This website is intended for healthcare professionals only.

Hospital Healthcare Europe
Hospital Pharmacy Europe     Newsletter    Login            

MALDI-MSI: a powerful tool in cancer research

Mass spectrometry imaging is a label-free method that allows the simultaneous investigations of hundreds of molecules in tissue

Alice Ly PhD
Michaela Aichler PhD
Axel Walch, MD PhD
Research Unit Analytical Pathology, Institute of Pathology, Helmholtz Zentrum München, German Research Center for Environmental Health (GmbH), Neuherberg, Germany
Mass spectrometry has emerged as an influential analytical tool in clinical research. The ability to simultaneously detect thousands of molecules within a single sample offers many advantages over traditional techniques that require labelling, such as immunoassays. In addition, different mass spectrometry technologies allow for the detection of molecules which may not be easily labelled, for example lipids, nucleotides, and other low molecular weight compounds. 
However, a major drawback of classical mass spectrometry is that the method largely requires analytes to be in a gas or liquid state, resulting in the loss of information regarding the spatial distribution of the detected molecules. This is particularly important given that it is not unexpected that a single tissue specimen would contain a large degree of morphological, cellular, and molecular heterogeneity. As such, microscopy is still required to determine the cellular origin of molecules detected by mass spectrometry. 
Matrix-Assisted Laser Desorption/Ionisation Mass Spectrometry Imaging (MALDI-MSI) is a technique that combines mass spectrometry with classical histology, resulting in a new quality of data in biochemical research and diagnostics. As an unbiased and label-free technique, MALDI-MSI has proven to be particularly useful in clinical studies, such as being used to discover new predictive and prognostic markers in cancer research.1,2
This review will first outline sample preparation steps and different MALDI-MSI platforms and as these directly influence the types of molecules that can be detected. Secondly, we will outline the application of MALDI-MSI in clinical research, before finally addressing future directions. 
MALDI imaging sample preparation and technological platforms
MALDI-MSI is capable of measuring a wide range of analytes, for example proteins, peptides, lipids and drugs, though tissue preparation plays a role in what can be detected in a given sample. The application has primarily been conducted using un-fixed frozen samples however protocols exist for the preparation of formalin-fixed, paraffin-embedded (FFPE) tissue for MALDI-MSI.3 The primary difficulty in measuring FFPE specimens is that the methylene cross-bridges which preserve the tissue need to be broken prior to measurement. 
Additionally, tissue processing steps for paraffin embedding, such as washing in solvents for example, ethanol, result in the elution of molecules such as lipids, while improper removal of paraffin prior to measurement can result in artefacts. However, given that there are many tissue archives with corresponding clinical information such as patient treatment and response, the measurement of these samples present an unparalleled potential resource for the discovery of new biomarkers. 
Tissue sections for MALDI-MSI are mounted onto glass slides which have a layer of indium tin oxide to render them electrically conductive. The section is then coated in a matrix; generally an organic acid which absorbs UV laser energy and enables molecular ionisation. Different matrices are suitable for the ionisation of particular molecules than others. Commonly used matrices include: 9-aminoacridine (9-AA) for metabolites and other small molecules;4 α-cyano-4-hydroxycinnamic acid (CHCA) for peptides and small proteins;5 and sinapinic acid for larger proteins.1 For a review of MALDI-MSI sample preparation methods, please refer to Norris and Caprioli.6
Following preparation, the sample is measured a predefined raster, with a mass spectrum generated at each measuring spot. Depending on the section size, this can produce several thousand spatially-correlated spectra. MALDI-MSI operates on the same principles of standard mass spectrometry, in that molecules are separated and identified based on their mass-to-charge ratio (m/z). Different mass analysers vary according to measurable mass range, mass resolution and accuracy, which can determine the type of molecule that can be detected (see Table 1). 
The most frequently used analyser for protein measurements is Time of Flight (TOF), while tandem TOF/TOF or Fourier-Transform Ion Cyclotron Resonance (FT-ICR) devices are more suitable for the separation of low molecular weight substances such as peptides, lipids, drugs, and metabolites. 
For a more detailed overview of the available analysers, please refer to the review of Pol et al.7 Solvents are used to remove the matrix from the section after measurement which is then histologically stained, typically with haematoxylin and eosin but immunohistochemical staining is also possible. A digital image of the stained section and the acquired mass spectra are then superimposed, allowing for the direct visualisation of the measurement on the sample, and thus the spatial distribution of different masses within the sample (see Figure 1). 
Fig. 1: MALDI imaging is able to analyse small tissue samples from patients, for example endoscopic biopsies. The subsequent unlabelled, histology-driven analysis allows the extraction of spatially resolved cell type-specific molecular signatures from a wide variety of classes. These can then be correlated with clinical endpoints that may therefore directly support the clinician in relevant areas such as diagnostics, response prediction, or disease outcome prediction. Used with permission from Elsevier.
Correlation of the measurement to histology also allows the extraction of spectra generated from ‘regions of interest’, for example, the spectra from tumour or healthy tissue only. This can then undergo further bioinformatics processing for patterns, such as clustering of similar spectra. 
