Subscriber access provided by Nottingham Trent University
Letter
Hepatocarcinoma Discrimination by Ratiometric Lipid Profiles Using Tip-Contact Sampling/Ionization Mass Spectrometry Xiaoming Chen, Jiaqi Gao, Tao Wang, Xinrong Jiang, Jiang Chen, Xiao Liang, and Jianmin Wu Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.9b02623 • Publication Date (Web): 29 Jul 2019 Downloaded from pubs.acs.org on July 30, 2019
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 6 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
Hepatocarcinoma Discrimination by Ratiometric Lipid Profiles Using Tip-Contact Sampling/Ionization Mass Spectrometry Xiaoming Chen1‡, Jiaqi Gao2‡, Tao Wang1, Xinrong Jiang1, Jiang Chen2, Xiao Liang2*, Jianmin Wu1* 1
Institute of Analytical Chemistry, Department of Chemistry, Zhejiang University, Hangzhou, 310058, P. R. China. Department of General Surgery, Sir Run Run Shaw Hospital, School of Medicine, Zhejiang University, Hangzhou 310016, P. R. China 2
ABSTRACT: Precise diagnosis at the molecular level is essential for the improvement of surgery and treatment. High-throughput and spatial-resolved mass spectrometric (MS) methods for in situ detection of metabolites on tissue samples can reveal the dysregulation of metabolism in abnormal tissue and help identification of tumor. We here report a non-destructive MS method named as tip-contact sampling/ionization (TCSI)-MS technology which can quickly acquire lipidomic information from liver tissue and thereby realize tumor identification. Using this technology, fatty acids and lipids at the liver tissue surface can be rapidly imprinted onto silicon nanowire tip attached with reduced graphene oxide (rGO) and sensitively detected by on-chip MS. With proper data pretreatment and statistical analysis, the clinical primary hepatocellular carcinoma (HCC) tissues can be discriminated from the non-tumor parts. In addition, we found that a panel of adjacent dual-peaks’ ratio can be used to build a prediction model in artificial neural networks (ANN), resulting in high accuracy (91.7%~98.3%) for tumor discrimination. Ratiometric TCSI-MS imaging using selected dual-peaks’ ratio can greatly enhance the spatial resolution of tumor margin. The feature ratiometric data of lipid molecules may guide the study of metabolism pathways involved in hepatocarcinoma and ultimately become new metabolic biomarkers in clinical diagnosis. The present work demonstrated that the TCSI-MS technology may pave a novel way for surgery guidance and precision diagnosis in tissue biopsy.
Hepatocellular carcinoma (HCC) is the major type of primary liver cancer,1 which has becoming the second most common cause of cancer related mortality worldwide after lung cancer.2 For most patients, surgical excision is the mainstay treatment for HCC.3 However, the recurrence of HCC still remains a challenging and the 5-year survival rate is around 47-53% even for very early-stage HCC (< 3 cm) patients who undergo surgery.4 Improved precision tumor diagnosis is essential for the optimization of surgical treatment and may contribute to the improved outcomes. Typically, liver cancer tissue discrimination and surgical resection evaluation are mostly dependent on the clinical experience and various medical imaging techniques.5, 6 Histological examination remains the gold standard in the diagnosis of liver cancer.7 Sometime postoperative immunohistochemical staining is required, but it takes a long time and a lot of resources to evaluate the expression of various biomarkers.8 Fluorescence molecular imaging (FMI) using specific molecules, such as indocyanine green (ICG), has been applied in surgery for liver tumor localization based on its slower metabolic rate in cancerous liver tissue.9 Nevertheless, benign liver disease, such as cirrhosis, also damages the function of liver, thus interfering with the decision-making of tumor diagnosis.9 Recently, intraoperative mass spectrometry (MS) diagnosis technologies based on the multidimensional molecular signatures of tumors have been emerging recently.10 In vivo detection on tissues for tumor discrimination was achieved by rapid evaporative ionization mass spectrometry (REIMS)11 and Spider mass spectrometry,12 but the electro-surgery or laser ablation process would result in thermal and mechanical damage of the analyzed tissue. MasSpec Pen, a nondestructive sampling probe, has been precisely designed to be integrated to the ESI-MS instrument for in vivo tissue extraction and online cancer diagnosis.13 Other solid phase micro-extraction (SPME) probes followed with MS analysis or touch spray mass spectrometry (TS-MS) have provided alternative offline strategies for metabolites detection at the surface of cancerous tissues.14, 15 Effective in situ sampling and extraction coupled with off-line MS can avoid to interfere with the operation. However, each probe can only be applied for one-point sampling and not appropriate for high-throughput detection or
