Despite significant advancements in medical science, cancer remains one of the most challenging diseases to diagnose and treat effectively. This challenge comes from a complex, multimodal molecular disease pathology that starts with genomic alterations that cascade to changes in cellular functions, in microenvironments, and in the interactions between cells, most of which are mediated by proteins and their functionally relevant post-translational modifications.
A promising frontier in cancer research therefore lies within the field of mass spectrometry (MS)-based proteomics (Figure 1). The technology enables the comprehensive study of biochemical processes from tissues down to single cells and even individual organelles in humans and model organisms. Recent innovations in proteomics techniques are enabling researchers to uncover new avenues for improving cancer diagnosis and treatment.
Resolving cancer biology through spatial proteomics
Spatial biology offers insights into intricate tissue architectures, local cellular biochemistry, and intercellular interactions. In single-cell next-generation sequencing, spatial resolution has been successfully used to investigate tumor microenvironments identifying phenotypically distinct and disease-relevant tumor types and immune states.2,3 Proteomics has matured to a point where over 10,000 proteins can be analyzed from bulk cellular material, and increased sensitivity and throughput has enabled researchers to resolve proteomics signatures across tissues4,5 and down to the single-cell level,6,7 and to elucidate the highly dynamic spatial context of proteins. This knowledge empowers researchers to comprehend multifaceted biological phenomena and decipher disease progression to pinpoint disease biomarkers and devise therapeutic interventions and personalized medicine approaches.
In the context of cancer, for example, it is particularly interesting to establish a functionally relevant read on individual cells in the tumor microenvironment and immune status. For instance, cytokines and chemokines may form gradients to attract or repel cells of the immune system, ultimately contributing to clinically relevant immune infiltration phenotypes. Moreover, profiling of extracellular vesicles in tumor microenvironments may reveal biomarkers and resolve cancer-immune interactions such as secretion of vesicles with cancer signatures to distract the immune response and contribute to immune evasion.
Robust and scalable proteomics workflows amenable to ultra-low-input material have shown broad implications in biomedicine, including early disease profiling and drug target and biomarker discovery, particularly around new modalities like T-cell engagers and CAR T cells.8 Increased sensitivity of liquid chromatography (LC)-MS workflows, enabled by highly efficient sampling of biomolecules through technological advances like trapped ion mobility spectrometry (TIMS),9 facilitate comprehensive and quantitative surveys that account for the smallest amounts of biological material and provide single-cell resolution.10,11
One emerging application is deep visual proteomics (DVP), which integrates artificial intelligence–guided image analysis of cell characteristics with automated laser microdissection of individual cells or nuclei, followed by LC-MS. In archived primary melanoma tissue, DVP can detect localized changes in protein composition as normal melanocytes transform into invasive melanoma. These changes reveal spatially evolving pathways during cancer progression, such as irregular mRNA splicing in advancing metastatic growth, accompanied by decreased interferon signaling and antigen presentation.
DVP’s capacity to maintain accurate spatial protein data within tissue samples has significant implications for molecular analysis in clinical settings.5 These highly sensitive workflows empower researchers to dissect tumor microenvironment from small amounts of clinical material such as fine-needle biopsies, and are also required to investigate other aspects of cancer biology such as blood-circulating cancer cells or secreted proteins and vesicles.
Investigating cancer-immune interactions through immunopeptidomics
The DNA in every cell in our body is estimated to undergo thousands of mutations per day.12 Although most of these mutations are efficiently corrected through an array of machineries, cells beyond repair can undergo suicide (apoptosis). Another line of defense is our immune system, which is trained to distinguish between normal cells and harmful cells, and to eliminate the latter, which include pathogen-infected cells and cancer cells.
One key principle is the recognition of self and foreign proteins based on fragments (immunopeptides) presented on the major histocompatibility complex (MHC) molecules on the cellular surface. This peptide presentation–recognition mechanism empowers T lymphocytes (T cells) to discern foreign patterns, triggering an immune response. Cancer cells have not only acquired mutations that drive the malignancy but are also characterized by increased genomic and proteomic instability, leading to presentation of novel peptides on the cell surface. Although the mutational landscape can be mapped with traditional sequencing technologies, peptides that are ultimately presented on the surface of cancer cells can only be determined directly by MS-based proteomics.
