René P. Zahedi(1) and Christoph H.Borchers(1-4)
(1)Segal Cancer Proteomics Centre, Lady Davis Institute, Jewish General Hospital, McGill University, 3755 Cote-Ste-Catherine Road, Montréal, QC H3T 1E2, Canada
(2)Gerald Bronfman Department of Oncology, Jewish General Hospital, McGill University, 3755 Cote-Ste-Catherine Road, Montréal, QC H3T 1E2, Canada
(3)University of Victoria – Genome British Columbia Proteomics Centre, Vancouver Island Technology Park, #3101- 4464 Markham St., Victoria, BC V8Z 7X8, Canada
(4)University of Victoria, Department of Biochemistry and Microbiology, Petch Building Room 207, 3800 Finnerty Rd., Victoria, BC, V8P 5C2, Canada
In recent years, precision medicine has opened new avenues for the targeted treatment of cancer (or other diseases), using the best available therapy. Upon identification of somatic genetic alterations (point mutations, deletions, amplifications, etc.), specific drugs can be selected that allow the targeted treatment of a tumour . Genomics-based precision medicine has been successfully applied to heal cancer patients by ensuring that each patient receives the optimal treatment, but response rates are still often unexpectedly low. For some cancers, up to 50 % of patients may not respond to the predicted targeted drug treatment . A variety of reasons contribute to these modest success rates. One reason is that the genome defines only the potential template of an organism (its genotype) and is rather static, while biological processes are highly dynamic. This dynamic is controlled not only by those biomolecules that are encoded in the genome (i.e., proteins), but also by metabolites and lipids. The proteome is defined as the complement of all proteins expressed by an organism, tissue or cell at a given time point and condition. Its composition depends not only upon which genes are active (i.e. protein translation), but also protein degradation and post-translational modifications that alter the structure of the protein and therefore its function and activity. Thus, the proteome of an organism reflects of the actual state of the organism (its phenotype) more accurately than does the genome, and the proteome depends on a multitude of factors: environment, diet, age, gender physical activity, health status, medication, etc.
To understand the causes as well as the consequences of disease on a molecular level, it is essential to understand which genes are actually being actively translated into proteins, which are the real drivers of biology.
Indeed, genomics will fail to distinguish the tadpole from the frog. Although these two are very distinct stages (phenotypes) in the life cycle of the very same organism, they share an identical genome but can be clearly differentiated by their proteomes. In a similar way, in many cases, the molecular basis of an individual cancer cannot be completely understood by using only genomic approaches. Therefore, the combination of genomic data with quantitative proteomic data on both protein expression and post-translational protein modifications — a novel approach termed proteogenomics  — is now thought to be a very promising approach for improving precision medicine so that it better predicts individualized treatment for cancer patients. This unique combination will enable researchers and clinicians to address several fundamental questions:
- Which genes and mutations are actually active and are being translated into proteins in a tumour? Although a tumour cell contains more than 10,000 different proteins  and tumours are an agglomerate of multiple cell types (cancerous and healthy), not every predicted mutation is actually present or relevant.
- How many copies of each protein are really present in the tumour? Depending on the protein, the copy number can vary from a few to a several million molecules per cell and copy number is often changed in tumours compared to healthy cells or tissues. Such changes can indicate increased or reduced activity of central pathways, and these changes are not readily predictable from the genome.
- Are protein activities or signalling pathways altered? Dysregulated signalling (protein phosphorylation) is one of the main causes of cancer, and many drugs target tyrosine kinases . Protein phosphorylation levels and alterations, however, cannot be predicted from the genome.
Proteogenomic analysis of tumour biopsies holds the unique potential to allow researchers and clinicians to fully understand the molecular basis of an individual tumour by combining precise information on its genotype and its phenotype. Ideally, this will allow clinicians to select the best available treatment strategy, or to develop new treatments that not only effectively kill all of the cancer cells, but also prevent relapse.
At the Segal Cancer Proteomics Centre, we have already begun to translate our proteogenomics strategy into the clinic. We have extensively analyzed colorectal cancer tumour specimens using state-of-the-art proteomics technologies based on nano-LC-MS/MS using a non-targeted “shotgun” approach which provides an unbiased identification and quantification of proteins. This has enabled the high confidence identification of more than 8,000 proteins from individual tumours. Aligning this data with the corresponding genomic data from the same tumour samples (the genotype), has allowed us to clearly verify the presence of predicted mutations within the tumour on the protein level (the phenotype). Using this approach, we have been able to verify the presence of known driver mutations, such as KRAS G12V, in the tumour samples. After the specific mutation has been verified, we were then able to develop very sensitive targeted mass spectrometry assays which allow us to determine the actual level of that mutation in individual tumours using absolute quantification methods with high sensitivity and precision. Our goal is to provide unique molecular signatures for individual tumours. This will allow us to better define why specific patients do not respond to treatments predicted based on their genomic profiles, and thus improve the selection of appropriate treatment strategies.
Dr. Christoph H. Borchers is a professor at both the University of Victoria and McGill University, and director of the proteomics centres at both universities. Dr. Borchers’ expertise includes the improvement, development, and application of proteomics and metabolomics technologies, with a major focus on techniques for quantitative proteomics and metabolomics for clinical diagnostics.
Dr. René P. Zahedi is the Associate Director of the Segal Cancer Proteomics Center at McGill University. His research has focused on the mass spectrometry-based detection and quantitation of post-translationally modified proteins, with an emphasis on cancer-related signalling pathways and diagnostics.
Letai A. Functional precision cancer medicine-moving beyond pure genomics. Nat Med. 2017;23(9):1028-35.
Heist RS, Christiani D. EGFR-targeted therapies in lung cancer: predictors of response and toxicity. Pharmacogenomics. 2009;10(1):59-68.
Boja ES, Rodriguez H. Proteogenomic convergence for understanding cancer pathways and networks. Clin Prot. 2014;11(1):22.
Bekker-Jensen DB, Kelstrup CD, Batth TS, Larsen SC, Haldrup C, Bramsen JB, et al. An Optimized Shotgun Strategy for the Rapid Generation of Comprehensive Human Proteomes. Cell Systems. 2017;4(6):587-99.e4.
Cohen P. Protein kinases–the major drug targets of the twenty-first century? Nature Reviews Drug Discovery. 2002;1(4):309-15.