localization of tumor information. Therefore, it is still in urgent need to develop novel techniques, which can not only combine nondestructive extraction of metabolites from tissue surface and direct MS detection together, but also conserve the molecular distribution information for tumor imaging. Our previous work indicated that vertical silicon nanowire (SiNWs) array is not only an ideal substrate for surface assisted laser desorption/ionization mass spectrometry (SALDI-MS), but also can be acted as a sampling chip when contacted with tissues.16 Through a simple contact-imprinting method, endogenous molecules, especially lipids from tissues, can be transferred onto the tip surface of SiNWs chip for subsequent MS analysis. In the present work, we establish “Tip-Contact Sampling/Ionization Mass Spectrometry” (TCSI-MS) technique to reveal the lipid profiles and spatial distribution on tissues surface. Application of TCSI-MS in the discrimination and localization of HCC tumor on clinical tissue samples was investigated, taking into consideration that liver with malignancy would undergo dysregulation of fatty acid synthesis and lipid metabolism.17 Herein, SiNWs attached with reduced graphene oxide (SiNWs@rGO) were fabricated and employed for in situ lipid molecules extraction from clinical liver
Scheme 1. Scheme of tip-contact sampling and ionization mass spectrometry (TCSI-MS) for HCC tissue lipid extraction and detection on the SiNWs@rGO chip.
ACS Paragon Plus Environment
Analytical Chemistry [PC+Na]+, [SM+K]+ and [TG+Na]+ ions, which dominated the profile of PCs, SMs and TGs, were considered (PSs molecular ions were not stable and were abandoned). Coupled with all [MH]- ions in negative ion mode, a total of 114 peaks were selected out. Furthermore, the MS intensity of each molecule was normalized within each class of lipids after MS/MS identification with the assumption that molecules sharing similar structures have similar LDI efficiency. In another data treatment method, the neighboring lipid peaks with only mono-unsaturation degree difference were regarded as internal standards for each other, just like the isotope-labeling technique for MS quantification.19 As a result, 86 ratio values calculated from neighboring two peaks were used as the elements for statistical analysis. In comparison, the medium RSD of the ratio values can be controlled within 17.3%, which satisfies the requirements for analysis of clinical samples in biomarker discovery.20 In contrast, the normalization within each lipid class of selected 114 peaks can slightly reduce the medium RSD to 22.2% (Figure S4). Thus, ratio value-based database of lipid molecules acquired from liver tissues can more stably reflect the tissue characteristics.
cancer tissues followed with high-throughput MS detection (Scheme 1, details can be seen in Supplementary Experimental Procedures). Compared to bare SiNWs, the presence of rGO on the tip of SiNWs can significantly enhance the TCSI-MS efficiency (Figure S1, S2). To evaluate whether TCSI-MS technology can discriminate HCC from non-cancerous parts using their molecular information, 20 clinical human HCC tissues (abbreviated as T) along with adjacent para-tumor (PT, distance from the cancer tissues = 2 cm) and normal liver tissues at the cutting margin in the surgery (N, distance from the cancer tissues > 2 cm) were tested with the TCSI-MS. The demographic information of patients and their diagnostic details can be referred to Table S1, S2. As shown in Figure 1 and Figure S3, both cancerous and non-cancerous (PT or N) tissues displayed similar lipid profiles in negative or positive ion mode. After peak alignment and MS/MS detection, five major classes, including fatty acids (FAs), phosphatidylethanolamines (PEs), phosphatidylinositols (PIs), phosphatidylcholines (PCs) and triacylglycerols (TGs) with different acyl chain length or saturation along with several discrete peaks of sphingomyelins (SMs) and phosphatidylserines (PSs) can be identified (Table S3). Adductive forms can be verified using their specific fragments and calculated m/z values.
0.50 0.25
0 200
0.00 200
300
300
861.5
0 700
700 400
400
750 750
500
500
800
850
800
600
600
859.5
835.5
303.2 305.3
500
833.5
0.75
762.5
1000
764.5 790.5
1500
714.5 738.5 740.5
1.00
279.2 281.2 Intens. [a.u.]
