Many of the diseases that attack our bodies lack viable, objective biomarkers.  That means there are delays in diagnosing them in patients.  Which clearly means the chances of succumbing to these diseases are clearly a risk.

That is why Drs. Julia Carrasco-Zanini (Queen Mary University of London) and Claudia Langenberg (University of Cambridge) garnered a team of 18 other professionals from across Europe and the USA to conceive of a new screening test.

A blood sample, analyzed for its protein components, was the chosen vehicle, as described in their paper, Proteomic Prediction of Common and Rare Disease, published in MedRxiv.  We already can study blood to discern diabetes and heart disease, but the goal of this team was to ferret out cancers and motor neuron diseases, among some 218 potential other disease states.  Using a database of some 41,931 folks (UK Biobank Pharma Proteomics Project (UKB-PPP), the largest proteomic collection to date), they focused on 3000 proteins that could be found in the blood samples.

Statistical models were developed using 70% of the census participants.  They then tested the model on the remaining 30% of the subjects.   Using both the protein-based model and standard clinical models, they were able to diagnose 52 medical conditions.  These included cardiomyopathy (heart muscle walls get stretched and thin) , pulmonary fibrosis (lung tissue scarring), bone marrow cancer, and motor neuron diseases.

Unfortunately, this process had a 10% false positive rate.  This is a bit high, but it still will allow patients to undergo routine quick screening, where the diseases may be discovered early.

But, there’s another issue.  Since the test population was primarily European, there needśto be testing of more diverse populations and geographies so that general acceptance of these protocols can be afforded all peoples.

Share this: