Over the past two decades, medicine has been moving towards data-driven diagnoses and personalized therapies based on molecular information.1 Oncology is a prime example, with an ever-growing range of precision drugs designed to target specific driver mutations in cancer cells. However, while these drugs have provided some benefits, they have failed to transform survival in most tumor types.
Frustratingly, identification of a particular genetic alteration is not enough to guarantee that a treatment will work. For example, drugs targeting the overactive V600E form of the signaling protein BRAF are effective in melanoma but have no activity in colorectal tumors with exactly the same mutation.
Attempting to treat disease based on a catalogue of genetic mutations is therefore likely to be an overly simplistic approach.
Genes don’t tell the whole story – proteomics can fill in the gaps
DNA sequencing is now rapid, reliable, low cost, and widely available, enabling rapid acquisition of large amounts of genetic and genomic information. Next-generation sequencing has been used in a range of clinical trials and has recently been approved for routine use by the US Food and Drug Administration (FDA).2
However, in practice, genes don’t always tell the whole story of what goes on in a cell or tissue at a molecular level (phenotype). Epigenetic regulation of gene activity and expression may alter the molecular pathways within a cell, resulting in dysfunction and disease even in the absence of underlying genetic alterations. As a result, genetic analysis is at best a proxy for the underlying molecular workings of a cell and may be misleading when it comes to selecting the best therapy.
Transcriptomics – sequencing of the RNA produced within a tissue or cell – is one way to estimate how genes are expressed, providing a more detailed insight into which cellular processes are dysfunctional in disease and aiding therapy selection.
Although transcriptomics provides a more realistic description of phenotype than genomic or genetic analysis, particularly while in a steady state. However, it is much less accurate during times of transition and turbulence – the correlation between transcription and translation is much weaker and RNA content alone is insufficient to predict protein abundance in many situations. For example, RNA may not be translated into a protein, and proteins can be modified after transcription in ways that can be unpredictable.3
As a result, there is often a mismatch between the predicted molecular profile of a cell based on RNA sequencing and its actual phenotype. Analyzing the protein content of cells (proteomics) would be more relevant for patient phenotyping and precision medicine. Yet protein analysis has historically been perceived as more technically challenging than DNA or RNA sequencing and requiring a large amount of starting material.
Proteomics reveals what’s going on inside
Faulty or dysregulated proteins lie at the heart of most disease pathways and are the usual target of precision therapies. Furthermore, there is a growing realization that the molecular chaos within a disease like cancer emerges as a result of complex dysregulated pathways and cannot be pinned on a handful of specific gene faults. Studying the entire set of proteins produced by a cell can therefore give a broad overview of the cell function and disease phenotype in a way that genomic analysis cannot.
Proteomics presents a significant challenge for analytical scientists: proteomes are highly complex, containing thousands of proteins, and it is easy to miss molecules that are present in a cell at low levels. However, researchers predict that decoding the proteome will impact the life sciences and clinical practice even more significantly than the genome revolution, so the challenge is worthwhile.
In recent years, mass spectrometry has emerged as the analytical method of choice for proteomics, due to its high sensitivity and specificity, providing reliable, high-throughput identification and quantification of proteins in biological samples.
Ultimately, proteomics could be used routinely to complement genomics and transcriptomics data in research and development, diagnosis, and treatment selection. But there are significant challenges that are only now being overcome.4,5,6
Shotgun proteomics can be a shot in the dark
The most widely used mass spectrometry technique is shotgun proteomics. This is usually based on a method called data-dependent acquisition (DDA), which samples subsets of hundreds of thousands of individual peptide fragments within a complex tissue such as a tumor.7
DDA selects these subsets randomly, so many rounds of analysis are required to ensure all the proteins present in a sample are detected and analyzed. Each experiment produces incomplete results and isn’t reproducible. This makes it difficult to accurately quantify the abundance of any particular protein.
These challenges have given mass spectrometry a reputation as a non-quantitative, low-throughput technique, preventing proteomics from finding clinical applications beyond basic research.8
Next-generation proteomics for clinical applications
An alternative approach – data-independent acquisition, or DIA – solves these problems. DIA follows a similar method to traditional shotgun proteomics, but every peptide in the sample is fragmented and analyzed in a single experiment in a parallel manner.
This ‘one shot’ method is much more attractive from the perspective of clinical applications as it generates comprehensive, reproducible and quantitative proteomic profiles with high throughput. However, the datasets produced through DIA are extremely complex and need to be separated out (deconvoluted) in order to understand the identities and abundance of all the proteins that are present in the original tissue.