1.0
1.25
700
900
800
800
m/z
900
900
N
1.0
3000
2000
1000
m/z
m/z
m/z
0 700
0 700
850
800
750
750
800
800
850
900
900
900
947.8
923.8
950
950
m/z m/z
923.8
PT
900
850
850
897.7 [TG 52:2+K]+ 907.8
881.8 879.7 881.8
844.5 853.7 [TG 50:2+Na]+
804.6
800
750
853.7 864.6 [PS 38:3+K]+
782.6 804.6 806.6 820.5 [PC 36:4+K]+
750
36:5+Na]+
713.5
0 700
0.0 700
907.8 [TG 54:3+Na]+
0.2 m/z
m/z
802.5 [PC
0.4
782.6 [PC 34:1+Na]+
Intens. [a.u.]
0.6
(f)
900
850
700
900
0.8
900
950
921.7 [TG 54:4+K]+
800
(e) x105
889.6 PI 38:2
861.6 PI 36:2 885.6 PI 38:4 887.6
700
900
1.0
1.0
T
900
897.7
600
900
800
1.2
PT
850
850
907.8
500
850
700
(d)
x104
800
800
881.8
800
600
750
869.7
400
500
900
750
700
853.7
750
400
850
0 700
804.6
300
300
800
m/z
802.6
700
750
m/z
782.6
0 200
0 200
700
863.5
1
900
739.5 754.5
1000
900
800
737.5
2000
800
700
711.5
3000
2
700
600
m/z
737.5 739.5 754.5
4000
900
900
711.5 [SM 32:2+K]+
5000
850
850
600
500
1.0
6000
2000
800
800
762.5 764.5 PE 38:5
6000
3
500
8000
Intens. [a.u.]
400
T
Intens. [a.u.]
400
300
(b)
4000
885.6
750
750
738.5 740.5 PE 36:3
1.0
x105
255.2
Intens. [a.u.] Normalized intensity
(c)
300
700
714.5 PE 34:2
0 200
0.0 200
255.2 279.2 303.2 281.2 Intens. [a.u.] 305.2 C20:3
0.5
0 700
887.6 909.5
500
835.5 PI 34:1
1000
1.0
833.5 PI 34:2 835.5 859.5 861.5 863.5 PI 36:1 885.5 887.6 PI 38:3
2000 1500
1.5
807.5 PI 32:1 833.5
2500
790.5
716.5 PE 34:1
2.0
3000
738.5 PE 36:4 740.5 762.5 PE 38:6 764.5
1.0
x105 2.5
281.2 C18:1 303.2 C20:4 Intens. [a.u.] 305.3
279.2 C18:2
Intens. [a.u.]
(a)
737.5 739.5 [SM 34:2+K]+ 754.5 760.6 [PC 34:1+H]+
A supervised statistical analysis method, linear discrimination analysis (LDA) was conducted for evaluation whether the data after pretreatment can be used for tumor identification, which was under the premise that T, PT and N were set as three distinctive groups. The result indicated that tumor tissue samples can be clearly discriminated from both para-tumor and normal liver tissues based on these ratiometric data (Figure 2). In contrast, the para-tumor and normal tissues partially overlap with each other. If PT and N were both presumed as non-T group, LDA result also showed an obvious discrimination of T, while PT and N were mixed together after linear transformation of ratiometric data (Figure S5A). Otherwise, incorrect precondition of group (e.g. T and N were set as the same group) would lead to a discrimination confusion (Figure S5B). After two-sample t test among T, PT and N groups, 22 ratios with significant difference (p < 0.01) between
In order to explore the intrinsic difference in the mass spectra of T and non-T groups, a robust data analysis strategy is needed. Due to the existence of different adducted molecular ions in positive-ion detection mode, their relative intensity was dependent on the salt composition at the tissue surface. On the other hand, lipids with different molecular structure have different LDI efficiency, ion suppression effect18 between different types of lipids cannot be neglected. These two factors can induce instability of acquired data if all detected peaks were selected and directly normalized within the entire detection range (Medium RSD = 30.7%, Figure S4a). Thus, appropriate adducted ions need to be selected followed with proper data pretreatment. In the present work, only
Intens. [a.u.]
