Sick child resting (© Tom Sickova - stock.adobe.com)
ITHACA, N.Y. – Researchers have developed a novel blood test that could transform how doctors diagnose and treat mysterious inflammatory conditions in children. The test, which analyzes cell-free RNA (cfRNA) in blood plasma, shows promise in distinguishing between various inflammatory syndromes that often present with similar symptoms, potentially leading to faster and more accurate diagnoses.
The study, published in the Proceedings of the National Academy of Sciences, focused on several inflammatory conditions that can be challenging to differentiate, including Kawasaki disease (KD), multisystem inflammatory syndrome in children (MIS-C), and various viral and bacterial infections. These conditions often share overlapping symptoms, making accurate diagnosis crucial for proper treatment.
The study, spearheaded by Iwijn De Vlaminck, associate professor of biomedical engineering at Cornell University, and lead author Conor Loy, an Ignite Fellow for New Ventures, used blood samples from 370 pediatric patients across four hospitals in the United States. The patients ranged from those with confirmed diagnoses of KD, MIS-C, viral infections, and bacterial infections to other hospitalized controls and healthy children.
Using advanced RNA sequencing techniques, the researchers analyzed the cfRNA profiles in these blood samples. Cell-free RNA are fragments of genetic material that circulate in the bloodstream, released by cells throughout the body. Unlike traditional blood tests that mainly reflect the immune response, cfRNA can provide insights into both the immune system’s activity and potential damage to various organs and tissues.
“When you analyze RNA in plasma, what you’re looking at is RNA from dying cells, and also RNA that’s been released from cells anywhere in the body,” Loy explains in a media release. “This gives you a huge advantage. In inflammatory conditions, there’s lots of cell death. Cells are, in some cases, exploding and their RNA gets released into plasma. By isolating that RNA and sequencing it, we can discover biomarkers for disease and backtrack where the RNA is coming from to measure cell death.”
The team’s analysis revealed distinct cfRNA signatures for different inflammatory conditions. Perhaps most notably, they developed a machine learning model that could differentiate between KD and MIS-C with a remarkable 98% accuracy. This is particularly significant because these two conditions can be especially difficult to distinguish clinically, yet require different treatments.
The researchers didn’t stop there. They expanded their approach to create a multi-class machine learning model capable of differentiating between KD, MIS-C, viral infections, and bacterial infections. This model achieved an impressive 80% accuracy in distinguishing between these four conditions.
Beyond just diagnosis, the cfRNA profiles also provided valuable information about organ involvement in these inflammatory conditions. The researchers found that cfRNA could indicate damage to specific organs like the liver, heart, and lungs, even in cases where traditional clinical tests didn’t show clear signs of injury.
“I think a lot of the novelty and the technical innovation, the engineering, is in the data analysis,” De Vlaminck says. “We’re able to quantify how much of the RNA is coming from different organs. How much is coming from the liver, or epithelial cells in the vascular system. By quantifying the sources, we can also learn about injury processes that are likely immune-related but happening in vascularized tissues.”
The potential implications of this research are far-reaching. Currently, diagnosing and differentiating between inflammatory conditions in children often relies on a combination of clinical symptoms, various blood tests, and sometimes invasive procedures. A single blood test that could accurately distinguish between these conditions and provide information about organ involvement could streamline the diagnostic process, lead to faster and more targeted treatments, and potentially improve outcomes for young patients.
However, the researchers caution that while the results are promising, more work is needed before this test can be used in clinical settings. The study was relatively small, and larger, more diverse patient populations will need to be studied to confirm the findings. Additionally, the technology and analysis methods used in the study would need to be adapted for use in typical hospital laboratories.
Despite these challenges, the study represents a significant step forward in the field of pediatric diagnostics. As our understanding of the information contained in cell-free RNA grows, it’s likely we’ll see more applications of this technology in medicine, potentially revolutionizing how we diagnose and treat a wide range of conditions.
Paper Summary
Methodology
The researchers collected blood samples from 370 children with various inflammatory conditions at four different hospitals. They extracted RNA from these blood samples and used a technique called RNA sequencing to analyze it. This process allowed them to see which genes were active in each sample. They then used sophisticated computer algorithms, including machine learning models, to identify patterns in this genetic data that could distinguish between different conditions. These models were trained on a portion of the data and then tested on separate datasets to ensure their accuracy.
Key Results
The study found that cfRNA profiles could distinguish between different inflammatory conditions with high accuracy. A model comparing Kawasaki disease and MIS-C achieved 98% accuracy. A more complex model differentiating between four conditions (KD, MIS-C, viral infections, and bacterial infections) achieved 80% accuracy. The cfRNA profiles also provided information about organ damage, correlating with clinical markers of liver and heart damage.
Study Limitations
The study had a relatively small sample size of 370 patients, and all were from hospitals in the United States. The results need to be confirmed in larger, more diverse populations. The technology used is also complex and would need to be simplified for use in typical clinical settings. The study also noted that the model had some difficulty distinguishing bacterial infections, possibly due to the variety of infection types in this category.
Discussion & Takeaways
The researchers believe this study provides proof of concept that cfRNA can be used to differentiate between inflammatory conditions in children and provide information about organ damage. They suggest that this could lead to a single blood test that could help diagnose these conditions more accurately and quickly than current methods. The study also identified specific genes that differ between conditions, which could provide new insights into the biology of these diseases.
Funding & Disclosures
The research was supported by grants from the National Institute of Child Health and Human Development, part of the National Institutes of Health (NIH).
