SLE is a systemic autoimmune disease with a complex etiopathogenesis. Although the underlying etiology is not clear, the genetic susceptibility plays a major causative role in the familial SLE [9]. A unique form of SLE recognized as a monogenic lupus is characterized by positive family history, early onset presentation, mainly with mucocutaneous manifestations, and experienced refractory disease course. Patients with monogenic lupus typically have underlying genetic defects affecting different immune system components. To date, more than 30 genes have been reported as a cause of monogenic lupus [13]. However, monogenic lupus remains a descriptive term. Of note, some forms of monogenic lupus, such as patients with C1q deficiency, may share phenotypic and immunopathogenic features with rare monogenic autoinflammatory disorders, which necessitate the use of criteria to classify patients with monogenic lupus [25, 26]. To the best of our knowledge, none of the available classification criteria for SLE have been examined in patients with monogenic lupus. More recently, the performance EULAR/ACR-2019 criteria proved to be efficient with significant sensitivity in classifying patients with sporadic cSLE regardless of the ethnicity. However, in comparison with various classification criteria, there are inconstant specificity estimates [20, 22, 27–29].
This study includes a large multinational cohort of monogenic lupus with various genetic defects. Despite the heterogeneity of the clinical presentation and evolving the phenotypic features, and the underlying genetic variants, the EULAR/ACR-2019 criteria efficiently classified monogenic lupus patients, irrespective of the diversity of the underlying genetic variants. In this cohort, the EULAR/ACR-2019 criteria yielded a high sensitivity, which means it has the potential to identify monogenetic lupus early. However, the specificity was not as high as expected, and this may be in line with what has been previously reported in sporadic cSLE. It allowed the classification in the majority, 89.8% patients met the EULAR/ACR-2019 criteria, compared with 93.9% patients met SLICC-2012.
For this study, the SLICC-2012 criteria was chosen as the gold standard instead of the ACR-1997 owing to its higher sensitivity across all age groups of SLE, and as stated earlier it was used as a framework in diagnosis by many physicians [20]. In addition, it classified early most of our cSLE controls, diagnosed by expert rheumatologists as cSLE; SLICC-2012 (90.9%) versus ACR-1997 (61.8%).
In monogenic lupus group, the difference between EULAR/ACR-2019 and SLICC-2012 fulfillment was primarily due to ANA negativity. Among patients who failed to meet the EULAR/ACR-2019 criteria, two patients with DNase1L3 scored more than 10 but they were ANA negative. It is worth noting that criteria domains, particularly immunological domains, were obtained at the initial diagnosis for the purpose of assessing the efficiency of early classification. This entry creation might reduce the sensitivity and increase specificity of EULAR/ACR-2019, particularly in cSLE and monogenic lupus, as autoantibodies can be detected later in the course of disease [6, 29]. Other observed differences in clinical domains’ fulfillment were hematological and neuropsychiatric manifestations. As three out of four items can be met in SLICC-2012 by having hematologic manifestations alone, this was combined into one domain in EULAR/ACR-2019. Additionally, some of the expanded neuropsychiatric manifestations counted in the SLICC-2102 are not considered in the EULAR/ACR-2019. This was illustrated in our study group as half of the patients who met SLICC-2012 neuropsychiatric criteria did not fulfill that of the EULAR/ACR-2019. Conversely, patients who met the EULAR/ACR-2019, but not the SLICC-2012, did so by scoring high weighted items however, with fewer features. The weighted scoring system also misclassified patients with sJIA and JDM who presented with arthritis and cutaneous manifestations. There was no difference in renal domain, as lupus nephritis class III and IV attained the highest weights in the EULAR/ACR-2019 and were enough to classify patients in the presence of ANA positivity, corresponding to stand-alone renal SLICC criteria. The significant difference in meeting immunological domain between the two criteria is largely due to the expanded scope of immunologic items in the SLECC-2012 compared to the EULAR/ACR-2019.
In real life, the data can be collected in a longitudinal manner. Thus, the criteria are cumulative and need not be present concurrently. It is agreed that the design of the classification criteria is primarily aimed at clinical research and is not proposed for diagnosis but is aimed at obtaining higher specificity, which usually comes at the expense of sensitivity. However, it is intended to be used as a diagnostic guide for the patients’ classification [29].
Our study has its limitations, primarily is that none of the previous classification criteria have been evaluated in monogenic lupus; hence, there is no true gold standard for classification or diagnosis. We realize that the expert physician’s opinion is not a standardized tool. However; it is the available practical “gold standard”. Furthermore, a more inclusive approach might be required in the future to have clearer idea, such as testing those criteria in cohort, including all monogenic inborn errors of immunity and dysregulation. Control cSLE patients have not been studied genetically, thus there might be a possibility of some of those patients carry monogenic or somatic mutations. It is a retrospective cohort; some immunological variables as lupus-specific antibodies were not obtained from the control group (sJIA and JDM), at the time of initial diagnosis. Although this work is a multicenter, multinational study, the majority of enrolled patients were Arab, racial influence has not been explored.