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 6
950
m/z
m/z
N
950
950
m/z
m/z
Figure 1. Representative mass spectra of HCC tumor tissue (a, b), para-tumor tissue (c, d) and normal liver tissue (e, f) from the same patient (Specimen No. 1) in negative-ion (a, c, e) and positive-ion (b, d, e) detection modes. 279.2 (C18:2) vs 281.2 (C18:1), 303.2 (C20:4) vs 305.3 (C20:2), 762.5 (PE 38:6) vs 764.5 (PE 38:5), 885.6 (PI 38:4) vs 887.6 (PI 38:3), 737.5 ([SM 34:3+K]+) vs 739.5 ([SM 34: 2+K]+), 802.5 ([PC 36:5+Na]+) vs 804.6 ([PC 36:4+Na]+) are examples of adjacent “dual-peaks”. All peaks detected in negative ion mode were appeared as [M-H]- ions.
ACS Paragon Plus Environment
2
Page 3 of 6
T and PT were selected out, whereas no feature difference existed between PT and N, verifying the negative characteristics of both samples in the clinical diagnosis (Figure S6, Table S4). Although single-peak dataset can also be applied to distinguish tumor and 25 feature peaks can be found (Figure S7, Figure S8, Table S5), change of adjacent “dual-peaks” ratio was much easier to be perceived in the mass profiles for sensitively detecting molecular dysregulation on tumor tissues (Figure S3). To test whether the characteristic ratio values can predict the cancerous HCC sample, an artificial neural network (ANN) model was constructed. An overall accuracy in the range of 93.3%~98.3% or 91.7%~98.3% can be achieved when all 86 ratios or 22 feature ratios were input in the model (Table 1), respectively. The area under the ROC curve (AUC) equals 0.9488~1, indicating excellent performance of the model for HCC tumor discrimination (Table 1). Though individual variation did exist, the specific pattern of feature ratio values showed a distinctive panel in tumors compared with paratumors, if the normal liver tissue of each patient was regarded as the background (Figure 3). These results further indicate that TCSI-MS coupled with ratiometric-methodology show its great potential in the off-line cancer diagnosis.
Figure 2. LDA result of T, PT and N groups based on 86 adjacent dual-peaks’ ratio values when T, PT and N were presumed as three distinctive groups. Triplicates were done for each tissue sample. Thus, 60 dots (20 × 3) can be found in each class (T or PT or N) in the LDA plot score. Table 1. Summary of sensitivity, specificity, accuracy and AUC in five successive ANN model Database
All ratio values
Sensitivity
85.0%~95.0%
Feature ratio values (p < 0.01) 85.0%~100%
Specificity
95.0%~100%
95.0%~97.5%
Accuracy
93.3%~98.3%
91.7%~98.3%
AUC
0.9588~0.9825
0.9488~1
The feature ratio of adjacent “dual-peaks” might be potential biomarkers related to HCC. An interesting phenomenon was observed that the peak intensity ratios of m/z = 281.2 (C18:1) to m/z = 279.2 (C18:2) and m/z = 305.3 (C20:3) to m/z = 303.2 (C20:4) kept at a relative low level for most of the PT and N tissues, while reaching a much higher level on tumor tissues (Figure S6c, Figure S9a). Due to the in source dissociation (ISD) of lipids in the MALDI instrument, the fatty acid profiles can be considered as the total composition of the acyl chains involved in the lipid mixture. Taking the pair PI 38:3 vs PI 38:4 as an example, it can be identified by MS/MS that PI 38:3 incorporates the C20:3 acyl chain and PI 38:4 molecule is composed of a C20:4 fatty acid chain, respectively (Table S3). Other ratios of lipids, including IPI 34:1/IPI 34:2, IPI 36:1/IPI 36:2 and ITG 50:1/ITG 50:2 show the same tendency with IC18:1/IC18:2 while IPE 36:3/IPE 36:4, IPI 36:3/IPI 36:4 and IPI 38:3/IPI 38:4 showed a positive correlation with IC20:3/IC20:4 (Figure S9b). C20:4 is usually corresponded to arachidonic acid (AA) and is the desaturated product of dietary linoleic acid (LA, C18:2).21 AA and LA are the core molecules involved in the two essential liver metabolism pathways: linoleic acid metabolism and arachidonic acid metabolism (Figure S9c) (http://smpdb.ca/; https://www.genome.jp/kegg/). AA is served as a precursor for eicosanoids, which are regarded as local hormones and modulators of cell metabolism and functions. The