Applications of MALDI-MSI
The identification of biomarkers that may act as predictive or prognostic factors, the stratification of patients into appropriate treatment groups, and prediction of response to treatment are all highly important for maximising survival rates of cancer patients. In each of these respects, MALDI-MSI has proven to be a particularly useful tool. 
A study conducted on resected gastric cancer tumours found a seven protein signature that correlated with unfavourable overall survival rates. Of these, three proteins S100-A6, HNP-1, and CRIP1 were identified and independently confirmed with immunohistochemistry. CRIP1 had previously not been associated with gastric cancer and was therefore identified as a new prognostic marker. A recent study utilised MALDI-MSI to examine differences between patients with Barrett’s carcinoma who responded to neoadjuvant treatment with cisplatin versus non-responder patients.2
Patients with a favourable clinical response had different proteomic signatures compared to those who did not respond. More specifically, MALDI-MSI revealed that COX7A2 could discriminate between responder and non-responder patients, and low COX7A2 protein expression was validated as a predictor of favourable response to cisplatin treatment. Having identified COX7A2 as a marker of response in Barett’s carcinoma, the protein was then also found to be dysregulated in other types of solid tumours. These two studies demonstrate the potential usefulness of MALDI-MSI in protein identification, patient stratification and response prediction. 
In addition to proteins, MALDI-MSI has been used to detect post-translational modifications, such as histone modifications in hepatocellular carcinoma.8 This study found modifications of histone H4 were more strongly found in tumour samples with microvascular invasion, and suggested that preoperative detection of these modifications may improve patient management. 
In addition to studies looking for predictive and prognostic markers, MALDI-MSI may be helpful in diagnosis. Two cancers with poor prognosis and high mortality rates are gastric cancer and hepatocellular cancer, both of which also lack early markers of malignancy. A MALDI-MSI study of gastric biopsies was able to discriminate between healthy and cancerous specimens with high accuracy.9 In addition, the authors identified differences in the protein profile that could distinguish between pathologic stage I (pT1N0M0) and more advanced tumours. 
A recent study examined the progression from cirrhosis to hepatocellular cancer (HCC).10 After first conducting MALDI-MSI on cirrhosis samples to identify patients at a high vs. low risk of developing HCC, the authors found that progression from cirrhosis to HCC correlated with increased expression of truncated monomeric ubiquitin. 
As previously discussed, measurement of archived FFPE tumour tissue is an emerging area in MALDI-MSI, as such the use of FFPE samples in clinical research is limited. A recent study used MALDI-MSI to measure FFPE breast and pancreatic cancer samples, and found that it was not only possible to discriminate between breast and pancreatic cancer specimens based on their spectral profile, but also that measurements of liver metastases could accurately determine the tumour of origin.11
Another MALDI-MSI study of FFPE tissue microarray of pancreatic tumours and principal component analysis – discriminant analysis (PCA-DA) found that the peptide profiles of pancreatic tumours could be classified based on their stages.12 More interestingly, the same study found that PCA-DA also discriminated additional stages which had not been assigned a pathological class. This finding highlights the possibility that MALDI-MSI would expand the understanding and classification of tumours beyond histological and morphological appearance. 
Future developments
The majority of MSI studies have concentrated on protein or peptide measurements due to the focus on finding new tumour prognostic markers but also from limitations on the mass range and sensitivity of what could be detected by MALDI-MSI devices. The development of high mass resolution instruments such as MALDI-FTICR has opened drug and metabolite imaging studies as a key area of development for MALDI-MSI. 
While drugs can be labelled and imaged via other methods such as positron emission tomography (PET), it is unknown how the labelling affects physical properties of the compound – issues that are important for understanding how the drug is metabolised and whether the drug has truly reached the target tissue. Studies of these issues have been examined in mouse models using MALDI-FTICR. 
Combined MSI-immunohistochemistry was used to examined the distribution of orally-administered afatinib, erlotinib and sorafenib in relation to the vascularisation of a squamous cell carcinoma xenograph.13 Co-registration of the MALDI signal to labelled blood vessels showed a relationship between drug signals and level of vascularisation and vessel size. 
Two separate studies have recently examined drug metabolite distributions in tissues. One study using colon adenocarcinoma-inoculated mice imaged the anti-angiogenic drug sunitinib and its metabolites were imaged the tumour, liver and kidney. 14 The other study examined the distribution of injected irinotecan and its active metabolite SN-38 in the organs of healthy mice and in two mouse models of colorectal cancer (Figure 2).15
Fig. 2: Different distributions of irinotecan and SN-38 in transversal whole-body tissue sections from mouse one hour post irinotecan injection. Panels showing H&E stains and MALDI-FT-ICR MS images of irinotecan (yellow: m/z 587.286 ±0.0002 Da) and SN-38 (blue: m/z 393.144 ± 0.0002 Da) indicate differences in the distributions of the molecules among various organs. Scale bar = 2mm.
