Roles of Lipid Profiles in Human Non-Small Cell Lung Cancer.

Aims: This review aims to identify lipid biomarkers of non-small cell lung cancer (NSCLC) in human tissue samples and discuss the roles of lipids in tissue molecular identification, the discovery of potential biomarkers, and surgical margin assessment.
Methods: A review of the literature focused on lipid-related research using mass spectrometry (MS) techniques in human NSCLC tissues from January 1, 2015, to November 20, 2020, was conducted. The quality of included studies was assessed using the QUADAS-2 tool.
Results: Twelve studies met the inclusion criteria and were included in the review. The risk of bias was unclear in the majority of the studies. The contents of lipids including fatty acids, phosphatidyl choline, phosphatidyl ethanolamine, phosphatidyl inositol, cardiolipin, phosphatidyl serine, phosphatidyl glycerol, ceramide, lysophosphatidylethanolamine, lysophosphatidylcholine, and lysophosphatidylglycerol differed significantly between cancer and healthy tissues. The sensitivity or specificity of the discrimination model was reported in 8 studies, and the sensitivity and specificity varied among the reported methods. The lipid profiles differed between adenocarcinoma and squamous cell carcinoma NSCLC subtypes.
Conclusion: In preclinical studies, MS analysis and multiple discrimination models can be combined to distinguish NSCLC tissues from healthy tissues based on lipid profiles, which provides a new opportunity to evaluate the surgical margin and cancer subtype intraoperatively. Future studies should provide guidance for selecting patients and discrimination models to develop an improved method for clinical application.

Introduction

Non-small cell lung cancer (NSCLC) is the leading cause of cancer death worldwide.[1] NSCLC, which includes adenocarcinoma (AC), squamous cell carcinoma (SCC), large-cell lung carcinomas, and adenosquamous carcinoma, accounts for approximately 85% of all lung cancers.[2] Currently, chest radiography, computerized tomography, bronchoscopy, endoscopic ultrasound, endobronchial ultrasound, and pathological diagnosis are the main diagnostic strategies for lung cancer.[3-5] The treatment options and overall survival rate of NSCLC depend on the stage. Surgical resection is the gold-standard treatment for early-stage NSCLC.[6-8] Unfortunately, 3.6% to 60% of patients with NSCLC suffer recurrence after surgery.[9-13] Furthermore, the recurrence rate of stage IB NSCLC patients with negative pathological margins was reported to be 12.7% to 13.79%.[14] This suggests that the lung cancer surgical margin determined by conventional histologic pathological diagnosis may not provide reliable information from onco-biology, and emerging evidence suggests that current onco-surgical techniques are frequently inadequate.[15-18] Therefore, there is an urgent need to develop a novel technique or device for accurately assessing the molecular margin based on tumor biology.

Lipidomics is a new field that focuses on lipids and their interactions in objects, tissues, and cells.[19] Lipids including fatty acids, glycerolipids, glycerophospholipids, sphingolipids, sterol lipids, prenol lipids, saccharolipids, and polyketides are currently understood as being relevant to tumor biology. Phospholipids including sphingolipids and phosphoglycerides are the major structural molecules of cell membranes and have important roles in maintaining barrier function and fluidity.[20] Phospholipids also act as the second messenger in intercellular and intracellular signal transduction.[21] Sphingolipids including sphingomyelin and glycosphingolipid regulate cancer cell death and survival by controlling cancer cell signal transduction.[22] Phosphoglycerides including phosphatidyl choline (PC), cardiolipin (CL), phosphatidyl ethanolamine (PE), phosphatidyl serine (PS), phosphatidyl glycerol (PG), and phosphatidyl inositol (PI) play important roles in activating cells, maintaining metabolism, and enhancing human immunity and regeneration.[20,23] In cancer cells, fatty acids (FAs) play crucial roles in membrane formation, energy storage, and signaling.[24]

Tumor molecular analysis has attracted considerable attention as it can provide additional information to incorporate cancer-related biomarkers into clinical decision-making.[25-27] The majority of analytical methods for determining lipid profiles are based on mass spectrometry (MS). Ambient MS analysis allows the extensive and rapid analysis of metabolites, resulting in multiparameter datasets containing quantitative information on a range of metabolites.[28,29] Recently, based on the direct profiling of molecules from various biological samples, MS methods for identifying tumor tissues have advanced rapidly.[5,30,31] In particular, the lipid profiling of biological tissues by MS has been developed for the molecular diagnosis of cancers.[5,30] This review focuses on recent applications of MS for lipid profiling in human NSCLC tissues and discusses the roles of lipids in discriminating cancer from healthy tissue, evaluating tumor stage and subtype, and assessing disease-specific biomarkers of NSCLC. Specifically, this review summarizes the applications of MS for the molecular diagnosis of NSCLC, including tissue molecular identification, the discovery of potential biomarkers, and surgical margin assessment. Finally, this review discusses the challenges and future perspectives for the use of MS in the precise clinical molecular diagnosis of NSCLC.

