K, Arai E, Maekawa K, Ishikawa M, Fujimoto H, Taguchi R, Matsumoto K, Kanai Y, Saito Y. Lipidomic signatures and associated transcriptomic profiles of clear cell renal cell carcinoma. Sci Rep. 2016;6(1):28932. doi:10.1038/srep28932.
- Zhu ZJ, Qi Z, Zhang J, Xue WH, Li LF, Shen ZB, Li ZY, Yuan YL, Wang WB, Zhao J. Untargeted metabolomics analysis of esophageal squamous cell carcinoma discovers dysregulated metabolic pathways and potential diagnostic biomarkers. J Cancer. 2020;11(13):3944–54. doi: 10.7150/jca.41733.
- Chen Y, Ma Z, Shen X, Li L, Zhong J, Min LS, Xu L, Li H, Zhang J, Dai L. Serum lipidomics profiling to identify biomarkers for non-small cell lung cancer. Biomed Res Int. 2018;7:5276240. doi: 10.1155/2018/5276240.
- Lu Y, Chen J, Huang C, Li N, Zou L, Chia SE, Chen S, Yu K, Ling Q, Cheng Q, et al. Comparison of hepatic and serum lipid signatures in hepatocellular carcinoma patients leads to the discovery of diagnostic and prognostic biomarkers. Oncotarget. 2017;9(4):5032–43. doi: 10.18632/oncotarget.23494.
- Lee GB, Lee JC, Moon MH. Plasma lipid profile comparison of five different cancers by nanoflow ultrahigh performance liquid chromatography-tandem mass spectrometry. Anal Chim Acta. 2019;1063:117–26. doi: 10.1016/j.aca.2019.02.021.
- Cheng F, Wen Z, Feng X, Wang X, Chen Y. A serum lipidomic strategy revealed potential lipid biomarkers for early-stage cervical cancer. Life Sci. 2020;260:118489. doi: 10.1016/j.lfs.2020.118489.
- Nagata S, Suzuki J, Segawa K, Fujii T. Exposure of phosphatidylserine on the cell surface. Cell Death Differ. 2016;23(6):952–61. doi: 10.1038/cdd.2016.7.
- Luo X, Cheng C, Tan Z, Li N, Tang M, Yang L, Cao Y. Emerging roles of lipid metabolism in cancer metastasis. Mol Cancer. 2017;16(1):76. doi: 10.1186/s12943-017-0646-3.
- Jones SF, Infante JR. Molecular pathways: fatty acid synthase. Clin Cancer Res. 2015;21(24):5434–8. doi: 10.1158/1078-0432.CCR-15-0126.
- Jesionowska A, Cecerska-Heryc E, Matoszka N, Dolegowska B. Lysophosphatidic acid signaling in ovarian cancer. J Recept Signal Transduct Res. 2015;35(6):578–84. doi: 10.3109/10799893.2015.1026444.
- Röhrig F, Schulze A. The multifaceted roles of fatty acid synthesis in cancer. Nat Rev Cancer. 2016;16(11):732–49. doi: 10.1038/nrc.2016.89.
- Qu Q, Zeng F, Liu X, Wang QJ, Deng F. Fatty acid oxidation and carnitine palmitoyltransferase I: emerging therapeutic targets in cancer. Cell Death Dis. 2016;7(5):e2226. doi: 10.1038/cddis.2016.132.
- Klupczynska A, Plewa S, Kasprzyk M, Dyszkiewicz W, Kokot ZJ, Matysiak J. Serum lipidome screening in patients with stage I non-small cell lung cancer. Clin Exp Med. 2019;19(4):505–13. doi: 10.1007/s10238-019-00566-7