Cancer Tissue Classification Using Supervised Machine Learning Applied to MALDI Mass Spectrometry Imaging

Overview

Journal Cancers (Basel)

Publisher MDPI

Specialty Oncology

Date 2021 Nov 13

PMID 34771551

Citations 8

Authors

Paul Mittal

Mark R Condina

Manuela Klingler-Hoffmann

Gurjeet Kaur

Martin K Oehler

Oliver M Sieber

Michelle Palmieri

Stefan Kommoss

Sara Brucker

Mark D McDonnell

Peter Hoffmann

Affiliations

Soon will be listed here.

Abstract

Matrix assisted laser desorption/ionization mass spectrometry imaging (MALDI MSI) can determine the spatial distribution of analytes such as protein distributions in a tissue section according to their mass-to-charge ratio. Here, we explored the clinical potential of machine learning (ML) applied to MALDI MSI data for cancer diagnostic classification using tissue microarrays (TMAs) on 302 colorectal (CRC) and 257 endometrial cancer (EC)) patients. ML based on deep neural networks discriminated colorectal tumour from normal tissue with an overall accuracy of 98% in balanced cross-validation (98.2% sensitivity and 98.6% specificity). Moreover, our machine learning approach predicted the presence of lymph node metastasis (LNM) for primary tumours of EC with an accuracy of 80% (90% sensitivity and 69% specificity). Our results demonstrate the capability of MALDI MSI for complementing classic histopathological examination for cancer diagnostic applications.

Citing Articles

Mass Spectrometry Imaging for Spatial Ingredient Classification in Plant-Based Food.

Vats M, Flinders B, Visvikis T, Dawid C, Hofmann T, Cuypers E J Am Soc Mass Spectrom. 2024; 36(1):100-107.

PMID: 39644241 PMC: 11697329. DOI: 10.1021/jasms.4c00353.

Spatially resolved metabolomics: From metabolite mapping to function visualising.

Min X, Zhao Y, Yu M, Zhang W, Jiang X, Guo K Clin Transl Med. 2024; 14(11):e70031.

PMID: 39456123 PMC: 11511672. DOI: 10.1002/ctm2.70031.

An Update on the Use of Artificial Intelligence in Digital Pathology for Oral Epithelial Dysplasia Research.

Alajaji S, Khoury Z, Jessri M, Sciubba J, Sultan A Head Neck Pathol. 2024; 18(1):38.

PMID: 38727841 PMC: 11087425. DOI: 10.1007/s12105-024-01643-4.

Classification of Pancreatic Ductal Adenocarcinoma Using MALDI Mass Spectrometry Imaging Combined with Neural Networks.

Kanter F, Lellmann J, Thiele H, Kalloger S, Schaeffer D, Wellmann A Cancers (Basel). 2023; 15(3).

PMID: 36765644 PMC: 9913229. DOI: 10.3390/cancers15030686.

Applications of spatially resolved omics in the field of endocrine tumors.

Hou Y, Gao Y, Guo S, Zhang Z, Chen R, Zhang X Front Endocrinol (Lausanne). 2023; 13:993081.

PMID: 36704039 PMC: 9873308. DOI: 10.3389/fendo.2022.993081.

References

Klein O, Kanter F, Kulbe H, Jank P, Denkert C, Nebrich G . MALDI-Imaging for Classification of Epithelial Ovarian Cancer Histotypes from a Tissue Microarray Using Machine Learning Methods. Proteomics Clin Appl. 2018; 13(1):e1700181. DOI: 10.1002/prca.201700181. View

Deininger S, Cornett D, Paape R, Becker M, Pineau C, Rauser S . Normalization in MALDI-TOF imaging datasets of proteins: practical considerations. Anal Bioanal Chem. 2011; 401(1):167-81. PMC: 3124646. DOI: 10.1007/s00216-011-4929-z. View

Meding S, Nitsche U, Balluff B, Elsner M, Rauser S, Schone C . Tumor classification of six common cancer types based on proteomic profiling by MALDI imaging. J Proteome Res. 2012; 11(3):1996-2003. DOI: 10.1021/pr200784p. View

Schwamborn K, Kriegsmann M, Weichert W . MALDI imaging mass spectrometry - From bench to bedside. Biochim Biophys Acta Proteins Proteom. 2016; 1865(7):776-783. DOI: 10.1016/j.bbapap.2016.10.014. View

Mittal P, Klingler-Hoffmann M, Arentz G, Winderbaum L, Kaur G, Anderson L . Annexin A2 and alpha actinin 4 expression correlates with metastatic potential of primary endometrial cancer. Biochim Biophys Acta Proteins Proteom. 2016; 1865(7):846-857. DOI: 10.1016/j.bbapap.2016.10.010. View

