SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

Objective

To conduct a prospective cohort study using anakinra, a recombinant IL-1 receptor antagonist (IL-1Ra), as first-line therapy in patients with new-onset systemic juvenile idiopathic arthritis (JIA).

Methods

Therapy with recombinant IL-1Ra (2 mg/kg) was initiated in 20 patients who fulfilled the International League of Associations for Rheumatology criteria for systemic JIA, before systemic steroid treatment was administered. Patients were monitored clinically and immunologically. The protocol contained a stop strategy for patients who met at least the adapted American College of Rheumatology 90% criteria for improvement in JIA (ACR Pediatric 90 [ACR Pedi 90]) after 3 months.

Results

We included consecutive patients with new-onset systemic JIA. The mean followup period was 32 months (range 12–54 months). At the 3-month time point, 85% of the patients showed an adapted ACR Pedi 90 response or had inactive disease; 75% of the patients achieved this response while receiving recombinant IL-1Ra alone. After 1 year, 17 of the 20 patients met the criteria for clinically inactive disease, and 13 of these patients met these criteria while receiving monotherapy with recombinant IL-1Ra. However, because of persistent disease activity, 7 of the 20 patients required additional therapy besides recombinant IL-1Ra. According to our stop strategy, 73% of patients with at least an adapted ACR Pedi 90 response at 3 months could stop recombinant IL-1Ra treatment within 1 year. After 2 years, 12 (86%) of 14 patients met the criteria for disease remission, either while receiving (n = 4) or not receiving (n = 8) medication. After 3 years, 10 (91%) of 11 patients met the criteria for disease remission, either while receiving (n = 2) or not receiving (n = 8) medication.

Conclusion

This is the first prospective study in which recombinant IL-1Ra was used as first-line therapy in patients with systemic JIA. We observed excellent responses in nearly all patients within 3 months. In the majority of responding patients, treatment with recombinant IL-1Ra could be stopped within 1 year, with remission being preserved during followup. In approximately one-third of patients, concomitant therapy was required for maintenance of clinical response.

Systemic juvenile idiopathic arthritis (JIA) is a severe subtype of JIA that accounts for ∼10% of JIA cases ([1]). It is characterized by systemic inflammation and arthritis. This systemic inflammation is manifested through a characteristic spiking fever, skin rash, hepatosplenomegaly, lymphadenopathy, serositis (e.g., pericarditis), and elevated laboratory parameters of inflammation such as the erythrocyte sedimentation rate (ESR), the C-reactive protein (CRP) level, the thrombocyte count, and the ferritin level ([2]).

Clinically and immunologically, systemic JIA clearly differs from other subtypes of JIA. It has become clear that systemic JIA is an autoinflammatory disease rather than an autoimmune disease like the other subtypes of JIA ([2-6]). As such, the innate immune system seems to play a more prominent role in the pathophysiology of systemic JIA. Indeed, since the report by Pascual et al ([7]) that the interleukin-1 (IL-1) pathway is crucial in JIA and that blockade with recombinant IL-1 receptor antagonist (IL-1Ra; anakinra) yields beneficial clinical responses even in patients with longstanding disease, evidence for the involvement of especially the IL-1/IL-18 pathway and the IL-6 pathway has accumulated ([7-12]). However, IL-1 levels are not clearly increased in the plasma of patients with new-onset systemic JIA, whereas patients with active disease do display elevated plasma levels of IL-6 and especially IL-18, as well as elevated levels of S100 proteins (especially S100A8, S100A9, and S100A12) ([13-17]). Moreover, patients with systemic-onset JIA show deficits in the number and function of natural killer (NK) cells ([6, 18-21]), rendering these patients susceptible to the life-threatening complication of macrophage activation syndrome (MAS) ([19, 20, 22-25]).