down-regulation of AA or LA in tumor tissue or serum samples have been found in some studies.21-22 The up-regulation of C18:1, C20:3 and PI 38:3 have been verified closely relevant to the de novo FA synthesis or lipogenesis,23 which is thought as a hallmark of cancer. Thus, the finding of the dysregulation of these “dual-peaks” may imply an abnormal linoleic acid and arachidonic acid pathway involved in the liver malignancy, which could be used as the potential diagnostic biomarker and the therapy target in the future. Dysregulation of these ratio values are also effective indicator of tumor margin when ratiometric TCSI-MS imaging was conducted on the HCC tumor tissue (Figure 4). A rectangle SiNWs@rGO chip (2.5 mm × 10 mm) was pressed onto the surface of this HCC tissue for TCSI-MS of both T and PT regions. Distinct difference of several feature ratios can be found on these two parts (Figure 4f, Figure S10), whose characteristics were in accordance with what have been indicated in the previous statistical results (Table S4). The peak pairs at C18:2 vs C18:1 and C20:4 vs C20:3, which dominated the negative-ion mode MS profiles, were selected for MS imaging (Figure 4d, 4e). The digital imaging of the two ratio value on the chips after mirroring conversion perfectly coincided with the H&E images (Figure 4b). Compared with single-peak MS imaging (Figure S10d), the ratiometric MS imaging significantly enhanced the contrast between
(b) T
(a)
Adjacent PT ①
(c)
②
(e) 281.2/279.2 (C18:1/C18:2) 305.3/303.2 (C20:3/C20:4)
Figure 3. Changes in the feature ratio values of adjacent lipids in 20 HCC tumors (a) and para-tumor tissues (b) versus matched normal tissues. The ratio values were expressed as a fold change pattern (Log2) by calculating the ratio of tumor or para-tumor’s “feature ratio value” to their matched normal tissue. Each column represents a specimen and the data used was the average of three replicates for each tissue sample.
0 2 0.5 1.2
Mirroring
Intens. [a.u.]
5 mm
(d)
x104 6 4 2
Intens. [a.u.]
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
0 x105 2
(f)
HCC-imaging-T-n 0:F7 MS, BaselineSubtracted
* C18:1/C18:2 Region ①: T C20:3/C20:4 * ** HCC-imaging-PT-n 0:E8 MS Raw
Region ②: PT
**
1
2.5 mm0 260
280
280
**
300
300
320
320
340
m/z 340 m/z
Figure 4. Ratiometric TCSI-MS for HCC tumor imaging. (a-c) Optical image of a HCC tissue (a), H&E stain result (b) and the TCS process by contacting a SiNW@rGO chip onto the surface of the HCC tissue (c). (d, e) Ratiometric imaging result of IC18:1/IC18:2 and IC20:3/IC20:4 detected on the rectangle SiNWs@rGO chip before (d) or after (e) mirroring treatment. (f) Representative negative-ion mode mass spectra detected on region ① (T) and region ② (PT) at the fatty acid range.
ACS Paragon Plus Environment
3
Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
cancer and normal tissue. The dysregulation of lipogenesis on HCC tumors can be easily and clearly visualized using the ratio of “dual-peaks” derived from the close related fatty acids biomarkers, indicating the advantage of ratiometric TCSI-MS imaging for tumor localization. In conclusion, we have developed a novel method for hepatocellular carcinoma discrimination based on TCSI-MS technique. Through TCSI-MS, various feature lipid molecules can be effectively extracted to the SiNWs@rGO chip for subsequent on-chip MS analysis. Especially, the usage of the pseudo-internal standard for the adjacent peaks can overcome the shortage of the bad stability of original MS data and make the result more reliable. Our results suggest that HCC identification and prediction can be accomplished with the adjacent dual-peaks’ ratiomatric data. TCSIMS imaging results can retain the molecular spatial distribution on tissues, which not only help the tumor margin decision but also confirms the dysregulation of fatty acid metabolism on hepatocellular carcinoma. Owing to its non-destructive, rapid and spatial resolved feature, TCSI-MS technique may become a new tool for exploring cancer-related biomarkers, precision cancer diagnosis, as well as intraoperative guiding.