Methods

A literature search of papers published in English was performed using Web of Science and PubMed (Medline). The search period was from January 1, 2015, to November 20, 2020, and the search focused on studies of lipid profiling in human NSCLC tissue. The search strategy employed a combination of MeSH terms: “mass spectrometry” [All Fields] AND (“lung neoplasms” [MeSH Terms] OR (“lung” [All Fields] AND “neoplasms” [All Fields]) OR “lung neoplasms” [All Fields] OR (“lung” [All Fields] AND “cancer” [All Fields]) OR “lung cancer” [All Fields]) AND (“lipids” [All Fields] OR “lipidate” [All Fields] OR “lipidated” [All Fields] OR “lipidates” [All Fields] OR “lipidation” [All Fields] OR “lipidations” [All Fields] OR “lipide” [All Fields] OR “lipides” [All Fields] OR “lipidic” [All Fields] OR “lipids” [MeSH Terms] OR “lipids” [All Fields] OR “lipid” [All Fields]). Two reviewers independently screened the potentially relevant studies by title and abstract via electronic search. The references of the returned papers were also searched to find additional potentially relevant articles. We included studies on the lipid profiles for human cancer and matched normal tissues from patients with NSCLC. Only studies that included MS as an analytical platform were included. The following studies were excluded: studies on biological samples other than tissue; cell line studies; animal studies; and studies without a control group. The following data were independently extracted from the selected studies by 2 reviewers: primary author; year of publication; study country/region; cancer subtype(s); analytical platform(s); discriminant model(s); sensitivity and specificity (if reported); and significantly different metabolites between the cancer and control groups. The main outcome measure was lipid abundance that was determined to be significantly different between cancer and normal samples. The quality of each included study was assessed using the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool. The risk of bias was unclear in the majority of studies.

Results

Study Characteristics

Twelve studies met the inclusion criteria and were selected for review. The search results are shown in Supplemental Figure 1. The analytical platforms used for lipid detection included liquid chromatography–MS,[32-35] electrospray ionization–MS,[36] nanoelectrospray ionization,[37] air flow-assisted desorption electrospray ionization MS imaging,[38,39] liquid extraction electrosonic spray ionization MS imaging,[40] matrix-assisted laser desorption/ionization MS (MALDI-MS),[35,36] MasSpec Pen,[16] desorption electrospray ionization,[16] tissue spray ionization MS,[41] and internal extractive electrospray ionization MS.[17] The cancer types included AC, SCC, large-cell lung carcinoma, and others (Table 1).

Table 1.

Summary of Included Studies.