Until recently, protocols for initial treatment in patients with systemic JIA relied on nonsteroidal antiinflammatory drugs (NSAIDs), corticosteroids, and methotrexate. Increasingly, biologic agents are being used, especially in steroid-resistant patients, with varying degrees of success. Unlike patients with polyarticular-onset JIA, patients with systemic-onset JIA, particularly those displaying clear systemic inflammation, do not respond very well to treatment with an anti–tumor necrosis factor α inhibitor such as etanercept ([26-29]).

Although multiple studies have shown clinical benefit of IL-1–targeted drugs such as recombinant IL-1Ra, not all patients respond equally well ([7, 9, 26, 30-32]). Very recently, 2 randomized placebo-controlled clinical trials showed a convincing benefit of both long-acting IL-1–blocking treatment (canakinumab) and IL-6–blocking therapy (tocilizumab) ([8, 12]). However, both trials included primarily patients with longstanding disease, and almost all of the patients had been treated or were still being treated with systemic steroids. Moreover, the protocols in these trials did not include a plan for tapering or stopping the drug once remission had been achieved.

The results of a retrospective multicenter study of recombinant IL-1Ra in systemic JIA ([33]) suggested that the timing of recombinant IL-1Ra therapy might be an important determinant of the response to this treatment. In that study, the 10 patients in whom recombinant IL-1Ra was started as first-line treatment (before the start of systemic steroids) had a relatively high response rate compared with patients with longstanding disease. This suggests that there might be a window of opportunity early in the disease course during which patients are more susceptible to a targeted therapy such as recombinant IL-1Ra. This concept was acknowledged in the 2011 American College of Rheumatology (ACR) recommendations for the treatment of patients with JIA, which allow initiation of recombinant IL-1Ra in patients with active fever and poor prognostic factors ([34]). Moreover, this concept is underscored by our own observation that in patients with new-onset systemic JIA treated with recombinant IL-1Ra, IL-18 levels normalized upon treatment, whereas in patients who received systemic steroids for longstanding disease, IL-18 levels remained increased even if the JIA appeared to be in clinical remission ([15, 21]). Therefore, we designed a clinical study in which patients with newly diagnosed systemic JIA were treated with recombinant IL-1Ra very early in the disease course (i.e., when they had failed to respond to NSAIDs but before the start of systemic steroids).

Our hypothesis for this study was that by targeting more specifically some key immunologic features of systemic JIA (both the IL-1 pathway and the deficient IL-18/NK cell axis) very early in the disease course, we would not only achieve high rates of response to recombinant IL-1Ra but would also be able to stop treatment in a substantial subset of patients. Accordingly, we included a stop strategy in our study, tapering recombinant IL-1Ra therapy at time point 3 months after the start of treatment if the patient had no fever and achieved inactive disease status while receiving recombinant IL-1Ra.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

Patients

In this prospective study, we treated 20 consecutive patients with new-onset systemic JIA who presented to our center between April 1, 2008 and January 1, 2012. Recombinant IL-1Ra was used as initial therapy, after failure to respond to indomethacin (an NSAID) but before the use of other disease-modifying antirheumatic drugs, systemic corticosteroids, or other biologic agents. For a diagnosis of systemic JIA, patients had to fulfill the International League of Associations for Rheumatology (ILAR) criteria ([35]). Prior to the start of recombinant IL-1Ra treatment, patients in whom systemic JIA was suspected were hospitalized for ≥3 days to confirm the diagnosis according to the ILAR criteria and to exclude other underlying causes and overt hemophagocytosis. Extensive evaluations were performed to determine the presence of infection (including blood cultures, serologic tests, polymerase chain reaction, purified protein derivative testing for tuberculosis) and malignancy, when appropriate (including screening blood tests, urinalysis [for neuroblastoma], radiologic and/or nuclear [positive emission tomograpy scanning] investigations for lymphoma, and bone marrow aspiration).