Recognizing this, the Human Immunopeptidome Project (HIPP) aims to map the entire human immunopeptidome and to help make immunopeptidomics widely accessible in clinics. Importantly, using quantitative immunopeptidomics on a large scale could be key in cancer immunotherapies, and also reveal individual tendencies toward common immune diseases and the response to new modalities like vaccines.13
Researchers have developed a high-throughput, sensitive, single-shot MS-based immunopeptidomics workflow that leverages TIMS time-of-flight (TOF) technology to identify the MHC-I and MHC-II tumor immunopeptidome and inform the development of cancer immunotherapies.14 The fast, sensitive, and accurate detection of MHC-presented peptides enabled creation of a databases containing over 150,000 immune peptides, including known and novel tumor cancer neoantigens. Based on this comprehensive data, MS-based de novo peptide sequencing algorithms like Novor 2.0 can detect neoantigens even from sources such as plasma and extracellular vesicles for which no direct genomics or transcriptomics information is available. Such newly identified antigens include rare mutation-derived neoantigens, which could serve as biomarkers or targets for future cancer immunotherapy.15
Discovering new cancer biology and biomarkers from a blood draw
Blood plasma is an easily accessible biospecimen routinely sampled from blood draws. Also, there are more than 100 FDA-cleared or FDA-approved clinical plasma or serum tests for assessing protein- or protein modification–based biomarkers. Consequently, plasma can serve as a rich source for early diagnosis.16 As blood connects to almost all tissues directly, blood plasma holds immense promise for minimally invasive cancer diagnostics by identifying specific protein and peptide signatures associated with different cancer types and stages.
One key challenge of plasma is the extraordinarily large dynamic range of the proteome with a minority of the proteins contributing most of the protein mass. Innovations in sample preparation as well as increasingly efficient MS workflows have enabled researchers to study the plasma proteome at scale and depth.17–19 Scientists can now meticulously identify and measure proteins with remarkable sensitivity, as recently reported in a preprint about a large-cohort lung cancer study.20 Using multianalyte classifiers, this study demonstrated 89%, 80%, and 98–100% sensitivity for all-stage, stage I, and stage III–IV lung cancer, respectively, at 89% specificity in a validation set.
Once identified, protein biomarkers can be efficiently interrogated with targeted LC-MS methods such as parallel reaction monitoring parallel accumulation serial fragmentation (prm-PASEF) that focus on the proteins and peptides of interest,21 whereas recently introduced methods such as guided data independent acquisition PASEF (g-dia-PASEF) combine targeted and discovery workflows, enabling focused attention on known biomarkers and unbiased surveys of the remaining proteome.22
To run studies at a scale of thousands or tens of thousands of samples, highly robust and scalable workflows are required. For instance, in the previously mentioned lung cancer study, 2,500 samples in almost 10,000 LC-MS runs across four distinct instruments were analyzed, demonstrating that robust data can be acquired, and the results integrated, across multiple LC-MS systems.
As blood plasma comprises a mixed proteome consisting of secreted proteins, tissue leakage proteins, proteins coming from platelets, as well as potential contaminants from the collection,23 much promise lies in discerning and dissecting these subproteomes. Instead of processing plasma as bulk material including all soluble and insoluble components, future efforts may target specific components, for example, extracellular vesicles, which can carry distinct biological insights. Of potential interest is the de novo sequencing of immunopeptides from circulating extracellular vesicles combined with that of immunopeptides presented by the MHC to identify cancer neoantigens. The absence of sequencing information and the very low abundance of MHC peptides pose key challenges that can be addressed with advanced de novo sequencing and highly sensitive LC-MS workflows, opening new diagnostic opportunities.24
Identifying new drug candidates through chemoproteomics
All biochemical processes in cells involve proteins, and modulating the abundance and localization of these proteins is a promising therapeutic avenue. Recently, scalable strategies have emerged that tackle protein targets previously deemed “undruggable.” For example, protein targets implicated in cancer are candidates for degradation via new modalities with high therapeutic value, such as targeted protein degradation.25
TIMS, in conjunction with PASEF, provides information on the collisional cross-sectional shape of molecules26 and is particularly useful in activity-based protein profiling (ABPP). In ABPP, proteins are targeted by reactive molecules (warheads), resulting in distinct masses and unique shape properties that can be distinguished based on their changed cross-sectional shape using TIMS-TOF technology. To realize the huge potential of these kinds of studies, chemoproteomics has moved from academia to industry, where many companies have disclosed programs in preclinical and early clinical development.27
As for all large screening approaches, achieving optimal throughput and workflow robustness is essential to effectively screen large compound libraries. Another crucial aspect of this screening process is the accurate identification and sensitive quantification of proteins. This information allows compound-induced degradation and modification to be determined reliably. High-throughput proteomics can acquire complex proteomes every couple of minutes, facilitating the capture of biological phenotypes, the identification of new druggable pathways, and the delineation of therapeutic interventions in complex diseases such as cancer.28,29
Conclusion
As the proteomic analysis of the cancer landscape evolves, new modalities and opportunities for early detection, precise classification, and tailored therapeutic interventions emerge that are fueled by advancements in sensitive MS workflows such as the combination of TIMS with fast TOF mass analyzers. This enables advanced acquisition modes, like PASEF, and novel machine learning strategies that, for example, facilitate identification of cancer neoantigens (Novor 2.0) from the smallest clinical isolates. Altogether, LC-MS-based proteomics offers unprecedented and unique insights into the molecular mechanisms driving tumorigenesis and holds the potential to transform patient outcomes through precision medicine. By integrating other molecular modalities accessible with LC-MS such as post-translational modifications, lipids, and metabolites, researchers are provided with a unique and versatile toolbox to drive new breakthroughs in cancer diagnosis and treatment.
