To the best of our knowledge, this is the most extensive pediatric case seriesdescribing patients with STAT1 GOF mutations under JAKinib therapyto date. Our study provides a detailed description of the clinical experience with this treatment approach in children and highlights the heterogeneity in terms of indications, dosing schedules and follow-up practices.
Infections were common in our cohort, correlating well with previous reports [3,8,9,10,16,17]. Before starting JAKinibs, most infections had been controlled; only P3 suffered fromCMV stomatitis. The previously published cases showed overall a more severe phenotype, with higher prevalence and severity of autoimmune manifestations and failure to thrive (Table 1). In addition, these patients had also received other immunosuppressive drugs.Thus, in most of themJAKinibs were not used as “first-line” therapy. In our series, JAKinibs were initiated at earlier disease stages, possiblyreflecting the positive experiences reported in the previous studies [3,8,9,10,16,17].
Reasons to start JAKinibs stated by the attendingphysicians of our cohort were similar to previous reports [3,8,9,10,16,17]. Beyond CMC, these included refractory autoimmune complications, progressive vasculopathy and lung disease (Table 1, Table S1). Furthermore, one patient received ruxolitinib for four months as a bridge to a subsequent HSCTprocedure[8, Table S-2].
Treatment, dosing and treatment response
In the setting of IEI, the appropriate dosing and interval remains to be established, as experience with these small molecule inhibitors in the pediatric age is very limited. Whilst the European Medicine Agency (EMA) has not yet approved ruxolitinib in children , the Food and Drug Administration (FDA) indicates their use for steroid-refractory acute graft-versus-host disease (GVHD) in children older than 12years of age in 2019 (recommended dose 5mg every 12 hours). 50 mg/m2/dayhas been indicated to be the maximum well-tolerated dose in children . Based on serial drug level determination and functional assays, 8 hours dose intervals have been recently suggested in a child with STAT3 GOF mutation associated with immune dysregulation (type 1 diabetes mellitus and interstitial lung disease). Interestingly, the dose needed and tolerated in this case report was high (2.2 mg/kg/day), being more than twice the dose compared to previous reports [8,21]and those used in our own cohort (see Table S-1 and S-2).
In our patients,ruxolitinib was used in 9/10 and barcitinib in 1/10 child, respectively. The attending physicians preferred ruxolitinib, given the larger literary experience in STAT1 GOF setting. We provide detailed dosing information (Table 1 and Table S-1) for our patients, including the starting and maximum doses. Our starting (0.28mg/kg/day vs 0.8mg/kg/day) and maximum doses (0.6 mg/kg/day vs 1.05 mg/kg/day) were lower than previously reported [3,8,9,10,16]. However, the absence of homogenous protocols(and dosing reported diversely as mg/kg/d, mg/day and/or m2/day) in the literature limits conclusive comparisons and should beunified in future studies.Dose adjustments in our cohort were performed mainly according to the clinical effect, absence of adverse events and in three cases supported by functional analysis using pSTAT1 stimulation assays (P3, P4, P9).
Collectively, CMC was the most prevalent disease manifestation (n=23) and JAKinib treatment was effective in almost all patients (overall response rate 20/22, Figure 1) within 2-8 weeks of treatment.Contrastingly, Acker et al. recently described a patient with only transient responses to JAKinib, administeredfor CMC, enteropathy and cytopenia .Importantly, in our cohort, the only patient receiving baricitinib did not show clinical improvement resulting in its discontinuation and switch to ruxolitinib.
In the absence of controlled prospective data, we suggest starting pediatric patients on 0.3-0.5mg/kg/day of ruxolitinibtwice per dayand then progressively increasing the dose by 0.1-0.2mg/kg/day every 2-4 weeks until achieving the expected clinical effect or occurrence of relevant side effects keeping in mind the suggested maximum dose of 50 mg/m2/day by Lohet al..
For the clinician, the patients and family it is important to know when to expect the Jakinib treatment to take effect. In our cohort, the cytopenias and CMC responded rather promptly (1-8 weeks), whereas others, such as keratitis and autoimmune hepatitis, required prolonged treatment courses (4-8 months). No improvement or worsening of cerebral aneurysms was observedin two patients. Unfortunately, the information available in the literature regarding treatment responses is often unspecific and incomplete. Where such data were provided,the time to response wassimilar to what was observed in our cohort requiring several weeks of therapy to achieve improvement [Table S-1].
Despite the combined data presented here, the number of pediatric STAT1 GOF patients treated with JAKinibs is still small. Furthermore, it is likely that the time to response might vary depending on the organ involved, severity and duration of the diseaseand JAKinibdosage. Therefore, larger, detailed, and prospective patient cohorts will need to address these aspects more consistently.