However, this study ultimately found that while both the parent drug and metabolite could be detected in the gastrointestinal tract, only irinotecan could be detected in the tumour itself, implying that either the metabolite was not present in the tumour or in such minute levels as to be undetectable. While the above given examples were animal studies, a 2012 study examined inhaled ipratropium uptake in the airway walls of COPD patients,16 indicating that such an approach can be taken from the basic research to the clinical research spheres. 
Even more promisingly, two recent studies demonstrated imaging of metabolites in FFPE samples using MALDI-MSI.17, 18 The measurement of endogenous metabolites from a variety of archived cancer patient tissues using MALDI-FTICR-MSI was first reported in 2015. The authors were not only able to distinguish between normal and tumour tissues but also discriminate between two different renal tumours.17 Interestingly, a novel independent prognostic factor for disease-free survival for oesophageal adenocarcinoma patients was also identified.17 A few months later, a separate group demonstrated MALDI-MSI could be used to detect drugs and drug metabolites in FFPE rabbit tissues.18
A label-free multiplexed platform that allows for the detection of hundreds to thousands of molecules from a single sample, MALDI-MSI has been used to detect new tumour biomarkers as well as increasing understanding of patient response. Continuing improvements in technology have also led to the investigation of drug distribution and metabolism, while continuing optimisation of sample preparation methods will allow the measurement of archived tumour samples. In all, MALDI-MSI is proving to be an incomparable tool for tissue-based research of tumour biology. 
The authors would like to thank the support of the SYS-Stomach consortium and Ministry of Education and Research of the Federal Republic of Germany (BMBF) grant No. 01ZX1310B.
  1. Balluff B et al. MALDI imaging identifies prognostic seven-protein signature of novel tissue markers in intestinal-type gastric cancer. Am J Pathol 2011;179(6):2720–9.
  2. Aichler M et al. Clinical response to chemotherapy in oesophageal adenocarcinoma patients is linked to defects in mitochondria. J Pathol 2013;230(4):410–9.
  3. Casadonte R, Caprioli RM. Proteomic analysis of formalin-fixed paraffin-embedded tissue by MALDI imaging mass spectrometry. Nat Protoc 2011;6(11):1695–1709.
  4. Miura D et al. Ultrahighly sensitive in situ metabolomic imaging for visualizing spatiotemporal metabolic behaviors. Anal Chem 2010;82(23):9789–96.
  5. Cohen SL, Chait BT. Influence of matrix solution conditions on the MALDI-MS analysis of peptides and proteins. Anal Chem 1996;68(1):31–7.
  6. Norris JL, Caprioli RM. Analysis of tissue specimens by matrix-assisted laser desorption/ionization imaging mass spectrometry in biological and clinical research. Chem Rev 2013;113(4):2309–42.
  7. Pol J et al. Molecular mass spectrometry imaging in biomedical and life science research. Histochem Cell Biol 2010;134(5):423–43.
  8. Pote N et al. Imaging mass spectrometry reveals modified forms of histone H4 as new biomarkers of microvascular invasion in hepatocellular carcinomas. Hepatology 2013;58(3):983–94.
  9. Kim HK et al. Gastric cancer-specific protein profile identified using endoscopic biopsy samples via MALDI mass spectrometry. J Proteome Res 2010;9(8):4123–30.
  10. Laouirem S et al. Progression from cirrhosis to cancer is associated with early ubiquitin post-translational modifications: identification of new biomarkers of cirrhosis at risk of malignancy. J Pathol 2014;234:452–63.
  11. Casadonte R et al. Imaging mass spectrometry to discriminate breast from pancreatic cancer metastasis in formalin-fixed paraffin-embedded tissues. Proteomics 2014;14(7-8):956–64.
  12. Djidja MC et al. Novel molecular tumour classification using MALDI-mass spectrometry imaging of tissue micro-array. Anal Bioanal Chem 2010;397(2):587–601.
  13. Huber K et al. Novel Approach of MALDI Drug Imaging, Immunohistochemistry, and Digital Image Analysis for Drug Distribution Studies in Tissues. Anal Chem 2014;86(21):10568–75.
  14. Torok S et al. Localization of sunitinib, its metabolites and its target receptors in tumour bearing mice: a MALDI mass spectrometry imaging study. Br J Pharmacol 2014;172(4):1148–63.
  15. Buck A et al. Distribution and quantification of irinotecan and its active metabolite SN-38 in colon cancer murine model systems using MALDI MSI. Anal Bioanal Chem 2015;407(8):2107–16.
  16. Marko-Varga G et al. Understanding drug uptake and binding within targeted disease micro-environments in patients: a new tool for translational medicine. Clin Transl Med 2012;1(1):8.
  17. Buck A et al. High-resolution MALDI-FT-ICR MS imaging for the analysis of metabolites from formalin-fixed, paraffin-embedded clinical tissue samples. J Pathol 2015;237(1):123–32.
  18. Bruinen AL et al. Mass Spectrometry Imaging of Drug Related Crystal-Like Structures in Formalin-Fixed Frozen and Paraffin-Embedded Rabbit Kidney Tissue Sections. J Am Soc Mass 2016;27(1):117–23.