Author, country/region and year of study	Analysis platform	Cancer type	Lipids of cancer compared to control
			Down-regulated	Up-regulated
Yingying Fan (China 2020)	UHPLC-MS	AC, SCC	FFA, CL, PI, SM, Cer	LPG, PG, TG, PEs, PCs
Yusuke Takanashi (Japan 2020)	LC-MS/MS	AC	TG (15:0_14:0_14:0)	SM (d35:1), Cer (d42:0)
Wenbo Cao (China 2020)	nanoESI-MS	NSCLC	C = C location PC(34:1), C = C location PC(36:1), C = C location PC(36:2)	sn-position isomers PC(34:1), sn-position isomers PC(38:5)
Haiyan Lu (China 2019)	iEESI-MS/MS	Lung cancer	PC(32:0)	PC(34:1)
Lei You (China 2020)	LC-MS	AC, SCC, others	–	LPC(20:1), PC(38:2), LPE(16:0), PE(34:3)
Min Zhang (China 2019)	AFADESI-MSI	AC, SCC	–	FA(20:5); FA(20:3); FA(20:2); FA(22:6); FA(22:5); FA(22:4); FA(22:2); FA(24:4); PC(16:0/18:1); PC(34:1)
Eyra Marien (Belgium 2015)	ESI-MS, MALDI-MS	AC, SCC, Large	SM(40:1),SM(42:1), SM(36:1)	PI(38:3), PI(40:3), PI(38:2)
Lei Guo (China 2020)	LE-ESSI-MSI	AC, SCC, Large, NSCC	SM(34:1), SM(42:2)	FA(20:3), FA(20:4), FA(22:4), FA(22:5), FA(22:6), PG(36:4), PI(34:1), PI(36:3), PI(36:4), PI(38:3), PI(38:4), PI(40:4), PI(40:5), PC(34:1), PC(34:3), PC(34:4), PC(36:3), PC(36:4), PC(38:3)
Jialing Zhang (America 2017)	MasSpec Pen, DESI-MSI	AC, SCC, others	PI(38:4), PE(36:1)	PI(36:1), PG(36:2), PG(34:1), FA(18:1）
Yusuke Muranishi (Japan 2019)	MALDI-IMS, LC-MS	AC	PC(16:0/16:0), SM (42:2)	–
Yiping Wei (China 2015)	TSI-MS	SCC	DPPC, POPC	DOPC, SAPC
Tiegang Li (China 2015)	AFADESI-MSI	AC, SCC	–	PC(16:0/18:2), PC(34:1)

Abbreviations: LC-MS, liquid chromatography mass spectrometry; ESI-MS, electrospray ionization mass spectrometry; nanoESI, nanoelectrospray ionization; AFADESI-MSI, air flow-assisted desorption electrospray ionization mass spectrometry imaging; LE-ESSI-MSI, liquid extraction electrosonic spray ionization mass spectrometry imaging; MALDI-MS, matrix-assisted laser desorption/ionization mass spectrometry; DESI-MSI, desorption electrospray ionization mass spectrometry imaging; TSI-MS, Tissue spray ionization mass spectrometry, iEESI-MS, internal extractive electrospray ionization mass spectrometry; AC, adenocarcinomas; SCC, squamous cell carcinomas; large, large cell lung carcinomas; PC, phosphatidyl choline; PE, phosphatidyl ethanolamine; PI, phosphatidyl inositol; CL, cardiolipin; PS, phosphatidyl serine; PG, phosphatidyl glycerol; LPE, Lysophosphatidylethanolamine; LPC, lysophosphatidylcholines; LPG, lysophosphatidylglycerol; FA, fatty acids; FFA, free fatty acids; SM, sphingomyelin; Cer, ceramide; TG, Triglyceride; DPPC, dipalmitoyl phosphatidylcholine; POPC, phosphatidylcholine; DOPC, oleoyl phosphatidylcholine; SAPC, arachidonic acid stearoyl phosphatidylcholine.

Lipid Profiling in Human NSCLC

FAs and phospholipids have been extensively studied in human NSCLC. The profiles of FAs were found to be significantly different between the cancer and control samples in 4 studies on NSCLC. There is a general trend of FA upregulation in cancer compared to normal across the majority of studies. However, Fan et al[32] reported that the abundance of free fatty acids (FFAs) was lower in lung cancer tissue compared to distal noncancerous tissue. Sphingomyelin (SM) was identified as a potential biomarker in 4 studies. In most of the studies, SM was downregulated in tumor tissue compared to normal tissue. However, Kahyo et al found that SM (d35:1) was increased in adenocarcinoma tissue compared to normal from the recurrent group patients.[33] The major structural lipids in membranes are glycerophospholipids including PC, PE, PI, CL, PS, and PG, which have been extensively examined in NSCLC research. Messenger lipids such as lysophosphatidylglycerol (LPG) and ceramide (Cer) were upregulated in tumor tissue compared to normal tissue. One study found that the level of Cer was lower in tumor tissue than in normal tissue due to the conversion of Cer into its derivatives.[32] Another 2 studies found that lysophosphatidylethanolamine, lysophosphatidylcholine (LPC), and LPG were upregulated in tumor tissue compared to in healthy tissue (Table 1).[32,34]

