A team of scientists from the University of Strathclyde (Scotland) led by Matthew J. Baker, from the Department of Pure and Applied Chemistry at the University of Strathclyde in Glasgow (United Kingdom), has succeeded in having a new blood test detect brain tumors with 87% accurate.
This is hopeful news, because although it is quite rare, brain cancer often has a poor outlook. According to the National Cancer Institute, about 0.6% of people will develop brain cancer or other cancers of the nervous system in their lifetime, and the 5-year survival rate is less than 33%. This is in part because tumors have very nonspecific symptoms, making them more difficult to distinguish from other conditions.
The new study brings a lot of hope in this regard. The experts used infrared light to create a “biological signature” of blood samples from patients and applied artificial intelligence to look for signs of cancer. The test correctly identified brain cancer in a cohort of 104 people 87% of the time.
“This is the first publication of data from our clinical feasibility study, and it is the first demonstration that our blood test works in the clinic,” explains Baker in the journal Nature Communications that publishes the work.
Diagnosis takes longer
“Diagnosing brain tumors is difficult, leading to delays and frustration for many people. The problem is that brain tumor symptoms are quite nonspecific , such as headaches or memory problems. It can be difficult for doctors to determine which people have more likely to have a brain tumor, “adds study co-author Paul Brennan.
The lack of cost-effective tests that can help doctors classify people with brain tumors in primary care also means that it takes longer to accurately diagnose brain cancer, something that only complicates things for patients.
A faster means of diagnosis
Thanks to the new method, a technique called Attenuated Total Reflection Fourier Transformed Infrared Spectroscopy (ATR-FTIR) and combined with machine learning technology to detect brain cancer, it is “a simple, label-free, non invasive, non-destructive “to analyze the biochemical profile of a blood sample without requiring extensive sample preparation, the authors explain.
The ATR-FTIR technique allowed the researchers to develop a biochemical “fingerprint” of brain cancer.
The researchers trained a machine learning algorithm to use these biochemical fingerprints to diagnose brain cancer in a retrospective cohort of 724 people. This cohort included people with primary and secondary cancers, as well as control subjects, without cancer. Later, they used the algorithm to predict brain cancer cases in a sample of 104 participants. Of these, 12 people had cancer, including four cases of glioblastoma, one of the most aggressive forms of brain tumor.
The findings revealed a sensitivity of 83.3% and a specificity of 87% for the blood test. “With this new test, we have shown that we can help clinicians quickly identify which people with these nonspecific symptoms should be prioritized for urgent brain imaging,” says Brennan. “This means faster diagnosis for people with a brain tumor and faster access to treatment.”
Referencia: Development of high-throughput ATR-FTIR technology for rapid triage of brain cancer. Holly J. Butler, Paul M. Brennan, James M. Cameron, Duncan Finlayson, Mark G. Hegarty, Michael D. Jenkinson, David S. Palmer, Benjamin R. Smith & Matthew J. Baker. Nature Communications volume 10, Article number: 4501 (2019) DOI: https://doi.org/10.1038/s41467-019-12527-5
