Saliva biomarkers show promise for tracking recovery from concussion injuries, and for predicting which patients will experience prolonged symptoms, according to recently published research in the Journal of Neurology.
The paper, titled “Saliva RNA Biomarkers Predict Concussion Duration and Detect Symptom Recovery: a Comparison with Balance and Cognitive Testing,” showed that that an algorithm based on non-coding RNA (ncRNA) molecules in the saliva outperformed current clinical toolsets in predicting persistent post concussion symptoms (PPCS). The predictive value was even greater when the saliva test was combined with existing balance and cognitive assessment tools.
The saliva-based algorithm was also found to be as accurate in identifying symptom recovery as currently used balance and cognitive testing. The authors concluded that the ncRNA saliva test could provide accurate expectations for recovery, stratify need for intervention, and guide safe return-to-activity decisions.
The study was a collaborative effort between New York-based molecular diagnostics company Quadrant Biosciences, and several hospital and academic research universities including SUNY Buffalo Jacobs School of Medicine and Biomedical Sciences, Penn State College of Medicine, Vanderbilt University, Harvard University, Colgate University, SUNY Upstate Medical, Adena Medical Center, Alberta Children’s Hospital Research Institute, and Boston Children’s Hospital.
Quantifying the Results
More than 3 million concussions occur each year, the majority occurring among children and young adults. Despite the prevalence of the injury, there are few clinically valid methods for its diagnosis or prognosis. Moreover, many of the existing toolsets are subjective in nature and susceptible to manipulation, and thus unable to accurately predict persistent post-concussion symptoms. As a result, there is great value in an objective biomarker panel that is not only accurate, but easily collected and measured.
In the present study, 505 saliva samples from 112 individuals between the ages eight to 24 years with mild traumatic brain injury were analyzed. The initial samples were collected less than two weeks post-injury, with follow-up samples collected over three weeks after the injury. At both appointments, computerized balance and cognitive tests were also performed.
Using machine learning tools, the researchers developed an algorithm using age and 16 ncRNAs that predicted PPCS with greater accuracy than the validated clinical toolset. In addition to outperforming the current toolset in predicting PPCS, the algorithm most accurately predicted PPCS when the ncRNAs were combined with age, balance and cognitive measures. (area under the curve (AUC) 0.86; 95% CI 0.84-0.88).
“This study provides important insights into the functional and biological differences between people who recover quickly and those who recover more slowly following a head injury,” says Greg Fedorchak, PhD, lead author on the paper and a research scientist at Quadrant Biosciences. “The ability to predict recovery time and recovery status using an objective saliva test removes a lot of the guesswork involved and lets the biology speak for itself. This has the potential to alleviate the burden of uncertainty for both patient and clinician alike, leading to better and more efficient care.”