ASSOCIATED CONTENT Supporting Information The Supporting Information is available free of charge on the ACS Publications website. Experimental procedures, supplementary results and discussion, material characterization, representative mass spectra, magnified spectra and MS imaging on HCC tissues, statistical analysis results and graphs of feature peaks and feature ratios, MS/MS identification of lipids (PDF)
AUTHOR INFORMATION Corresponding Author *Jianmin Wu, Email:
[email protected] *Xiao Liang, Email:
[email protected] Author Contributions ‡X.
M. Chen and J. Q. Gao contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENT This study is supported by grants from the National Natural Science Foundation of China (NSFC, Grants No. 21874118, 21575127), the Natural Science Foundation of Zhejiang Province (Grants No. Y18H160118) and Scientific Research Projects of Zhejiang Education Department (Grants No. Y201738040).
REFERENCES (1) Guichard, C.; Amaddeo, G.; Imbeaud, S.; Ladeiro, Y.; Pelletier, L.; Maad, I. B.; Calderaro, J.; Bioulac-Sage, P.; Letexier, M.; Degos, F.; Clement, B.; Balabaud, C.; Chevet, E.; Laurent, A.; Couchy, G.; Letouze, E.; Calvo, F.; Zucman-Rossi, Integrated Analysis of Somatic Mutations and Focal Copy-Number Changes Identifies Key Genes and Pathways in Hepatocellular Carcinoma. Nat. Genet. 2012, 44, 694-698. (2) Torre, L. A.; Siegel, R. L.; Ward, E. M.; Jemal, A. Global Cancer Incidence and Mortality Rates and Trends--An Update. Cancer Epidemiol., Biomarkers Prev. 2016, 25, 16-27.
Page 4 of 6
(3) Mokdad, A. A.; Singal, A. G.; Yopp, A. C. Treatment of Liver Cancer. J. Am. Med. Assoc. 2016, 315, 100-100. (4) Altekruse, S. F.; McGlynn, K. A.; Dickie, L. A.; Kleiner, D. E. Hepatocellular Carcinoma Confirmation, Treatment, and Survival in Surveillance, Epidemiology, and End Results Registries. Hepatology 2012, 55, 476-482. (5) Bipat, S.; van Leeuwen, M. S.; Comans, E. F.; Pijl, M. E.; Bossuyt, P. M.; Zwinderman, A. H.; Stoker, J. Colorectal Liver Metastases: CT, MR Imaging, and PET for Diagnosis—MetaAnalysis. Radiology 2005, 237, 123-131. (6) Kruskal, J. B.; Kane, R. A. Intraoperative US of the Liver: Techniques and Clinical Applications. Radiographics 2006, 26, 10671084. (7) Ishak, K. G.; Anthony, P. P.; Sobin, L. H., Histological Typing of Tumours of the Liver. 1994. (8) Yamashita, T.; Forgues, M.; Wang, W.; Kim, J. W.; Ye, Q.; Jia, H.; Budhu, A.; Zanetti, K. A.; Chen, Y.; Qin, L. X.; Tang, Z. Y.; Wang, X. W. EpCAM and Alpha-Fetoprotein Expression Defines Novel Prognostic Subtypes of Hepatocellular Carcinoma. Cancer Res. 2008, 68, 1451-1461. (9) Kokudo, N.; Ishizawa, T. Clinical Application of Fluorescence Imaging of Liver Cancer Using Indocyanine Green. Liver Cancer 2012, 1, 15-21. (10) Zhao, Y. Y.; Miao, H.; Cheng, X. L.; Wei, F. Lipidomics: Novel Insight into the Biochemical Mechanism of Lipid Metabolism and Dysregulation-Associated Disease. Chem. Biol. Interact. 