Novel algorithms have been recently developed that solve this deconvolution challenge. This has allowed the generation of DIA results that are unbiased, more complete (almost 10,000 proteins in a single shot), precisely quantifiable and reproducible.5-7
Figure 1: Data Independent Acquisition (DIA) discovery proteomics workflow. Image credit: Biognosys
Proteomics on trial
In April 2019, Roche presented results from a simulated study mimicking the implementation of discovery proteomics in a clinical trial. The team used HRM to analyze 30 representative colorectal cancer samples, successfully identifying and quantifying more than 9,000 proteins with high reproducibility. Importantly, the amount of tumor material required for proteomic analysis was comparable to that needed for transcriptomics, providing the foundations for the use of discovery proteomics in a clinical trial setting.8-10
Proteomics will drive the shift to personalized medicine
Proteomics is considerably more complex than DNA and RNA sequencing but provides a more accurate reflection of the molecular state of cells and tissues. Proteomics using DIA approaches such as HRM is now the closest approximation of protein expression profiling available to clinical and discovery scientists.
Right now, clinical researchers can use these techniques for discovery proteomics in patient phenotyping and stratification, biomarker research, drug and target discovery, pathway modeling, and mechanisms of action studies. For researchers without access to equipment or the expertise required to bring proteomics into their work, Biognosys offers reliable protein profiling that meets good clinical practice criteria for clinical applications.11
The field of proteomics has matured significantly over the past decade and advances such as DIA and HRM are finally overcoming the hurdles of quantitative precision, reproducibility and scalability and enabling the routine use of high throughput mass spectrometry in clinical applications.
In the future, it is expected that every clinical sample will be routinely protein-profiled as part of the diagnosis and therapy selection process.5-7 Bringing proteomics to clinical trials and practice will improve disease phenotyping, diagnostics and effective treatment selection, ultimately increasing the efficiency of the drug discovery pipeline and bringing benefits to patients.
- Doostparast Torshiz, A., & Wang, K. (2018) Next-generation sequencing in drug development: target identification and genetically stratified clinical trials. Drug Discov. Today. 23(10), 1776–1783 doi: 10.1016/j.drudis.2018.05.015
- Siu, L.L., Conley, B.A., Boerner, S., & LoRusso, M.P. (2015) Next-Generation Sequencing to Guide Clinical Trials. Clin Cancer Res. 21(20), 4536–4544 doi: 10.1158/1078-0432.CCR-14-3215
- Liu, Y., Beyer, A., and Aebersold, R. (2016) On the Dependency of Cellular Protein Levels on mRNA Abundance. Cell. 165 (3) 535-550 doi: 10.1016/j.cell.2016.03.014
- Chen, R., & Snyder, M. (2013) Promise of Personalized Omics to Precision Medicine. Wiley Interdiscip Rev Syst Biol Med. 5(1), 73–82 doi: 10.1002/wsbm.1198
- Duarte, T.T., & Spencer, C.T. (2016) Personalized Proteomics: The Future of Precision Medicine. Proteomes. 4(4), 29 doi: 10.3390/proteomes4040029
- Rinner, O. (2016) Next-generation Proteomics from an Industrial Perspective. Chimia. 70(12), 860–863 doi: 10.2533/chimia.2016.860
- How does DIA differ from classic shotgun proteomics and DDA? (2018, March 15) Retrieved from https://help.biognosys.com/help/how-does-dia-differ-from-classical-shotgun-proteomics
- Schubert, O.T., Röst, H.L., Collins, B.C., Rosenberger, G., & Aebersold, R. (2017) Quantitative proteomics: challenges and opportunities in basic and applied research. Nat. Protoc. 12(7), 1289–1294 doi: 10.1038/nprot.2017.040
- Roche Presents Data from Biognosys for Use of Discovery Proteomics in Clinical Trials (2019, April 4) Retrieved from https://www.biognosys.com/press-release-roche-presents-data-from-biognosys-for-use-of-discovery-proteomics-in-clinical-trials
- Ducret A. (2019) Clinical Proteomics Enters Clinical Trials: A Generic GCP-compliant Workflow for the Routine Analysis of FFPE Tissue Samples by Quantitative Mass Spectrometry. Presented at the Annual Congress in Clinical Mass Spectrometry (US), Palm Springs, CA.
- Roche Partners with Biognosys, Caprion to Demonstrate Proteomics Readiness for Clinical Trials (2019, April 23) Retrieved from https://www.genomeweb.com/proteomics-protein-research/roche-partners-biognosys-caprion-demonstrate-proteomics-readiness#.XOlJRi3Mw_W