A new classification system that incorporates clinical and serologic data may be useful in the classification of idiopathic inflammatory myopathies, results of a recent analysis suggest.
By analyzing the patterns of relationships between 47 variables in this observational, retrospective cohort study, investigators identified 4 clusters of patients that corresponded to known idiopathic inflammatory myopathy subtypes.
Myositis-specific antibodies played a key role in predicting whether a patient belonged in a cluster, according to investigators, who noted that myositis-specific antibodies known to be associated with certain subgroups fell into the corresponding clusters they identified.
“This emphasizes that muscle biopsy may no longer be necessary for diagnosis of idiopathic inflammatory myopathies in patients with [myositis-specific antibodies] and corresponding phenotypes,” said the investigators, led by Kubéraka Mariampillai, PhD, of the Université Pierre et Marie Curie, Institut National de la Santé et de la Recherche Médicale (INSERM), Paris, France.
The study, described in JAMA Neurology, was based on data for 260 patients in the database of the French Myositis Network. The mean age of the patients was 62 years, and 63% were women.
Investigators conducted a multiple correspondence analysis based on 47 selected variables, including age, race, historical and recent diagnoses, dermatologic changes, creatine kinase levels, myositis-specific antibodies, and pathologic characteristics, among others.
Based on that analysis, they identified four subgroups corresponding to known entities: Dermatomyositis, inclusion body myositis, immune-mediated necrotizing myopathy, and antisynthetase syndrome.
Using decisional algorithm trees, investigators found that myositis-specific antibodies “played a key role in estimating connection to a cluster,” while by contrast, the pathologic data were “dispensable,” Dr. Mariampillai and her associates said.
The best tree omitted variables related to muscle biopsy and had a 78% correct estimation looking just at antisynthetase antibodies, dermatomyositis rash, and finger flexor scores of 3 or less, investigators said.
“The classification quality of the tree was appreciated on the basis of all classification criteria, with an overall sensitivity of 77.0% and a specificity of 92.0%,” the investigators said.
Patients with polymyositis were present in the study data, but fell mainly in the clusters corresponding to immune-mediated necrotizing myopathy and antisynthetase syndrome.
“This finding indicates that patients with polymyositis do not represent a subgroup of patients, and use of this term should probably be discontinued,” Dr. Mariampillai and coinvestigators said.
The study was supported by Association Française contre les Myopathies, and by CSL Behring, which partly funded the development of electronic case report forms. Dr. Mariampillai and colleagues reported no conflicts of interest related to this work.
SOURCE: Mariampillai K, et al. JAMA Neurol. 2018 Sep 10. doi: 10.1001/jamaneurol.2018.2598.