PressSplit
Proteins in the blood can provide information about the risk of disease. © IMAGO/Westend61
Research groups are developing blood tests that can provide several different diagnoses simultaneously.
Frankfurt – Using a simple blood test to determine the individual risk of certain diseases or at least to detect them at a stage when they are not yet causing any symptoms – and this for a wide range of ailments: This possibility would revolutionize diagnostics in a similar way to how X-rays or magnetic resonance imaging have done.
The advantages are obvious: if you know your risk of developing a disease, you can take countermeasures and perhaps prevent it from breaking out. Early detection often means less stressful treatment and improves the chances of recovery. This type of diagnosis would also save the health system money on expensive imaging procedures and treatment. However, such tests would have to be very reliable in order not to lull people into a false sense of security or cause unnecessary concern and trigger further tests. Keywords: false negative or false positive findings, overdiagnosis.
Detecting diseases earlier using markers in the blood
Research teams around the world are working on tests that meet these requirements. Some have already been developed for cancer, which are designed to detect several types of tumor at the same time using markers in the blood. However, none have yet been able to pass all the tests to be suitable for screening. Neither the German Cancer Research Center nor the German Cancer Society recommend such tests to date.
New blood test can detect risk for 67 diseases
But this kind of next-generation diagnostics is not just about cancer. A new blood test is designed to be able to identify the risk of 67 different diseases. It was developed by researchers at the Berlin Institute of Health at the Charité, Queen Mary University of London, University College London, the University of Cambridge and the British pharmaceutical company Glaxo Smith Kline. The diseases that the test is designed to predict include types of cancer such as multiple myeloma and non-Hodgkin lymphoma, motor neuron disease, in which nerve cells are progressively damaged, pulmonary fibrosis and dilated cardiomyopathy, a pathological enlargement of the heart muscle. The researchers have published their results in the specialist journal Nature Medicine published.
The test detects the signs of the respective disease risk using protein signatures. Background: Just one drop of blood contains thousands of proteins that can provide information about diseases. A well-known example of this is troponin, the measurement of which provides information about whether someone has had a heart attack. Or, less reliably: the PSA value, which can indicate prostate cancer. Or, less specifically: the C-reactive protein, which forms in the event of inflammation or infection.
Proteins in human blood can indicate diseases – many are still unknown
There are probably many other proteins in human blood that could indicate diseases. However, many of them are still unknown. The trick is to find out which ones can act as markers. The researchers used the UK Biobank Pharma Proteomics as a basis for this. They analyzed the data of around 3,000 plasma proteins from more than 40,000 randomly selected blood samples, which were linked to information from the participants’ electronic health records. In the next step, the team filtered out five to 20 proteins that seemed best suited to predicting certain diseases.
“It is a huge step that we can now identify new markers for screening and diagnosis from the thousands of proteins that circulate in human blood and are measurable,” says Claudia Langenberg, lead author of the study and head of the Computational Medicine working group at the Berlin Institute of Health and director of the Institute for Precision Medicine at Queen Mary University. For Robert Scott, Vice President and Head of the Human Genetics and Genomics Department at Glaxo Smith Kline, such proteomics techniques also offer important starting points for “increasing success rates and efficiency in drug research and development.”
Next, the test will be tested in people with and without symptoms of the disease, as well as in different ethnic groups.
Blood plasma-based screening procedure
Researchers at the Ludwig Maximilian University of Munich (LMU) and the Max Planck Institute of Quantum Optics are also working on a blood plasma-based screening method that can detect several health conditions with just one measurement. This team relies on infrared light and machine learning.
Infrared spectroscopy can be used to analyze the molecular composition of substances. “It is like taking a fingerprint from molecules,” says a statement from LMU. The technology provides “detailed information” via molecular signals for complex biofluids such as blood plasma. Companies also use blood tests.
Blood samples are examined using infrared spectroscopy
Although the method has long been used in chemistry and industry, it has so far been rarely used in medical diagnostics. The basis here was more than 5,000 blood samples that had been taken as part of a previous research project in the Augsburg area. The researchers examined them using infrared spectroscopy. The team then used machine learning to analyze the correlation between the molecular fingerprints measured with infrared and the medical data.
The multi-stage computer algorithm used is able to distinguish between different health conditions, including prediabetes, a precursor to type 2 diabetes that is often overlooked. If prediabetes is detected, the development of real diabetes could be prevented. According to the researchers, the algorithm was even able to filter out people who were healthy and remained so for years after the blood sample was taken.
The new study lays the foundation for the molecular infrared fingerprint to become a “routine part of health examinations” in the future, according to the LMU press release. It could “help doctors to identify and treat diseases more efficiently.” With more refined technology, even more applications are conceivable in the future: up to “personalized health screening, in which individuals regularly have their health checked and identify potential problems long before they become serious.” (pam)
