A genomic risk score (GRS), based on DNA analysis, may reliably help diagnose children with juvenile idiopathic arthritis (JIA) and its various subtypes, a large study reports.
The study, “Genomic risk scores for juvenile idiopathic arthritis and its subtypes,” was published in the journal Annals of the Rheumatic Diseases.
JIA encompasses all forms of arthritis occurring before age 16. The underlying cause is not well understood, and its diagnosis is based on symptoms that can vary widely among patients. Currently, no JIA-specific tests help with a diagnosis.
Recent studies have suggested an inheritable component to JIA because it possesses many features of other autoimmune disorders, including shared genes that increase susceptibility and greater risk in patients’ family members. As such, genetics may be used to predict risk and aid in diagnosis.
GRSs have been used in other conditions, including celiac disease, to predict patients from unaffected individuals with high specificity. Such methods are based on changes to the genetic code called single nucleotide polymorphisms (SNPs), which consist in the alteration of one of the building blocks of DNA (nucleotides).
Here, researchers in Australia, the U.S. and the U.K. set out to build a GRS that can identify people susceptible to JIA.
“This study aims to create a GRS which in-principle could be used to support the current clinical JIA diagnosis practice,” the scientists wrote.
The GRS was developed by analyzing DNA from a large group from the U.K., which included 2,758 JIA patients and 5,187 matched controls without the condition.
To validate the GRS, two additional groups were included; one from the Children’s Hospital of Philadelphia (CHOP), which consisted of 1,229 patients in the U.S. and Norway, plus 5,512 children as controls; and the other from the Childhood Arthritis Risk Factor Identification study (CLARITY) in Australia, composed of 558 cases and 704 controls. All participants were of European ancestry.
Computational methods were used to calculate GRS, which selected the optimal number of SNPs based on the model with the highest average area under the receiver operating characteristic curve (AUC) — a measure of the usefulness of a test to distinguish between those with or without JIA.
The GRS for JIA had an AUC of 0.671 in the U.K. group, which indicated an acceptable ability to distinguish between JIA patients and controls. Assessments in the other two groups validated the method, as the AUCs in the CHOP group (0.657) and the CLARITY group (0.671) were similar to the result in participants from the U.K.
A measure called odds ratio was calculated, in which a value above one represents greater likelihood of having JIA. Results showed that the odds of having JIA were 1.83 times higher in the CHOP group and 2.01 times greater in the CLARITY group relative to controls.
Next, the team extended the analysis to construct GRSs specific to different JIA subtypes.
The GRS with the strongest ability to predict a JIA subtype was for enthesitis-related JIA, which had an AUC value of 0.82 in the U.K. group, 0.84 in CLARITY, and 0.7 in the CHOP participants. The GRS for oligoarticular JIA also was a strong predictor with an AUC range from 0.72 to 0.77.
In contrast, the weakest subtype GRSs were for systemic JIA (0.528) and undifferentiated JIA (0.542), a JIA subtype with no or multiple characteristics of other JIA subtypes.
“A strength of this study is that the JIA GRS was developed on a U.K. data set and externally validated in two independent studies in Australia and the USA, indicating the robustness of the score,” the investigators wrote.
“Therefore, in the hands of a primary care doctor, a diagnostic algorithm based on a JIA-GRS may provide a more timely, accessible and reliable means of assessing children with musculoskeletal symptoms who may be JIA cases,” they added.
According to the research team, a limitation of the study was that it did not assess the JIA GRS in individuals of non-European ancestries.