Sensitivity and Specificity of the Discrimination Model in Human NSCLC

The sensitivity and specificity of the diagnostic discrimination model are critical factors. The reviewed studies used the following models to differentiate between tumor tissue and normal tissue (Table 1): partial least squares discriminant analysis (PLS-DA; n = 2); principal component analysis (PCA; n = 3); orthogonal projections to latent structures discriminant analysis/orthogonal partial least-squares discriminant analysis (OPLS-DA; n = 5); PCA and linear discriminant analysis (PCA-LDA; n = 1); and partial least squares linear discriminant analysis (PLS-LDA; n = 1). The sensitivity or specificity of the discrimination model was reported in 8 studies (Table 2). The sensitivity or specificity varied among these studies. In 3 studies, the sensitivity of the discrimination model reached 100%,[17,33,36] while only one study reported a specificity of 100% (for the OPLS-DA in Li et al[39]). Interestingly, Zhang et al reported sensitivities of 85.2% for AC and 82.1% for SCC using the OPLS-DA model.[38]

Table 2.

The sensibility and Specificity of Discriminant Model of Included Studies.

Author, country/region and year of study	Discriminant model	Cancer type	Sensibility (%)	Specificity (%)
Yingying Fana (China 2020)	PLS-DA	AC, SCC	94	67
Yusuke Takanashi (Japan 2020)	PCA	AC	100	80
Haiyan Lu (China 2019)	OPLS-DA	Lung cancer	100	51
Lei You (China 2020)	PLS-DA	AC, SCC, others	94.10	84.40
Min Zhang (China 2019)	OPLS-DA	AC, SCC	85.2% of AC, 82.1% of SCC	–
Eyra Marien (Belgium2015)	PCA-LDA	AC, SC, Large	100	96.70
Jialing Zhang (America 2017)	PCA	AC, SCC, others	97.90	95.70
Tiegang Li (China 2015)	OPLS-DA	AC, SCC	93.50	100

Abbreviations: PLS-DA, partial least squares-discriminant analysis; PCA, principal component analysis; OPLS-DA, orthogonal projections to latent structures discriminant analysis/orthogonal partial least squares discriminant analysis; PCA-LDA, principal component analysis and linear discriminant analysis; PLS-LDA, partial least squares linear discriminant analysis; AC, adenocarcinoma; SCC, squamous cell carcinoma; large, large-cell lung carcinomas.

Lipid Profiling in Human NSCLC Subtypes

Abundant evidence has shown that the lipid profile is closely related to the NSCLC subtype. Due to differences in research design, analysis methods, discrimination models, and so on, the lipid profiles reported for human NSCLC subtypes show great variations among studies. In 4 of the reviewed studies, the lipid profiles of patients with AC and SCC were analyzed (Table 3). Fan et al reported that PE(20:0_18:1), DG(18:0_20:3), DG(18:2_18:2), PE(22:0_20:4), PE(18:0_20:4), DG(18:1_18:2), DG(36:2), PC(18:1_20:4), DG(16:0_18:2), and Cer(d18:1/26:0) were increased in AC patients compared to SCC patients.[32] Another study found that [PC(34:0) + Na]+, [phosphorylcholine + K]+, and [phosphorylcholine + Na]+ were upregulated in AC patients compared to SCC patients.[38] In contrast, You et al[34] found that the FFA22:1, FFA24:1, LPC20:1, and PC38:2 were downregulated in AC compared to SCC.

Table 3.

Lipids profiling research in human NSCLC Subtype.

Author, country/region and year of study	Analysis platform	Discriminant model	Differential molecules	Lipids of AC compared with SCC
				Down-regulated in AC	Up-regulated in AC
Yingying Fan (China 2020)	UHPLC-MS	PLS-DA	PE(20:0_18:1), DG(18:0_20:3), PE-O 39:4, DG(18:2_18:2), PE(22:0_20:4), PE(18:0_20:4), DG(18:1_18:2), DG 36:2, PC(18:1_20:4), DG(16:0_18:2), Cer(d18:1/26:0)	–	PE(20:0_18:1), DG(18:0_20:3), DG(18:2_18:2), PE(22:0_20:4), PE(18:0_20:4), DG(18:1_18:2), DG(36:2), PC(18:1_20:4), DG(16:0_18:2), Cer(d18:1/26:0)
Lei You (China 2020)	LC-MS	PLS-DA	FFA22:1, FFA24:1, LPC 20:1, PC 38:2, LPE 16:0, PE 34:3, SM 35:2	FFA22:1, FFA24:1,LPC 20:1, PC 38:2,	LPE 16:0, PE 34:3, SM 35:2
Min Zhang (China 2019)	AFADESI-MSI	OPLS-DA	[PC(34:0) + Na]+, [phosphorylcholine + K]+, [phosphorylcholine + Na]+	–	[PC(34:0) + Na]+, [phosphorylcholine + K]+, [phosphorylcholine + Na]+
Eyra Marien (Belgium2015)	ESI-MS MALDI-MS	PCA-LDA	PI36:4, PS40:8, PS42:9, PC40:2, PE44:5, PI36:3, PE42:2, SM36:2	–	–