Baricitinib, a potent JAK1/JAK2 inhibitor has shown good tolerability in rheumatologic diseases and other monogenic interferonopathies [22,23]. To date,one case report indicated efficacy in an adult patient with STAT1 GOF suffering from recurrent aphthae, as well as oral and esophageal CMC.Contrastingly, in our cohort P2 failed to show any improvement after two months of treatment with 4mg/m2. Curiously, upon switching to ruxolitiniba fast, completeand sustained remission of CMCand partial remission ofaphthae after 3 months of treatment were observed. Whether baricitinibis inferior or not in the control of the disease manifestations in STAT1 GOF compared to ruxolitinib remains to be determined.
Assessing disease activity using immune deficiency and dysregulation activity (IDDA) score
The IDDA score is a promising tool to assess disease activity and burden in the setting of immune dysregulatory diseases [14,15]. It allows for intraindividual, longitudinal monitoring by using a number of relevant clinical parameters and has been added as a voluntary option to the European Society for Immunodeficiencies (ESID) registry . We applied the score for the first time to patients with STAT1 GOF obtaining lower numbers (15.99) when compared to those reported for lipopolysaccharide (LPS)-responsive and beige-like anchor protein (LRBA) deficient patients proceeding to transplant (32.9) or those remaining under conventional immunosuppressive therapy (20.8) (Table 1, ). A significant reduction in the IDDA score was observed after initiation of JAKinib therapy for all patients with initial IDDA score >10, suggesting a substantial decline in the disease activity due to JAKinib introduction.
Adverse events and monitoring
Overall, the occurrence of adverse events potentially related to JAK inhibition were rare in our cohort. In fact, only one patient experienced an increased frequency of bacterial infections. Contrastingly, the reports in the literature for STAT1 GOF on JAKinibmention higher rates of urinary infections [26,27] and other less frequent infectious complications, such as herpes virus reactivation ,tuberculosis and/or other atypical mycobacterial infections [26,27,28], JC virus (four fatal cases)[29,30,31,32], pneumocystis jirovecii, hepatitis B [34,35] and toxoplasmosis . This discrepancy might be attributed to an earlier introduction of JAKinibs in our cohort compared to their predominant use asa rescue strategy following the failure of other immunosuppressive regimens in the previously reported cases [3,8,10,16,17].
Although no published guidelines exist, we observed a surprisingly consistent approach chosen by the individual participating centers in terms of investigations performed prior and during JAKinib therapy (Figure S1, Table S1). These parameters most likely reflect concerns based on the published experience with JAKinibs in other scenarios, such as myelofibrosis, arthritis and graft versus host disease (GVHD), as well as STAT1 GOF cases [26,27], FDA and EMA recommendations [18,19]. They include screening for infectious complications and monitoring for organ toxicity.In the absence of an easy-to-perform assay to determine ruxolitinibserum levels and the lack of the well-defined correlation between drug levels and clinical response, other biomarkers have been explored to monitor the drug effect/clinical response, such as phosphorylated STAT1 levels (pSTAT1) and IL17 production in T lymphocytes (TH17). Whilst some studies suggest a correlation between normalization of these markers [3,11,37],others reported a clear discrepancy . This might be due to differences in timing of sampling, sample preparation and assay protocols. In future studies, harmonized treatment and monitoring protocols are needed toconsistently evaluate the role of these and other biomarkers in patients with IEI underJAKinib therapy.
In our cohort drug levels were not performed.All participating centers stated an overall interest to perform JAKinibs level testing but did not have test availability at their institutions.
Importantly, none of the patients described here experienced severe adverse events such as thromboembolism or pulmonary hypertension. Interestingly,one patient (P4), who was started on ruxolitinib despite suffering from pulmonary hypertension, showed a marked improvementallowingthe reduction of chronic medication forpulmonary hypertension, as well as the suspension oflong-term oxygen supplementation.
Our recommendation prior to starting the JAK inhibition in pediatric patients with STAT1 GOF is toobtain a complete medical history,aimingto identify previous, active or chronic infections and potential underlying organ damage. We also suggest applying early and extensive diagnostic and therapeutic strategies when suspecting viral, bacterial and/or fungal infections including blood, urine, stool, aspirate samplesand biopsies from affected tissues/organs, if indicated, to minimize the risk of severe and preventable infectious complications.
In the specific setting of JAK inhibition in (pediatric) STAT1 GOF patients, the role of primary or secondary antimicrobial, antiviral and antifungal prophylaxis remains to be established. Most authors suggest antimicrobial prophylaxis in patients with recurrent (respiratory) infections. Systematic prevention of herpes virus infections is more controversial but should be considered in those patients with a history of systemic infection, severe lymphopenia as well as a history of long-term immunosuppression. In our cohort, immunoglobulin replacement therapy, antimicrobial and antiviral prophylaxis were prescribed according to the initial immunological work-up andwere not part of a specific strategy to prevent infections under JAKinib therapy.