Rauser S, Deininger S, Suckau D, Hofler H, Walch A . Approaching MALDI molecular imaging for clinical proteomic research: current state and fields of application. Expert Rev Proteomics. 2010; 7(6):927-41. DOI: 10.1586/epr.10.83. View

Verbeeck N, Caprioli R, Van de Plas R . Unsupervised machine learning for exploratory data analysis in imaging mass spectrometry. Mass Spectrom Rev. 2019; 39(3):245-291. PMC: 7187435. DOI: 10.1002/mas.21602. View

Perez-Riverol Y, Csordas A, Bai J, Bernal-Llinares M, Hewapathirana S, Kundu D . The PRIDE database and related tools and resources in 2019: improving support for quantification data. Nucleic Acids Res. 2018; 47(D1):D442-D450. PMC: 6323896. DOI: 10.1093/nar/gky1106. View

Nampei M, Horikawa M, Ishizu K, Yamazaki F, Yamada H, Kahyo T . Unsupervised machine learning using an imaging mass spectrometry dataset automatically reassembles grey and white matter. Sci Rep. 2019; 9(1):13213. PMC: 6744563. DOI: 10.1038/s41598-019-49819-1. View

10.

Arentz G, Mittal P, Zhang C, Ho Y, Briggs M, Winderbaum L . Applications of Mass Spectrometry Imaging to Cancer. Adv Cancer Res. 2017; 134:27-66. DOI: 10.1016/bs.acr.2016.11.002. View

11.

Behrmann J, Etmann C, Boskamp T, Casadonte R, Kriegsmann J, Maass P . Deep learning for tumor classification in imaging mass spectrometry. Bioinformatics. 2017; 34(7):1215-1223. DOI: 10.1093/bioinformatics/btx724. View

12.

Gustafsson J, Eddes J, Meding S, Koudelka T, Oehler M, McColl S . Internal calibrants allow high accuracy peptide matching between MALDI imaging MS and LC-MS/MS. J Proteomics. 2012; 75(16):5093-5105. DOI: 10.1016/j.jprot.2012.04.054. View

13.

Winderbaum L, Koch I, Mittal P, Hoffmann P . Classification of MALDI-MS imaging data of tissue microarrays using canonical correlation analysis-based variable selection. Proteomics. 2016; 16(11-12):1731-5. DOI: 10.1002/pmic.201500451. View

14.

Gustafsson O, Eddes J, Meding S, McColl S, Oehler M, Hoffmann P . Matrix-assisted laser desorption/ionization imaging protocol for in situ characterization of tryptic peptide identity and distribution in formalin-fixed tissue. Rapid Commun Mass Spectrom. 2013; 27(6):655-70. DOI: 10.1002/rcm.6488. View

15.

Zhang Y, Liu X . Machine learning techniques for mass spectrometry imaging data analysis and applications. Bioanalysis. 2018; 10(8):519-522. DOI: 10.4155/bio-2017-0281. View

16.

Mittal P, Klingler-Hoffmann M, Arentz G, Winderbaum L, Lokman N, Zhang C . Lymph node metastasis of primary endometrial cancers: Associated proteins revealed by MALDI imaging. Proteomics. 2016; 16(11-12):1793-801. DOI: 10.1002/pmic.201500455. View

17.

Casadonte R, Kriegsmann M, Zweynert F, Friedrich K, Baretton G, Bretton G . Imaging mass spectrometry to discriminate breast from pancreatic cancer metastasis in formalin-fixed paraffin-embedded tissues. Proteomics. 2014; 14(7-8):956-64. DOI: 10.1002/pmic.201300430. View

18.

Schramm T, Hester Z, Klinkert I, Both J, Heeren R, Brunelle A . imzML--a common data format for the flexible exchange and processing of mass spectrometry imaging data. J Proteomics. 2012; 75(16):5106-5110. DOI: 10.1016/j.jprot.2012.07.026. View

19.

Mittal P, Klingler-Hoffmann M, Arentz G, Zhang C, Kaur G, Oehler M . Proteomics of endometrial cancer diagnosis, treatment, and prognosis. Proteomics Clin Appl. 2015; 10(3):217-29. DOI: 10.1002/prca.201500055. View

20.

Boyle S, Mittal P, Kaur G, Hoffmann P, Samuel M, Klingler-Hoffmann M . Uncovering Tumor-Stroma Inter-relationships Using MALDI Mass Spectrometry Imaging. J Proteome Res. 2020; 19(10):4093-4103. DOI: 10.1021/acs.jproteome.0c00511. View