Javad Wahadat

Systemic lupus erythematosus (SLE) is a severe and potentially life-threatening autoimmune disease. SLE that begins in childhood (cSLE) typically exhibits a more severe clinical course than SLE in adults. This has significant implications for the quality of life, educational achievements, and employment opportunities of these patients. More than half of the patients fail to secure paid employment in young adulthood. By the time patients reach the age of 30, 5-10% of them will have died as a direct result of SLE, mainly due to infections or cardiovascular complications. At that age, the majority of surviving patients have also incurred irreversible damage from both the disease itself as its pharmacological treatment. Specifically, prednisone poses a significantly increased risk of developing damage such as cataracts, osteoporosis, osteonecrosis, growth delay, or even growth cessation. Therefore, it is important to develop therapeutic strategies that can reduce disease activity while minimizing the detrimental effects of both the disease and medication.

There are three potential strategies that can contribute to improving treatment. A recently developed therapeutic strategy is the so-called “treat-to-target” (T2T) concept. Here, treatment is adjusted based on disease activity to achieve a predetermined goal (remission or low disease activity). Although this concept has already been elaborated for SLE in adults, its feasibility in children with SLE has not been explored to date. Second, therapeutic interventions can be optimized by identifying measurable biomarkers that contribute to pathogenesis and are therefore associated with disease activity. Type I Interferons (IFN-I) are a group of proteins that play an important role in the pathogenesis of SLE. Therefore, these proteins represent a promising biomarker for disease activity. With the advent of drugs that can block the IFN-I pathway, the presence of these proteins can be used to select patients who may benefit from these new drugs. In addition to the IFN-I pathway, recent studies have associated other protein and/or gene profiles with the degree of disease activity in SLE. Therefore, a third strategy to optimize the treatment of SLE may involve combining the protein and gene profiles associated with disease activity.

In this thesis, through various studies, we will confirm that the aforementioned strategies can play a crucial role in optimizing the treatment of children with SLE.

In Chapter 1, we delineate the clinical characteristics of cSLE, emphasizing the absence of current diagnostic criteria for this condition and the utilization of various methods to assess disease activity. Additionally, we discuss the tools employed for measuring disease activity, flares, and disease-related damage. Furthermore, we address the latest therapeutic targets, with a focus on the T2T approach that has shown positive outcomes in adults with SLE. Regarding the pathogenesis, we discuss the role of various triggers in the onset of the disease and their contribution to its maintenance. We also discuss how the innate and adaptive part of the immune system collaborate and result in the so-called self-amplifying immunological loop, recognized as a hallmark of SLE pathogenesis. This chapter concludes by discussing the pivotal role of IFN-I. As previously mentioned, IFN-I is regarded as a key player in the pathogenesis of SLE. These proteins not only serve as promising candidate biomarkers for monitoring disease activity but are also employed as a therapeutic target in clinical trials, being inhibited by various medications.

The pharmacological treatment of SLE patients aims to achieve disease remission. Both clinical remission and the more recently formulated state of low disease activity, known as “Low Lupus Disease Activity State” (LLDAS), are therapeutic goals associated with a reduced risk of disease flares and organ damage. In Chapter 2, we demonstrate that LLDAS is an attainable treatment goal for children with SLE. In the Rotterdam cohort, all children achieved LLDAS within a median time of 6 months. We also demonstrate that the early introduction of immunosuppressants can limit the use of corticosteroids. Thus, the pursuit of LLDAS appears to be a feasible objective in cSLE and may potentially lead to improved long-term outcomes, including reduced organ damage and fewer disease flares.

While the clinical features of SLE provide valuable insights into disease activity and the occurrence of disease flares, there is a need for specific biological parameters that better reflect the underlying abnormal biological pathways. In Chapter 3.1, we demonstrate that a subset of patients with cSLE exhibits an elevated expression of the so-called IFN-I signature, which corresponds to increased levels of IFN-I proteins. Remarkably, in these patients, the expression of endoplasmic and cytosolic receptors responsible for IFN-I production appears to be elevated despite the use of various medications. We were able to inhibit the IFN-I signature by blocking the TBK-1 protein. This protein is part of the IFN-I signalling pathway. Therefore, TBK-1 may potentially serve as a novel therapeutic target in SLE in the future.