Daniel Hornburg, PhD, is vice president, biomarkers and precision medicine, Torsten Mueller, PhD, is business development manager, proteomics, and Stefan Foser, PhD, is vice president, global pharma, at Bruker Daltonics.
References
1. Shi R, Tang Y-Q, Miao H. Metabolism in tumor microenvironment: Implications for cancer immunotherapy. MedComm 2020; 1(1): 47–68. DOI: 10.1002/mco2.6.
2. Hornburg M, Desbois M, Lu S, et al. Single-cell dissection of cellular components and interactions shaping the tumor immune phenotypes in ovarian cancer. Cancer Cell 2021: 39(7): 928–944.e6. DOI: 10.1016/j.ccell.2021.04.004.
3. Mayer S, Milo T, Isaacson A, et al. The Tumor Microenvironment Shows a Hierarchy of Cell-Cell Interactions Dominated by Fibroblasts. Nat. Commun. 2023; 14(1): 5810. DOI: 10.1038/s41467-023-41518-w.
4. Tüshaus J, Sakhteman A, Lechner S, et al. A region-resolved proteomic map of the human brain enabled by high-throughput proteomics. EMBO J. 2023; 42(23): e114665. DOI: 10.15252/embj.2023114665.
5. Mund A, Coscia F, Kriston A, et al. Deep Visual Proteomics defines single-cell identity and heterogeneity. Nat. Biotechnol. 2022; 40(8): 1231–1240. DOI: 10.1038/s41587-022-01302-5.
6. Mayer RL, Mechtler K. Immunopeptidomics in the Era of Single-Cell Proteomics. Biology (Basel) 2023; 12(12): 1514. DOI: 10.3390/biology12121514.
7. Brunner A-D, Thielert M, Vasilopoulou C. Ultra-high sensitivity mass spectrometry quantifies single-cell proteome changes upon perturbation. Mol. Syst. Biol. 2022; 18(3): e10798. DOI: 10.15252/msb.202110798.
8. Makhmut A, Qin D, Fritzsche S, et al. A framework for ultra-low-input spatial tissue proteomics. Cell Syst. 2023; 14(11): 1002–1014.e5. DOI: 10.1016/j.cels.2023.10.003.
9. Vasilopoulou CG, Sulek K, Brunner A-D. Trapped ion mobility spectrometry and PASEF enable in-depth lipidomics from minimal sample amounts. Nat. Commun. 2020; 11(1): 331. DOI: 10.1038/s41467-019-14044-x.
10. Ctortecka C, Clark NM, Boyle B, et al. Automated single-cell proteomics providing sufficient proteome depth to study complex biology beyond cell type classifications. Nat. Commun. 2024; 15: 5707. DOI: 10.1038/s41467-024-49651-w
11. Leduc A, Koury L, Cantlon J, Slavov N. Massively parallel sample preparation for multiplexed single-cell proteomics using nPOP. bioRxiv 2023; DOI: 10.1101/2023.11.27.568927.
12. Chatterjee N, Walker GC. Mechanisms of DNA damage, repair, and mutagenesis. Environ. Mol. Mutagen. 2017; 58(5): 235–263. DOI: 10.1002/em.22087.
