Mapping the Plasma Proteome: Unlocking Biomarkers for Precision Medicine

Advances in high-throughput proteomics now allow researchers to profile circulating plasma proteins at unprecedented depth and scale – accelerating efforts to identify novel biomarkers for the early detection of disease, monitoring and predicting response to treatment.

However, the complexity of the plasma proteome poses challenges: it contains a diverse array of proteins that vary considerably in abundance.

“Proteins exhibit significant inter-individual variability, particularly in blood, so large-scale analyses are crucial to uncover disease-specific changes and transition these discoveries into the clinic,” says Christoph Messner, assistant professor for precision proteomics at the University of Zurich, Switzerland.

Researchers employ a range of next-generation tools for high-throughput analysis of hundreds to thousands of blood samples to identify and validate potential disease-related biomarkers.

“We rely on mass spectrometry (MS)-based proteomics to detect and quantify the highly abundant proteins, while multiplex affinity-based assays allow us to profile those at lower concentrations,” says Fredrik Edfors Arfwidsson, assistant professor and docent in biotechnology at the KTH Royal Institute of Technology, Sweden. “By combining these complementary technologies, we can capture a more comprehensive view of the plasma proteome.”

Next-generation tools for proteomics

MS-based proteomic workflows are well-established in research laboratories and are routinely used for biomarker discovery and profiling. Over the past decade, dramatic advances in technology have led to faster, more sensitive instruments capable of identifying thousands of proteins and peptides in a single run.

“You can scan much faster now, which has significantly increased throughput,” says Messner. “The latest instruments also enable unprecedented proteome depth, allowing us to detect a vast number of proteins within remarkably short measurement times.”

In recent years, the introduction of data-independent acquisition (DIA) methods has also been a game changer for MS analyses.

“This was a really important step forward,” expresses Messner. “DIA enables shorter measurement gradients – reducing analysis times and increasing throughput. The results are also more reproducible than conventional data-dependent acquisition (DDA) methods.”

Advancing Quantitative Proteomics With Next-Gen DIA

Explore strategies for increasing throughput without compromising data quality
View App Note

Advertisement

Affinity-based assays that use binding to target proteins to enable the simultaneous detection and quantification of hundreds or even thousands of specific proteins from a single sample, offer a powerful alternative to MS-based proteomics. The number of assays on these highly-multiplexed proteomic platforms has increased ten-fold over the past 15 years, rising from fewer than 1,000 to over 11,000.

One of these high-throughput proteomics platforms uses the proximity extension assay (PEA), which combines oligonucleotide-linked antibodies with quantitative real-time PCR.

“These multiplex panels have expanded dramatically, thanks to antibody barcoding,” explains Edfors Arfwidsson. “We can now measure up to 5,400 proteins from just a single drop of blood, or less than 10 microliters of plasma.”

Another tool for large-scale proteomics employs aptamers – nucleotide-based compounds with high protein-binding specificity and sensitivity – instead of antibodies. The latest version enables the simultaneous measurement of over 11,000 proteins from just 55 microliters of plasma.*

Large-scale plasma proteomics

As a Scientific Director of the Human Blood Atlas, part of the wider Human Protein Atlas, Edfors Arfwidsson is at the forefront of efforts to use advanced proteomic tools for the large-scale profiling of plasma proteins in health and disease.

“We initially focused on profiling circulating proteins in healthy individuals, which revealed that each person has a unique proteomic fingerprint that remains remarkably stable over time,” says Edfors Arfwidsson. “This insight led us to expand our work and start to explore different disease states.”

In 2015, the team published a landmark study analyzing the plasma proteome in cancer patients. Using the PEA, they measured 1,463 proteins in minute amounts of blood from more than 1,400 cancer patients across 12 cancer types collected at the time of diagnosis and before treatment.

“By applying machine learning, we identified biomarker panels that could effectively differentiate between different cancer types and even stage colorectal cancers,” says Edfors Arfwidsson. “This was achieved from just one drop of blood.”

Building on this work, the Disease Blood Atlas was launched at the most recent Human Proteome Organization (HUPO) annual meeting. This open-access resource contains next-generation blood profiling data across 59 different diseases and 6,121 patients, covering cardiovascular, metabolic, cancer, psychiatric, autoimmune, infectious and pediatric diseases.