Nearly all SLE patients included in various scientific studies on the IFN-I signature to date were treated with one or more (immunosuppressive) medications. Interestingly, the majority of these patients still exhibited an elevated IFN-I signature despite these treatments. To investigate this, we established an in vitro model in Chapter 3.2 in which cells were activated through two IFN-I inducing routes, the endoplasmic and/or cytosolic pathways, followed by testing the effect of standard of care medications in SLE. As the tested medication appeared to have no effect on IFN-I production via the cytoplasmic pathways, the sustained production of IFN-I is likely occurring through these receptors.

Several studies suggest that the IFN-I signature is a potential biomarker for disease activity. However, multiple longitudinal studies have yielded conflicting results. Chapter 3.3 describes a comprehensive longitudinal cohort study. In this study, we assessed both the IFN-I signature and the concentration of IFN-I proteins. IFN-alpha2, one of the key IFN-I proteins, was measured using a highly sensitive Single-molecule assay called Simoa, which enables the measurement of very low concentrations of IFN-alpha2. Our results demonstrated that the IFN-I signature remained relatively stable over a three-year period and was not significantly associated with disease activity. However, we were able to show that the concentration of IFN-alpha2 in the blood, measured with Simoa, fluctuated over time and was associated with changes in disease activity. This highlights the importance of the method used to measure IFN-I activity for its applicability in the clinical setting as a biomarker for disease activity.

The IFN-I signature is not the only gene expression profile associated with SLE. Previous studies have demonstrated multiple gene expression profiles associated with cSLE. A combination of these profiles can be used to classify patients into more homogeneous groups based on their disease activity. However, the techniques used in the published landmark study are often too complex and costly for routine clinical practice. In Chapter 4.1, we translated these gene expression profiles into four simpler gene signatures that could be more easily applied in clinical practice. We demonstrated that a combination of these signatures could classify patients into groups with similar disease activity profiles. In a follow-up study, Chapter 4.2, these gene signatures were combined with specific proteins also associated with SLE. Subsequently, a computer model was used to classify patients into different groups based on these biological factors (gene expression and proteins). This approach resulted in three distinct clusters of patients with different disease activity profiles. Furthermore, we examined, through 40-color flow cytometry, whether the identified clusters were characterized by specific cell types in peripheral blood. Interestingly, multiple cell types were found to differ between these clusters. Together, these two studies emphasize the idea that future treatment decisions should not only be based on clinical characteristics but should also incorporate biological parameters.

In Chapter 5, we compare our findings with the existing literature. Several recent studies have demonstrated that achieving LLDAS is a feasible goal in cSLE. However, it is important to note that the current definition of LLDAS is tailored to adults. Thanks to the efforts of an international task force, definitions for children have now been established that still closely resemble those for adults, facilitating the comparability of future studies between childhood onset and adult onset SLE patients. While the results are predominantly positive for the T2T (Treat-to-Target) approach, there are still ideas to further refine this strategy. One of these concepts involves incorporating a measurable biological parameter into the T2T criteria. Ideally, this biomarker should be associated with disease activity, thereby enabling the achievement of the T2T strategy from a biological perspective. IFN-I appears promising in this regard, but candidates from multiple levels (genomics, transcriptomics, proteomics, etc.) seem to be emerging for this purpose. With the advent of new analysis methods, including artificial intelligence, research that combines biomarker-candidates from multiple levels appears to be even more relevant. However, the question remains whether this approach is feasible in practice, considering factors such as cost and technical complexity. Therefore, it remains the responsibility of researchers to develop techniques that are both cost-effective and feasible for clinical practice.