The panelists statistically tested 982 candidate criteria for systemic JIA-related MAS and evaluated how well these criteria accurately classified the MAS patients. The 37 best-performing criteria and eight criteria derived from existing literature were examined and evaluated by the panel. They achieved consensus on the classification of 391 of the 428 patient profiles, or 91.4%. The panelists reached 82% consensus on the final MAS classification criteria. In validation analyses, the criteria’s sensitivity was 0.73, and their sensitivity was 0.99. In addition, there was high agreement (0.76) between the classification of MAS or not MAS obtained using the criteria and the original diagnosis of the patients by their treating physicians.
Dr. Schneider believes that the new classification criteria for MAS in systemic JIA could also be suitable for use in patients with adult-onset Still’s disease, which many rheumatologists consider the same disease as systemic JIA in patients with onset of disease at 16 or older. However, this extrapolation should be studied formally, he urges. He doesn’t think that the criteria could be used in either adult or juvenile patients with MAS in the context of other diseases where it can be a complication, such as SLE.
Some Caveats to Consider
How should clinicians utilize these new classification criteria in their systemic JIA patients? These are not diagnostic criteria, which can reflect all the different possible clinical features of a disease, Dr. Schneider stresses. The panel evaluated hundreds of candidate criteria, and some that were eliminated in the process may still be useful in some clinical situations.
“The specificity of these criteria has not yet been properly validated in a different cohort of patients,” he says. Some laboratory tests for some markers that could be useful were not included, such as levels of soluble CD25, an interleukin 2 receptor. “Soluble C25 levels may still add valuable information to be considered in the evaluation of MAS in a patient with systemic JIA. It’s available at many academic centers. There are also other markers that were not included in the proposed criteria, such as soluble CD163 levels, that may be helpful.”
The panel’s report noted some caveats to consider when applying these new criteria. The real patient profiles used by the experts to base their opinions of various candidate criteria included relatively limited information, says Dr. Schneider. Most importantly, these profiles did not include changes in the patients’ clinical and laboratory features over time.
