The pace of precision medicine is accelerating all the time. The use of molecular information about a person’s condition to make treatment decisions is becoming increasingly common, particularly in oncology where genomic profiling is now standard clinical practice.
Recent advances include transitioning from using tissue or tumor biopsy samples for molecular profiling to using plasma samples known as liquid biopsies, which detect circulating cells or DNA from tumors. Blood is much easier and less invasive to obtain than tissue samples, and particularly useful when a tumor is inaccessible for biopsy. Furthermore, patient profiling using regular blood draws can give a more comprehensive overview of a person’s dynamic health than relying on tissue biopsies at infrequent intervals.
While much attention has been focused on using blood samples for genomic analysis, plasma contains much more information about a person’s condition than DNA alone.
By revealing the ultimate phenotypic output of the genome acting together with the environment, the tens of thousands of proteins that make up the plasma proteome can give a more accurate picture of the functional status of a person’s health than genomic profiling alone. This disparity between the richness of biological information available from the proteome compared with the genome has significant clinical implications.
The link between genotype and phenotype can be challenging to predict, so targeted therapies developed based on genomic or even transcriptomic profiling can often miss the mark. It’s not uncommon to see a lack of benefit in patients whose genomic profile is predictive of response, or differing results to the same treatment in patients with seemingly similar profiles.
Integrating proteomics into biomarker discovery can therefore give a more accurate representation of clinical phenotype and biological responses than relying on genomics or transcriptomics alone.
The plasma proteome is highly complex and heterogeneous, containing many tens of thousands of proteins with a wide dynamic range of concentrations and post-translational modifications. As a result, identifying and validating relevant and clinically valuable protein biomarkers can be challenging.
Biomarker discovery and validation also require consistent measurements across thousands of heterogeneous patient samples collected at multiple timepoints – something that previous proteomics technologies have been unable to achieve. Analytical techniques must also be sensitive enough to pick out small changes in the expression of biologically significant rare proteins against the background noise of abundant yet irrelevant molecules. Furthermore, affinity-based proteomics approaches rely on looking for the ‘usual suspects’ in plasma, based on panels of existing proteins, which limits the discovery of informative novel biomarkers.
These limitations have meant that the plasma proteome has remained a largely untapped clinical resource, until now.
New advances in proteomics technology and processes, led by Biognosys, have made unbiased and repeatable high-throughput plasma proteomics a reality. Developments in sample preparation, chromatography, and mass spectrometry mean that large-scale clinical studies with multiple longitudinal samples are now possible.
Unlike its predecessors, our data-independent acquisition mass spectrometry (DIA-MS) workflow analyses every peptide within a sample, providing unbiased analysis and unprecedented depth for true discovery and innovation. This complete analysis of all peptides ensures reproducible results that have previously eluded protein biomarker discovery.
Our Hyper Reaction Monitoring (HRM™) technology allows DIA-MS to be applied with high throughput in large-scale studies involving thousands of samples from clinically relevant biofluids, including blood plasma, cerebrospinal fluid (CSF) and urine.
The feasibility and power of large-scale plasma proteomics for biomarker discovery were proven in the DiOGenes study – one of the largest clinical proteomics studies to date, involving multiple plasma samples from more than 500 people taking part in a weight loss trial.
Together with Nestlé, we successfully analyzed more than 1,500 plasma samples using high-throughput DIA-MS at a rate of over 30 samples per day per instrument. In total, our workflow achieved unprecedented robust quantitative analysis at scale, precisely measuring over 500 individual proteins and identifying 20 potential biomarkers to take forward for further investigation and validation. Additionally, we were able to identify patterns of protein glycation with direct relevance to the mechanism of action of the weight loss intervention, which would not have been detectable using other approaches.
More recently, our plasma proteomics workflow was applied in the ongoing Phase 2 PRINCE trial, sponsored by Parker Institute for Cancer Immunotherapy (PICI) in collaboration with Bristol-Myers Squibb, Apexigen, Inc. and Cancer Research Institute. In this trial, people with pancreatic cancer were treated with combinations of immuno- and chemotherapy. Their plasma proteomes were analysed during the trial, allowing us to identify proteomic signatures associated with clinical outcomes, which will be presented at an upcoming oncology conference.
Our experts are continuing to push the boundaries of large-scale plasma proteomics. We have recently announced our roadmap for taking large-scale plasma proteomics even further by increasing the depth of analysis.
Currently, we can quantify around 1,700 proteins per sample, but with upcoming developments in our proteomics workflows and updated data analytics algorithms empowered with Artificial Intelligence (AI) and Machine Learning (ML), we are increasing our analysis depth to up to 3,000 proteins by the end of 2021.
The future of personalized and precision medicine is unlikely to rely on one profiling technique alone. That’s why we’re also working on integrating proteomics with other approaches to harness the power of multi-omics. In pursuit of this goal, we are supporting the large-scale Cancer Scout research initiative, together with University Medical Center Göttingen and Siemens Healthineers, which will combine genomics, proteomics, digital pathology and AI to support personalized cancer medicine – a concept known as proteogenomics.
After being overshadowed by advances in genomics and transcriptomics over the past decade, it’s now time for large-scale plasma proteomics technologies to shine. Unprecedented reproducibility, sensitivity, throughput, and clinical transferability puts proteomics in prime position to drive forward the discovery of novel, biologically relevant biomarkers that will transform diagnosis and treatment for a huge range of health conditions in the future.