Thousands of new biomarker discoveries are published every year. Very few of them ever reach a clinical lab. That gap between discovery and clinical adoption is a persistent problem in precision medicine, but advanced proteomics is helping close it.
For years, the narrative around precision medicine centered on genomics. However, with the exception of genetic mutations in cancer, our genome is fixed at birth, whereas proteins respond continuously to stress, injury, infection, and disease. When a protein is discovered somewhere unexpected, such as a structural protein leaking into urine or a fragment released during non-programmed cell death, it can signal a breakdown in biological processes. The ability to identify the proteins associated with a particular disease can help detect that disease earlier and more precisely than other available diagnostic tools.
One clear illustration of the potential impact of disease detection through protein biomarkers is tumor detection. Cancer cells divide rapidly and imprecisely, and that imprecision releases abnormal proteins into circulation. But with insights from proteins, a clinician could diagnose disease and begin treatment early, well before the first symptoms appear. That is the promise of protein biomarkers, but more must be done to bring it from the research lab into the clinic.
Advances in protein detection
Three developments are accelerating the shift from biomarker discovery to clinical diagnostics. Firstly, mass spectrometry workflows can profile complex biological samples with precision. Secondly, proximity extension assays now allow researchers to measure large, targeted protein panels across extensive cohorts with high reproducibility. Finally, biobanks, including the UK Biobank, Singapore’s PRECISE, and repositories held by the National Institutes of Health, have reached a scale that enables researchers to link specific protein signatures to real health outcomes across large, diverse populations.
Combining advanced proteomics solutions and large biobank samples has already yielded impactful results. For example, biobank data helped confirm the relationship between the PCSK9 gene and cholesterol regulation, a finding that reshaped cardiovascular treatment. In multiple sclerosis, a blood test built around 18 validated protein biomarkers now gives clinicians a way to track disease activity and adjust treatment over the course of a patient's life, rather than relying solely on periodic imaging or symptom reports. A recent Cell study, which integrated genetic and proteomic data from more than 78,000 individuals, provided new insights into biological mechanisms that could accelerate drug target identification, enable earlier diagnosis, and support more precise therapeutic decisions.
Smaller, sharper panels win
Large biobank studies are critical to generating protein biomarker candidates. A single analysis of a population-level cohort can flag thousands of proteins associated with specific diseases or traits. That scale, however, doesn’t hold the same weight once the goal shifts from biomarker exploration to clinical use. A successful protein diagnostic must answer a specific clinical question rather than survey the entire proteome.
Biomarkers that have proven effective in lung cancer diagnosis offer a useful illustration. A "universal" panel intended to cover every angle of a biologically complex disease would likely include too many proteins to measure effectively, degrading performance across settings, making interpretation harder, and increasing assay cost and regulatory burden. Instead, a panel built around a specific, well-defined clinical decision, matched to how clinicians actually practice, tends to outperform a broader one. Work out of the University of Liverpool (Davies et al., 2023) illustrates this winnowing process. The researchers started with a large pool of candidate proteins and narrowed it to a panel of 240, about 10% of the original candidates, that could identify disease risk one to three years before diagnosis.
Panel size is not a minor technical detail here. It shapes assay complexity, cost, and regulatory pathway. For biopharma companies and assay developers, the focus is shifting from identifying more candidates to building scalable, dependable tests around the candidates that matter most.
Challenges in moving from research to diagnostics
Despite analytical innovations, protein biomarker discovery still faces hurdles as it moves from a research lab to the clinic. One key challenge is that reproducibility across different populations and disease states is difficult to establish. Additionally, many of the most informative protein signals occur at low concentrations and are meaningful only within a specific biological context, making them difficult to measure reliably at a clinical scale. Even a biomarker with strong analytical validity still must clear cost, operational, and regulatory hurdles before it can work as a diagnostic tool.
Overcoming these barriers requires a workflow in which discovery platforms, validation studies, assay development, and regulatory strategy are designed to interlock, rather than exist as separate efforts that happen to converge on the same protein.
Protein biomarkers beyond diagnosis
Beyond developing diagnostic tools, protein biomarkers are already shaping how clinical trials stratify patients, and validated protein panels are becoming the foundation of companion diagnostics. Once a biomarker is established, clinicians have a baseline against which to measure progress as treatment proceeds. Levels that decline over time can confirm that a therapy is working. Levels that remain elevated or climb again can signal a relapse before symptoms return. For example, biomarker data can help a clinician determine whether extending chemotherapy after surgery justifies the added toxicity for a given patient, and it can flag elevated relapse risk early enough to escalate care while sparing lower-risk patients from treatment they don't need.
Enabling the future of precision medicine with proteomics
Proteomic technology keeps advancing, and biobank-scale data keeps expanding, but neither can close the translation gap alone. Closing it will require researchers to narrow their questions, build smaller, more specific panels and validate them across diverse populations before an assay ever reaches a clinical setting. It also requires the tools and protocols to dig deeper when a protein appears somewhere unexpected, since that is often where the earliest biological signal resides.
The opportunity ahead for proteomics isn't a larger catalog of biomarkers. It's a shorter, more disciplined path from a protein signal to a decision a physician can act on, made earlier, with less invasive testing and better outcomes for the patient on the other end. As technologies continue to advance and governments and biopharma companies invest more in biobank-enabled research, we will see the impact of proteomics in realizing the power of precision medicine at scale.
Stephen Harding is Vice President and General Manager, Protein Diagnostics, Thermo Fisher Scientific