At the annual meeting, a team of researchers presented a new method to accurately assist in the early diagnosis of ankylosing spondylitis (AS). Their study found that a genetic risk score can help clinicians identify the condition earlier and at lower cost than current testing methods.
AS is a common cause of chronic back pain and may affect the skin, intestines, and eyes. The disease commonly begins during patients’ teens or 20s, is far more common in men than women, and has a strong genetic risk.
Zhixiu Li, PhD, of Queensland University of Technology Institute of Health and Biomedical Innovation in Australia, said the typical delay between the onset of AS symptoms and diagnosis is eight to 10 years and that patients often receive inappropriate treatment during the delay. Early diagnosis and intervention have been shown to improve outcomes, but early diagnosis can be challenging.
“In early disease, magnetic resonance imaging (MRI) imaging is the current gold standard for diagnosis, but it is very expensive, and many patients with early changes on an MRI scan don’t go on to get AS,” said Dr. Li. “Thus, we think genetic profiling may even be informative, particularly in early disease, but also potentially prior to onset of symptoms.”
Genetic risk score (GRS) uses thousands of genetic variants to calculate a person’s susceptibility for a particular disease. Dr. Li’s team of researchers examined two models of GRS to determine whether they could serve as an early diagnostic tool in AS. The first model was based on samples of European descent (n=7,742 patients with AS and n=14,542 controls). The second model was based on samples of East Asian descent (n=6,001 patients with AS and n=4,943 controls).
The researchers compared their models with standard testing for AS and reported results using area under the curve (AUC). In the European GRS model, the AUC was 0.92, with 83% sensitivity and 92% specificity compared with 0.87 with human leukocyte antigen-B27 alone. In the East Asian GRS model, the AUC was 0.95, with 91% sensitivity and 95% specificity.
“Our findings show that GRS has high discriminatory capacity and could be of clinical utility in early diagnosis at [a] lower cost than MRI,” said Dr. Li. “In addition, [it] could also be applied to identify individuals with [a] substantial risk of developing AS before major symptoms appear. It could also be applied to patients with arthritic symptoms but without a clear diagnosis.”
The researchers are working to improve their models and examining how GRS performs when combined with other clinical features or test results.