Treatment with indomethacin (an NSAID) was initiated in all patients, and 1 patient with systemic-onset JIA experienced significant improvement within 1 week (resolution of fever and clearly diminished inflammation and symptoms of arthritis). Therefore, this patient was not entered into the subsequent study protocol in which recombinant IL-1Ra was used as first-line therapy. In 20 patients with systemic JIA, no response to NSAIDs was observed after at least 1 week of treatment. Failure to respond to indomethacin was defined as the persistence of both fever and arthritis after 7 days of treatment. In these patients, recombinant IL-1Ra was started at a dosage of 2 mg/kg/day via subcutaneous injections, with a maximum dosage of 100 mg/day. Laboratory parameters that characterize systemic JIA, including blood counts, ESR, CRP, ferritin, S100A8/S100A9 serum complex (S100A8/9), S100A12, and soluble IL-2R levels, and cytokine profiles were assessed before the start of recombinant IL-1Ra treatment (day 0) and at time points 3 days, 30 days, 3 months, 6 months, 12 months, 24 months, and 36 months.

The clinical response to treatment was evaluated at the given time points during the first 6 months and every 3 months thereafter, according to the adapted ACR Pediatric 30 (Pedi 30), Pedi 50, Pedi 70, and Pedi 90 criteria for improvement in JIA ([36]). An adapted ACR Pedi 30 response is indicated by the absence of fever and at least 30% improvement from baseline in at least 3 of any of the 6 core set of variables, with no more than 1 of the remaining variables worsening by >30%; the adapted ACR Pedi 50, Pedi 70, and Pedi 90 are similarly defined (indicating 50%, 70%, and 90% improvement, respectively).

Clinically inactive disease was defined as no active arthritis, no systemic features, no uveitis, a normal ESR (≤20 mm/hour), and a physician's global assessment indicating no disease activity (defined as a score of ≤10 mm on a 100-mm visual analog scale) ([37]). Clinical remission “on medication” was defined as inactive disease for at least 6 months while treatment with recombinant IL-1Ra was ongoing ([38]). Clinical remission “off medication” was defined as clinical remission for at least 12 months without the use of medication ([38]). In patients showing an adapted ACR Pedi 90 response or clinically inactive disease while receiving recombinant IL-1Ra monotherapy at time point 3 months, an attempt was made to taper recombinant IL-1Ra, according to the stop strategy. These patients were given the same dose of recombinant IL-1Ra according to an alternate-day regimen during month 4, followed by a complete withdrawal of medication 1 month later (if at least an adapted ACR Pedi 90 response was maintained). Patients who experienced a recurrence of symptoms were switched back to the daily regimen, and tapering was attempted again 3–6 months later.

Patients were prospectively monitored for adverse events or side effects such as infection and local skin reactions. This study was approved by the University Medical Center Utrecht ethics committee (study no. 08-215), and informed consent was given by either the parents or guardians of the patients.

Extracellular cytokine production

Various cytokines in plasma were measured using a multiplex immunoassay as described previously ([15, 39]). For all assays, measurement and data analysis were performed using a FlexMap3D system in combination with Bio-Plex Manager software version 6.1.1, using a 5-parameter curve-fitting system (Bio-Rad).

Myeloid-related protein enzyme-linked immunosorbent assay (ELISA).

In all patients, serum S100A8/A9 and S100A12 levels were measured at University Children's Hospital Muenster, using a double-sandwich ELISA system as previously described ([40]). The readers of the laboratory assays were blinded with regard to the diagnosis and the inflammatory activity of the patients.