2015, 240, 220-238.. (11) Júlia, B.; László, S. S.; James, K.; Lewis, M. R.; Muirhead, L. J.; Kirill, V.; Reza, M.; Balázs, D.; László, D.; Ara, D. Intraoperative Tissue Identification Using Rapid Evaporative Ionization Mass Spectrometry. Sci. Transl. Med. 2013, 5, 153-154. (12) Saudemont, P.; Quanico, J.; Robin, Y. M.; Baud, A.; Balog, J.; Fatou, B.; Tierny, D.; Pascal, Q.; Minier, K.; Pottier, M.; Focsa, C.; Ziskind, M.; Takats, Z.; Salzet, M.; Fournier, I. Real-Time Molecular Diagnosis of Tumors Using Water-Assisted Laser Desorption/Ionization Mass Spectrometry Technology. Cancer Cell 2018, 34, 840-851. (13) Zhang, J.; Rector, J.; Lin, J. Q.; Young, J. H.; Sans, M.; Katta, N.; Giese, N.; Yu, W.; Nagi, C.; Suliburk, Nondestructive Tissue Analysis for Ex vivo and In vivo Cancer Diagnosis Using a Handheld Mass Spectrometry System. J. Sci. Transl. Med. 2017, 9, eaan3968. (14) Pirro, V.; Llor, R. S.; Jarmusch, A. K.; Alfaro, C. M.; CohenGadol, A. A.; Hattab, E. M.; Cooks, R. G. Analysis of Human Gliomas by Swab Touch Spray-Mass Spectrometry: Applications to Intraoperative Assessment of Surgical Margins and Presence of Oncometabolites. Analyst 2017, 142, 4058-4066. (15) Goryńska, P. Z.; Chmara, K.; Goryński, K.; Paczkowski, D.; Harat, M.; Bojko, B. A New Strategy for Brain Tumour Metabolomic Analysis. Med. Res. J. 2018, 3, 15-22. (16) Chen, X.; Wang, T.; Lin, L.; Wo, F.; Liu, Y.; Liang, X.; Ye, H.; Wu, J. Tip-Enhanced Photoinduced Electron Transfer and Ionization on Vertical Silicon Nanowires. ACS Appl. Mater. Interfaces 2018, 10, 14389-14398. (17) Lee, S.; Mardinoglu, A.; Zhang, C.; Lee, D.; Nielsen, J. Dysregulated Signaling Hubs of Liver Lipid Metabolism Reveal Hepatocellular Carcinoma Pathogenesis. Nucleic Acids Res. 2016, 44, 5529-5539. (18) Knochenmuss, R. Ion Formation Mechanisms in UV-MALDI. Analyst 2006, 131, 966-986. (19) Ritorto, M. S.; Ewan, R.; Perez-Oliva, A. B.; Knebel, A.; Buhrlage, S. J.; Wightman, M.; Kelly, S. M.; Wood, N. T.; Virdee, S.; Gray, N. S.; Morrice, N. A.; Alessi, D. R.; Trost, M. Screening of DUB Activity and Specificity by MALDI-TOF Mass Spectrometry. Nat. Commun. 2014, 5, 4763. (20) Wang, S. Y.; Kuo, C. H.; Tseng, Y. J. Batch Normalizer: A Fast Total Abundance Regression Calibration Method to Simultaneously Adjust Batch and Injection Order Effects in Liquid Chromatography/Time-of-Flight Mass Spectrometry-based Metabolomics Data and Comparison with Current Calibration Methods. Anal. Chem. 2013, 85, 1037-1046. (21) Chaudry, A. A.; Wahle, K. W.; Mcclinton, S.; Moffat, L. E. Arachidonic Acid Metabolism in Benign and Malignant Prostatic
ACS Paragon Plus Environment
4
Page 5 of 6 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry Tissue In vitro: Effects of Fatty Acids and Cyclooxygenase Inhibitors. Int. J. Cancer 1994, 57, 176-180. (22) Zhang, Y.; Qiu, L.; Wang, Y.; Qin, X.; Li, Z. High-Throughput and High-Sensitivity Quantitative Analysis of Serum Unsaturated Fatty Acids by Chip-based Nanoelectrospray Ionization-Fourier Transform Ion Cyclotron Resonance Mass Spectrometry: Early Stage Diagnostic Biomarkers of Pancreatic Cancer. Analyst 2014, 139, 1697-1706.
(23) Peck, B.; Schug, Z. T.; Zhang, Q.; Dankworth, B.; Jones, D. T.; Smethurst, E.; Patel, R.; Mason, S.; Jiang, M.; Saunders, R.; Howell, M.; Mitter, R.; Spencer-Dene, B.; Stamp, G.; McGarry, L.; James, D.; Shanks, E.; Aboagye, E. O.; Critchlow, S. E.; Leung, H. Y.; Harris, A. L.; Wakelam, M. J. O.; Gottlieb, E.; Schulze, A. Inhibition of Fatty Acid Desaturation is Detrimental to Cancer Cell Survival in Metabolically Compromised Environments. Cancer Metabolism 2016, 4, 6.
ACS Paragon Plus Environment
5
Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 6
TOC Figure
ACS Paragon Plus Environment
6
