Blood Proteins Can Predict The Risk Of Developing More Than 60 Diseases

Authored by Marina Zhang via The Epoch Times (emphasis ours),

Over 60 diseases can be predicted just by looking at proteins in the blood, a study published Monday finds.

These proteins provided more accurate predictions for 52 out of 67 diseases than current clinical tests.

“Measuring one protein for a specific reason, such as troponin to diagnose a heart attack, is standard clinical practice. We are extremely excited about the opportunity to identify new markers for screening and diagnosis from the thousands of proteins circulating and now measurable in human blood,” lead author professor Claudia Langenberg, director of the Precision Healthcare University Research Institute at Queen Mary University of London, said in a press release.

Postdoctoral researcher Julia Carrasco-Zanini-Sanchez, who is also the study’s first author, told The Epoch Times that the study was prompted by her team’s prior research on a disease related to impaired glucose control.

“[The condition] is basically a form of prediabetes that you can only detect when you do what’s called an oral glucose tolerance test, but not through HBA1c (blood glucose testing) or fasting glucose testing,” she said.

“We started working with proteomics (large-scale study of proteins) to try to develop a test ... to predict the outcome of this oral glucose tolerance test without having to perform it because it’s routinely not done in clinical practice.”

Ms. Carrasco-Zanini-Sanchez said that their prior study led them to wonder if other diseases could be predicted using proteins.

She said that their current model predicts disease development in 10 years’ time.

“[Ten years] is kind of a long time frame for some of the diseases that we’re studying … a three- or five-year prediction timeline would be a bit more relevant. However, the data is still not large enough, which is why ... all of them are trained for 10-year incidents,” the first author said.

Ms. Carrasco-Zanini-Sanchez, who did her doctorate research on proteomics, told The Epoch Times that she hoped the study’s proteomics tests would be used to screen specialized populations most at risk for the disease in question rather than the entire population.

52 Diseases Identified

The study, published in Nature Medicine, used data from the UK Biobank and analyzed over 3,000 different blood proteins for 218 different diseases.

Over 40,000 people were recruited to have a sample of their blood taken for proteomics analysis.

These people were then followed for 10 years through their electronic health records to see what diseases they would develop.

For those who did end up developing various diseases, by studying the protein levels they had 10 years ago, researchers determined protein signatures for over 60 diseases.

Each protein signature is made up of five to 20 different proteins.

The researchers developed a clinical model to predict the risk for various diseases, which included information such as age, sex, and body mass index, among other factors.

On top of that model, they added the protein signature, disease biomarkers, or genetic risk scores to make three other models and compared the results.

With the protein signature model, the authors found significant improvements to the predictions for 52 diseases. They include celiac disease, dilated cardiomyopathy, liver cirrhosis, multiple myeloma, COPD, dementia, Sjogren’s disease, and prostate cancer, among others.

The authors highlighted that the biomarker model for prostate cancer, which is currently screened for by measuring a person’s prostate-specific antigen, was outperformed by their protein signature model.

They also identified certain proteins that only predicted one disease; for example, TNF receptor superfamily member 17, a protein responsible for B-cell development, was highly specific for predicting multiple myeloma.

Model Based on One-Time Sample

Participants’ blood samples were only taken once at the start of the study. Their health outcomes were followed through their electronic health records for 10 years.

Ms. Carrasco-Zanini-Sanchez said it is unlikely that blood protein levels would change too drastically.

“There are not that many studies with repeated proteomic sampling. But the ones that exist do show that there are quite stable levels or stable measurements of the proteins. Very large variations are mostly associated with changes in risk factors or the environment that may, on itself, predispose you to a different disease.”

Ms. Carrasco-Zanini-Sanchez envisioned that their test may help doctors make better judgments on diagnosing and treating high-risk groups.

“If we think about screening the entire population [for celiac disease] ... about one person in 100 people will basically go on to develop celiac disease,” she said, adding that many people may need to be tested to only help one person.

However, in specific population groups, like those with autoimmune disease, their risks of developing celiac disease are higher.

“This is kind of the general framework in which we envision,” she said, “It’s just about finding the right population in which you would apply this sort of test realistically.”

The first author said that doctors in the United States may also use the proteomics test to screen their own patients for diseases during checkups.