NK cell cytotoxicity assays

Cells from the NK cell–sensitive human erythroleukemia cell line K562 (ATCC) were maintained in RPMI 1640 supplemented with 2 mmoles/liter glutamine, 100 units/ml penicillin/streptomycin, and 5% (volume/volume) fetal calf serum at a concentration of 1 × 106 cells/ml. Peripheral blood mononuclear cells (4 × 105) were seeded in round-bottomed 96-well plates and precultured for 16 hours, with or without various concentrations of IL-18. The cytotoxicity of NK cells was assessed in standard 51Cr-release assays using NK cell–specific K562 cells as targets. Target cells were labeled with 100 μCi of Na251CrO4 (Amersham Biosciences). After 16 hours of preincubation, the cells were mixed with labeled target cells (4 × 104) and incubated for 4 hours in triplicate in a final volume of 200 μl. Supernatant was harvested and evaluated using an automated gamma counter. Maximum release of 51Cr was obtained by lysing the labeled target cells in 3% Triton X-100 (Sigma-Aldrich). Spontaneous 51Cr release was determined in wells containing labeled target cells incubated with medium alone. Results are shown as the percent release, which was calculated (in counts per minute) as follows: (test 51Cr release − spontaneous 51Cr release)/(maximum 51Cr release − spontaneous 51Cr release). Patient samples were always tested using healthy donor samples as a control. Low NK cell lytic function was arbitrarily defined as NK cell lysis of <15%.

Statistical analysis

Results are expressed as the median (interquartile range), when appropriate. All statistical analyses were performed using SPSS software version 12.0.1. P values less than 0.05 were considered significant.

RESULTS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

Effect of first-line recombinant IL-1Ra therapy on response rates and specific markers of inflammatory disease

In this prospective, uncontrolled, observational cohort study, we treated 20 consecutive patients with new-onset systemic JIA with recombinant IL-1Ra as first-line therapy, after failure of NSAID (indomethacin) monotherapy. Figure 1 shows a schematic diagram of the protocol and time line of the study during the first year. The protocol included a stop strategy, in which recombinant IL-1Ra treatment was tapered at time point 3 months in patients achieving at least an adapted ACR Pedi 90 response (treatment was tapered for 4 weeks and stopped thereafter). All patients were followed up for at least 12 months, with a mean followup after starting recombinant IL-1Ra of 32 months (range 12–54 months). The mean number of arthritic joints prior to the start of recombinant IL-1Ra treatment was 4.4 (range 1–14). The clinical data, as well as laboratory or immunologic parameters of disease activity at the time that recombinant IL-1Ra was started, are shown in Table 1.

image

Figure 1. Schematic diagram showing the study protocol and response to treatment during the first year. The white boxes indicate patients who achieved and maintained at least an adapted American College of Rheumatology Pediatric 90 (ACR Pedi 90) response. The shaded boxes indicate patients in whom additional therapy besides recombinant interleukin-1 receptor antagonist (rIL-1Ra) was required to achieve or maintain a clinical response. SJIA = systemic juvenile idiopathic arthritis; NSAID = nonsteroidal antiinflammatory drug; MTX = methotrexate.

Download figure to PowerPoint

Table 1. Characteristics of the patients with JIA at the start of recombinant IL-1Ra therapy*
  1. Except where indicated otherwise, values are the median (interquartile range). JIA = juvenile idiopathic arthritis; IL-1Ra = interleukin-1 receptor antagonist; CRP = C-reactive protein; ESR = erythrocyte sedimentation rate; NK = natural killer.

No. male/no. female13/7
Age at disease onset, mean (range) years7.9 (1.1–15.3)
No. of arthritic joints at diagnosis, mean (range)4.4 (1−14)
Exanthema present at diagnosis, %85
Overt serositis present at diagnosis, %10
CRP, mg/liter (normal 0−10)176 (126−268)
ESR, mm/hour (normal 0−12)130 (108−140)
Platelets, ×109/liter (normal 150–450)575 (409−772)
Ferritin, mg/liter (normal <250)798 (233−2,351)
Soluble IL-2R, units/liter (normal <600)1,890 (1,054−3,354)
IL-18, pg/liter5,399 (1,837−8,923)
NK cell function, % lysis7.1 (3.3−13.0)
S100A12, ng/ml11,895 (5,075−19,250)
S100A8/9, ng/ml32,235 (19,800−82,800)

