The study shows that ‘protein signatures’ can accurately forecast the emergence of 67 diseases, which include multiple myeloma, non-Hodgkin lymphoma, motor neurone disease, pulmonary fibrosis, and dilated cardiomyopathy.
Protein ‘signatures’, obtainable through a blood sample, can help predict the occurrence of 67 diseases such as multiple myeloma, non-Hodgkin lymphoma, motor neurone disease, pulmonary fibrosis, and dilated cardiomyopathy. Utilizing protein data for disease prediction may enhance early diagnosis, prompt treatment initiation, and ultimately lead to better patient outcomes.
Studies involving thousands of proteins derived from a small blood sample highlight how proteins can effectively anticipate the emergence of various diseases.
This research, published in Nature Medicine, was conducted through an international collaboration that includes GSK, Queen Mary University of London, University College London, Cambridge University, and the Berlin Institute of Health at Charité Universitätsmedizin, Germany.
The research team analyzed data from the UK Biobank Pharma Proteomics Project (UKB-PPP), the largest proteomics investigation conducted so far, analyzing around 3,000 plasma proteins from a randomly chosen group of over 40,000 UK Biobank participants. The protein data is integrated with the participants’ electronic health records. Utilizing advanced analytical methods, the authors identified a ‘signature’ of 5 to 20 key proteins necessary for predicting each disease.
The study highlights that these protein ‘signatures’ can predict 67 diseases, including multiple myeloma, non-Hodgkin lymphoma, motor neurone disease, pulmonary fibrosis, and dilated cardiomyopathy.
Moreover, the protein prediction models outperformed traditional models that rely on standard clinical data. Predictions based on blood cell counts, cholesterol levels, kidney function, and diabetes tests (such as glycated hemoglobin) were less effective than the protein-based models in most cases.
The advantages of assessing and discussing the risk of future heart attacks and strokes through ‘cardiovascular risk scores’ are well recognized. This research introduces new potential for predicting a variety of diseases, including rare ones that commonly take a long time to diagnose, providing novel chances for timely diagnostics.
These findings need to be validated across different demographics, including individuals both with and without disease symptoms, as well as among various ethnic groups.
Lead author Professor Claudia Langenberg, Director of the Precision Healthcare University Research Institute (PHURI) at Queen Mary University of London and Professor of Computational Medicine at the Berlin Institute of Health at Charité Universitätsmedizin, stated:
“Typically, clinicians measure a specific protein, such as troponin for diagnosing heart attacks. We are incredibly enthusiastic about the possibility of discovering new markers for screening and diagnosis from the extensive collection of proteins that are now measurable in human blood. There is an urgent need for proteomic research in diverse populations to validate our findings and develop effective tests that can measure disease-relevant proteins in an affordable and clinically standardized manner.”
First author Dr. Julia Carrasco Zanini Sanchez, who was a research student at GSK and the University of Cambridge at the time and is now a postdoctoral researcher at PHURI, remarked:
“Several of our identified protein signatures performed comparably to or even surpassed some proteins already tested for their potential use in screening, like the prostate-specific antigen for prostate cancer. Thus, we are highly excited about the possibility that our protein signatures may enable earlier detection and ultimately improve prognoses for numerous diseases, including serious conditions like multiple myeloma and idiopathic pulmonary fibrosis. We uncovered a lot of encouraging examples, and the next phase is to prioritize specific diseases for further evaluation of their proteomic predictions in a clinical context.”
Co-lead author Dr. Robert Scott, Vice President and Head of Human Genetics and Genomics at GSK, added:
“One major challenge in developing new drugs is identifying patients who are most likely to gain from these treatments. This research highlights the potential of employing expansive proteomic technologies to find individuals at high risk across many diseases, which aligns with our strategy of using technology to enhance our understanding of human biology and disease. Further efforts will advance these insights and enhance our knowledge of how they can effectively support greater success rates and increased efficiency in drug discovery and development.”