“It’s fascinating to observe how some well-known blood-based biomarkers behave across different diseases,” says Edfors Arfwidsson. “For example, we found that a cancer biomarker was also upregulated in certain infections.”

Other large-scale proteomics studies are generating independent datasets that will enable cross-validation of candidate biomarkers across different populations.

One such study examined the impact of common genetic variation on circulating blood proteins and their role in disease. The researchers measured the abundance of 2,293 proteins in 54,219 participants, uncovering over 14,000 associations between common genetic variants and plasma proteins, over 80% of which were previously unknown. This vast dataset is now accessible to scientists around the world through the UK Biobank.

Another team applied high-throughput MS to map the plasma proteome for sepsis, a life-threatening condition caused by a dysregulated host response to infection leading to organ failure. By analyzing 2,612 samples from 1,611 patients, they identified potential biomarkers that could help pave the way for precision medicine approaches to improve sepsis diagnosis and treatment.

Researchers from the UK and China found that loneliness is linked to a higher risk of illnesses such as heart disease, stroke, type 2 diabetes and susceptibliity to infection. They drew this conclusion after studying data from over 42,000 participants across nearly 3,000 plasma proteins in the UK Biobank.

The scale of these plasma proteomics studies continues to grow, promising to revolutionize our understanding of diseases and their treatments. The UK Biobank recently launched the world’s most comprehensive study of plasma proteins to date, aiming to measure up to 5,400 proteins across 600,000 samples, including those from half a million participants and 100,000 follow-up samples collected up to 15 years later. This unparalleled large-scale initiative will create a first-of-its-kind database, enabling researchers to investigate how changes in blood protein levels during mid-to-late life influence an individual’s disease risk.

Translational challenges

While large-scale proteomics studies are very powerful tools for plasma biomarker discovery, significant barriers remain in translating these findings into clinical applications.

A big challenge lies in integrating and analyzing the enormous datasets generated to identify panels of protein biomarkers associated with specific disease phenotypes.

“We have a core team of four bioinformaticians working solely on these datasets, and we’re only just scratching the surface of their potential,” says Edfors Arfwidsson. “We’ll probably spend another 5 or 10 years integrating these data with other resources.”

The adoption of artificial intelligence (AI) and machine learning tools has significantly enhanced the capacity to interpret these complex datasets. For example, in MS-based proteomics, machine learning – particularly deep learning – can now predict experimental peptide measurements from amino acid sequences alone. However, while these tools excel in pattern recognition and predictive modeling, their success relies on access to large, high-quality annotated datasets that are often limited in proteomics.

Beyond data handling and analysis, achieving regulatory approval poses additional challenges. Validation of protein biomarkers requires rigorous evidence of their accuracy, specificity, sensitivity and reproducibility across diverse patient populations. However, the lack of standardized experimental protocols, data formats and quality control measures often undermines reproducibility and comparability across studies.

“These advanced proteomics technologies, combined with AI and machine learning, are generating a lot of promising data – but we’re still missing the tools we need to validate these findings,” says Edfors Arfwidsson. “We’re identifying promising biomarker panels, but it’s really hard to cross-validate them because of the lack of standardization.”

Addressing these hurdles will require coordinated efforts between researchers, industry and regulatory bodies. Establishing standardized frameworks and robust validation protocols is critical to ensuring biomarkers can transition from the lab to the clinic without compromising patient safety.

An exciting future

Next-generation proteomics is enabling researchers to explore the plasma proteome with unprecedented depth and scale to deepen understanding of human health and disease.

“It’s an exciting time to work in this field,” says Messner. “As instrumentation and methodologies continue to advance rapidly, proteomics will become increasingly important for clinical applications.”

Blood-based protein biomarkers hold the potential to revolutionize healthcare by enabling earlier disease detection, personalized treatment strategies and more effective monitoring of therapeutic responses.

“In the near future, I can foresee that we’ll be routinely profiling the plasma proteome to detect early signs of disease or other markers that can provide information about a person’s health trajectory,” predicts Edfors Arfwidsson. “An annual blood test to assess risk factors or monitor specific diseases would be feasible to deploy at scale.”

*This article is based on research findings that are yet to be peer-reviewed. Results are therefore regarded as preliminary and should be interpreted as such. Find out about the role of the peer review process in research here. For further information, please contact the cited source.