Abbreviations: LC-MS, liquid chromatography mass spectrometry; ESI-MS, electrospray ionization mass spectrometry; AFADESI-MSI, air flow-assisted desorption electrospray ionization mass spectrometry imaging; MALDI-MS, matrix-assisted laser desorption/ionization mass spectrometry; AC, adenocarcinoma; SCC, squamous cell carcinoma; PC, phosphatidyl choline; PE, phosphatidyl ethanolamine; PI, phosphatidyl inositol; PS, phosphatidyl serine; LPE, lysophosphatidylethanolamine; LPC, lysophosphatidylcholines; FFA, free fatty acid; SM, sphingomyelin; Cer, ceramide; DG, diacylglycerol; PLS-DA, partial least squares discriminant analysis; PCA, principal component analysis; OPLS-DA, orthogonal projections to latent structures discriminant analysis/orthogonal partial least squares discriminant analysis; PCA-LDA, principal component analysis and linear discriminant analysis; PLS-LDA, partial least squares linear discriminant analysis.

Discussion

Lipids Profile Research and Surgical Margin Assessment in Human NSCLC

Lipids are involved in energy storage, are components of the cell membrane, and serve as messengers in signal transduction and molecular recognition processes. In cancer cells, lipid metabolism is disordered, which can affect numerous cellular processes such as cell growth, proliferation, differentiation, and motility. Lipids are a major component of cell membranes, and increasing evidence suggests that lipids play a crucial role in the occurrence and development of NSCLC.[30] Since lung cancer is the leading cause of cancer-related death among both men and women, more and more research is focusing on the role of lipid metabolism in lung cancer.[42-45] Increasing evidence suggests that disordered lipid metabolism is important when evaluating tumor stage and metastasis, determining the effectiveness of treatment, and developing new therapeutic targets and disease-specific biomarkers.[46-48] Various biological samples including tissues, surgical aerosols, and sera have been used in studies on lipid metabolism in lung cancer.[5,30] The serum-based lipidomic metabolism of NSCLC has been used to identify serum lipid biomarkers for the early screening and detection and NSCLC and the assessment of NSCLC subtypes.[32,49,50] Interestingly, differences in lipid metabolism in surgical aerosols have been used to determine the boundary of lung cancer intraoperatively based on rapid evaporative ionization MS; using this approach, a sensitivity of 97.7% and a specificity of 96.5% were achieved for the binary classifications of cancer and healthy tissue.[51] In studies on human NSCLC tissue, excellent models for the discrimination of cancer and healthy tissues have been established based on lipids. These methods provide a new opportunity for the real-time, in situ evaluation of surgical margins during surgery based on molecular characterization.

In previous studies, type II alveolar epithelial cells, a type of lung stem cell, were transformed into growing monoclonal lung tumors during active KRAS mutation.[52] Pulmonary surfactant, a complex mixture of phospholipids (85% phosphatidylcholine) and surfactant proteins, was synthesized and secreted by type II epithelial cells.[53] Sin et al suggested that pro-surfactant protein B can serve as a biomarker to predict lung cancer.[54] Furthermore, Umeda et al[55] reported that higher levels of serum surfactant protein D were correlated with a lower number of distant metastases and prolonged overall survival and progression-free survival. The phospholipids in pulmonary surfactant are mainly in the form of dipalmitoylphosphatidylcholine (DPPC). In a previous study, we found that in SCC patients, cancer tissue contained less DPPC than healthy tissue.[41] These findings suggest that abnormal pulmonary surfactant and the destruction of type II cell homeostasis are closely related to the occurrence of lung cancer.

