Blood protein profiling combined with genomics advances rare disease diagnosis

For millions of people living with rare diseases, the moment a genetic test comes back negative or inconclusive can feel like a door slamming shut. Genome sequencing has transformed rare disease diagnosis over the past decade, but it has also revealed its limits: a large proportion of patients still walk away without answers, even after their DNA has been thoroughly analyzed. Researchers from Queen Mary University of London, the Berlin Institute of Health at Charité, and Genomics England have now demonstrated a way forward—by measuring proteins in the blood alongside genomic data, they can resolve cases that sequencing alone leaves in the dark.

The challenge at the heart of this work is a category of genetic findings called variants of uncertain significance, or VUS. These are genetic changes that show up on a sequencing report but carry no clear verdict: is this variant actually causing the disease, or is it just a harmless difference? Without that answer, a patient remains undiagnosed. The researchers reasoned that proteins might provide the missing piece. A genetic variant that disrupts a gene's function should leave a trace in the bloodstream—an unusually high or low level of the protein that gene produces. If a patient carries a suspicious genetic variant and also has abnormal levels of the corresponding protein, that combination offers evidence that the variant is genuinely harmful.

To test this idea, the team analyzed blood samples from people with rare diseases who had already undergone genome sequencing through Genomics England's 100,000 Genomes Project but still lacked a diagnosis. They measured nearly 1,500 proteins in each sample and cross-referenced the results with the genomic data. The approach worked. By combining protein measurements with genetic information, researchers were able to resolve previously uncertain findings and provide diagnoses for some patients. The method also identified candidate genetic variants and genes that could explain other patients' conditions and merit further study.

One striking example involved hereditary hemorrhagic telangiectasia, a rare inherited disorder affecting blood vessels. Cases that had remained undiagnosed after genomic analysis alone became clearer when protein data was added to the picture. In another case, researchers identified a rare genetic variant in a gene called TIE1, which plays a role in blood vessel function. A patient with an unexplained inherited heart disorder carried this variant and also had exceptionally low levels of the TIE1 protein in their blood. The patient's father carried the same variant and had the same condition. When researchers grew cells from the patient in the laboratory, those cells showed markedly reduced levels of the TIE1 protein and abnormal signaling. This convergence of evidence—genetic, proteomic, and cellular—suggested a new gene-disease link, though the researchers emphasize that such findings are candidates requiring further confirmation, not yet established causes.

The practical advantages of this approach are significant. Unlike some existing methods for studying genetic effects, which require skin biopsies and the cultivation of patient cells in a laboratory, this method uses blood samples. Blood is easier to obtain, less burdensome for patients, and potentially more scalable. As Dr. Julia Carrasco-Zanini of Queen Mary University noted, the goal is not to search endlessly for new variants but to make better sense of the genetic information already in hand. If a variant is accompanied by an unusual protein level, that combination tells a story the genome alone cannot.

The researchers are clear about the current limitations. Today's proteomic technology measures only a fraction of the roughly 20,000 proteins the human body produces, and not every disease-relevant protein can be reliably detected in blood. Not all disease-causing variants will change protein levels in the bloodstream. This means the approach cannot yet provide answers for every undiagnosed patient. But as the technology improves—measuring more proteins with greater sensitivity and establishing better reference ranges—the potential grows. Professor Claudia Langenberg described the opportunity as bringing different layers of biological information together: the genome provides the blueprint, while proteins reveal how that blueprint is being translated into biology in an individual patient.

The next phase will require larger studies across a wider range of rare diseases and more diverse populations. Improvements in proteomic technology and the development of robust clinical standards will be necessary before this approach becomes routine in diagnosis. But the proof of principle is now established. For patients and families navigating the long, frustrating search for a rare disease diagnosis, this work suggests that the answer may not require finding new genetic variants at all—it may simply require looking at the ones already discovered in a new way.