13. Vizcaíno JA, Kubiniok P, Kovalchik KA, et al. The Human Immunopeptidome Project: A Roadmap to Predict and Treat Immune Diseases. Mol. Cell. Proteomics 2020; 19(1): 31–49. DOI: 10.1074/mcp.R119.001743.
14. Phulphagar KM, Ctortecka C, Jacome ASV, et al. Sensitive, high-throughput HLA-I and HLA-II immunopeptidomics using parallel accumulation-serial fragmentation mass spectrometry. Mol. Cell. Proteomics 2023; 22(6): 100563. DOI: 10.1016/j.mcpro.2023.100563.
15. Hoenisch Gravel N, Nelde A, Bauer J, et al. TOFIMS mass spectrometry-based immunopeptidomics refines tumor antigen identification. Nat. Commun. 2023; 14: 7472. DOI: 10.1038/s41467-023-42692-7.
16. Geyer PE, Holdt LM, Teupser D, Mann M. Revisiting biomarker discovery by plasma proteomics. Mol. Syst. Biol. 2017, 13(9): 942. DOI: 10.15252/msb.20156297.
17. Ferdosi S, Stukalov A, Hasan M, et al. Enhanced Competition at the Nano-Bio Interface Enables Comprehensive Characterization of Protein Corona Dynamics and Deep Coverage of Proteomes. Adv. Mater. 2022; 34(44): e2206008. DOI: 10.1002/adma.202206008.
18. Viode A, van Zalm P, Smolen KK, et al. A simple, time- and cost-effective, high-throughput depletion strategy for deep plasma proteomics. Sci. Adv. 2023; 9(13): eadf9717. DOI: 10.1126/sciadv.adf9717.
19. Metatla I, Roger K, Chhuon C, et al. Neat plasma proteomics: getting the best out of the worst. Clin. Proteomics 2024; 21(1): 22. DOI: 10.1186/s12014-024-09477-6.
20. Koh B, Liu M, Almonte R, et al. Multi-omics profiling with untargeted proteomics for blood-based early detection of lung cancer. medRxiv 2024; DOI: 10.1101/2024.01.03.24300798.
21. Brzhozovskiy A, Kononikhin A, Bugrova AE, et al. The Parallel Reaction Monitoring-Parallel Accumulation-Serial Fragmentation (prm-PASEF) Approach for Multiplexed Absolute Quantitation of Proteins in Human Plasma. Anal. Chem. 2022; 94(4): 2016–2022. DOI: 10.1021/acs.analchem.1c03782.
22. Lesur A, Bernardin, F, Koncina E, et al. Quantification of 782 Plasma Peptides by Multiplexed Targeted Proteomics. J. Proteome Res. 2023; 22(6): 1630–1638. DOI: 10.1021/acs.jproteome.2c00575.
23. Geyer PE, Voytik E, Treit PV, et al. Plasma Proteome Profiling to detect and avoid sample-related biases in biomarker studies. EMBO Mol. Med. 2019; 11(11): e10427. DOI: 10.15252/emmm.201910427.
24. Wahle M, Thielert M, Zwiebel M, et al. IMBAS-MS Discovers Organ-Specific HLA Peptide Patterns in Plasma. Mol. Cell. Proteomics 2024; 23(1): 100689. DOI: 10.1016/j.mcpro.2023.100689.
25. Zhao L, Zhao J, Zhong K, et al. Targeted protein degradation: mechanisms, strategies and application. Signal Transduct. Target. Ther. 2022; 7: 113. DOI: 10.1038/s41392-022-00966-4.
26. Meier F, Köhler ND, Brunner A-D, et al. Deep learning the collisional cross sections of the peptide universe from a million experimental values. Nat. Commun. 2021; 12(1): 1185. DOI: 10.1038/s41467-021-21352-8.
27. Békés M, Langley DR, Crews CM. PROTAC targeted protein degraders: the past is prologue. Nat. Rev. Drug Discov. 2022; 21: 181–200. DOI: 10.1038/s41573-021-00371-6.
28. Messner CB, Demichev V, Wang Z, et al. Mass spectrometry-based high-throughput proteomics and its role in biomedical studies and systems biology. Proteomics 2023; 23(7-8): e2200013. DOI: 10.1002/pmic.202200013.
29. Kaspar-Schoenefeld S, Krieger JR, Martelli C, et al. High-throughput proteome profiling with low variation in a multi-center study using dia-PASEF. bioRxiv 2024: DOI: 10.1101/2024.05.29.596405.