Lipid Profiles for NSCLC Screening

Previous studies suggest that plasma phospholipids including phosphatidylcholine and LPC are potential biomarkers of lung cancer.[56] Louis et al reported that the concentrations of SM and phosphatidylcholine were decreased in the blood plasma of lung cancer patients compared to normal.[57] Increasing research is focusing on the analysis of serum lipid profiles and the identification of potential serum phospholipids as biomarkers for the diagnosis and screening of human NSCLC.[49,50] Serum phospholipid may be suitable as biomarkers for convenient lung cancer screening without the need for radiation. Due to the complex composition of blood plasma, the specificity and sensitivity of reported detection methods vary, making these methods difficult to apply in clinical diagnosis.

Lipid Profiling for the Evaluation of Human NSCLC Stage and Subtype

The need for clinical treatment for lung cancer is decided based on the tumor stage and subtype. Lipid metabolism of NSCLC subtypes is critical for understanding the mechanisms, identifying potential diagnostic biomarkers, and targeted therapy. The differences in the characteristics of lipids between human AC and SCC tissues have been studied. In 2012, the lipid profiles of AC and ACC tissues were detected by MALDI-MS,[58] providing a strategy to discriminate the 2 major histologic types of NSCLC. Subsequently, in 2015, Marien et al[36] revealed that the lipids in AC and SCC tissues were different. Increasing evidence suggests that lipids are potential biomarkers of AC and SCC.[32,34,36,38] Interestingly, Takanashi et al[33] found that SM (d35:1) can be used as a predictor for lung AC recurrence after surgery. The different lipid profiles of AC and SCC tissues provide a way to evaluate tumor stage and prognosis and guide treatment decisions for different tumor subtypes.

Advantages and Applications of Lipid Profiles to Diagnosis Human Non-Small Cell Lung Cancer

In situ and real-time identification of tumor margins are essential for tumor surgery to obtain reliable curative resection and accurate prognosis, as well as to minimize losses of healthy tissue.[41] Frozen section histology is the gold-standard method to accurately determining tumor margin during surgery, but this approach suffers from shortcomings, such as time-consuming (mean turnaround time of 30-40 min), subjective interpretation of the results (diagnosis results depend on the experience of the pathologist), increased risk of surgery (during this time the patient remains under anesthesia), and so on.[41,51,59] Lipids are one of the important units for the human, and it plays a key role in activating cells, maintaining metabolism, enhancing human immunity and regeneration, maintaining cell barrier function, controlling cell signal transduction and so on.[20-24] Most importantly, lipids constitute the outermost structure of cells, which is beneficial for the diagnosis of diseases. MS analysis is increasingly applied to rapid and accurate lipids diagnosis of cancer, based on the advantages of real-time and direct analysis of lipids information from biological samples without or minimal sample pretreatment.[5] As mass spectrometry can provide rich lipid information on and in the tissue which can be correlated with pathology. In situ and real-time lipid profiles analysis by MS may be an alternative to standard frozen-section histology. Due to no/minimal sample pretreatment, numerous MS techniques allow the direct lipid profiling from tissues, showing potential in the real-time diagnosis of lung cancer.[16,17,32-41] The application of lipid biomarkers and multiple discrimination models to lung cancer detection has only been demonstrated in laboratory-based studies. Unfortunately, the risk of bias for patient selection was unclear in the reviewed studies. It is unclear whether consecutive patient data were included or whether highly selected samples were chosen. The application of lipid biomarkers and multiple discrimination models to lung cancer detection has only been demonstrated in laboratory-based studies. Farther, there is a lack of standards, including study designs, models, objects, methods, and analyses, which might lead the data was chosen to report. Future researches should design random, double-blind, multicenter, use a unified distinction model to evaluate the feasibility of lipid profile for diagnosis NSCLC compare with the immunohistochemistry results.

Conclusion and Perspectives

The review demonstrates the variations in the relative abundances of the metabolites of phospholipids and FAs in tissue samples from patients with NSCLC. MS can be combined with different discrimination models to distinguish cancer tissue from healthy tissue based on differences in lipid composition, providing a new way to evaluate surgical margins in situ and in real time in intraoperative settings. In situ and real-time lipid profiles analysis by MS may be an alternative to standard frozen-section histology. Future studies should provide details on how the patients and discrimination model(s) were selected and develop a suitable analysis method for clinical application.

Click here for additional data file.

Supplemental material, sj-tiff-1-tct-10.1177_15330338211041472 for Roles of Lipid Profiles in Human Non-Small Cell Lung Cancer by Zhang Jianyong, Huang Yanruo, Tang Xiaoju, Wei Yiping and Luo Fengming in Technology in Cancer Research & Treatment