Clinically, recombinant IL-1Ra treatment was very effective, resulting in normalization of body temperature within 3 days in 18 (90%) of 20 patients. We observed normalization of the classic parameters of inflammation, such as the CRP and ferritin levels and the ESR, and achievement of an adapted ACR Pedi 90 response in 80% of patients receiving recombinant IL-1Ra monotherapy, within 30 days of treatment (Table 2 and Figure 2). We also observed normalization of the more specific parameters of (subclinical) disease activity in systemic JIA, namely IL-18, S100A8/A9, and S100A12, in virtually all responders at time point 3 months (Table 2). Therefore, first-line therapy with recombinant IL-1Ra resulted in high response rates and normalization of (subclinical) inflammation in most patients within 3 months.

Table 2. Circulating markers of inflammation in the plasma or serum of patients with juvenile idiopathic arthritis during recombinant IL-1Ra treatment*
Time pointStatus at followup (no. of patients)CRP, mg/literFerritin, μg/literIL-18, pg/literS100A12, ng/mlS100A8/9, ng/ml
  1. Twenty patients were evaluated at all time points except 2 years (n = 14) and 3 years (n = 11). Patients receiving rescue therapy required the initiation of concomitant steroids/disease-modifying antirheumatic drugs or a switch to other biologic agents due to an insufficient response to recombinant interleukin-1 receptor antagonist (IL-1Ra) therapy alone. Values are the median (range). CRP = C-reactive protein; NA = not applicable.

  2. a

    One patient died of macrophage activation syndrome 2 years after the start of recombinant IL-1Ra therapy and subsequent failure to respond to steroids, tocilizumab, and canakinumab.

Prior to start of treatmentNA176 (56–410)798 (111–15,384)5,399 (395–29,350)11,895 (1,395–146,000)32,235 (4,130–497,000)
Day 3Responder (n = 18)53 (13–248)401 (85–3,088)3,760 (40–23,710)22,160 (7,210–122,249)7,115 (2,150–26,300)
 Rescue therapy (n = 2)204 (110–298)2,312 (201–4,424)2,522 (33–5,011)NANA
Day 30Responder (n = 16)2 (1–46)81 (17–845)1,212 (23–8,509)210 (23–1,300)875 (320–4,120)
 Rescue therapy (n = 4)21 (2–85)240 (157–2,804)2,838 (764–8,133)3,058 (180–8,445)7,015 (1,730–92,175)
Day 90Responder (n = 15)2 (2–81)19 (9–340)71 (7–3,387)145 (15–480)920 (340–2,630)
 Rescue therapy (n = 5)2 (2–8)39 (23–74)3,470 (1,541–5,985)270 (190–26,900)1640 (1,340–68,700)
1 yearResponder (n = 13)6 (2–42)37 (4–63)73 (37–287)72 (58–241)379 (360–1,450)
 Rescue therapy (n= 7)3 (1–53)27 (24–41)950 (300–2,760)312 (40–1,060)2,323 (380–6,970)
2 yearsResponder (n = 8)5 (2–8)24 (16–36)19 (9–81)97 (11–370)546 (290–2,410)
 Rescue therapy (n = 6)2 (1–23)44 (9–57)2,062 (414–3,711)480 (120–7,680)3,010 (535–34,780)
3 yearsResponder (n = 7)3 (2–5)24 (20–30)35 (15–115)120 (100–220)770 (370–1,220)
 Rescue therapy (n = 3)a2 (1–15)24 (18–38)NANANA
image

Figure 2. Adapted American College of Rheumatology Pediatric 70 (ACR Pedi 70) and Pedi 90 responses at 0, 3, 6, 12, 24, and 36 months. One-year followup data were available for 20 patients, 2-year data were available for 14 patients, and 3-year data were available for 11 patients. Scores for the entire cohort (including patients requiring additional therapy) are shown. Note that adapted ACR Pedi 70 and Pedi 90 scores as well as adapted ACR Pedi 70 and Pedi 90 scores for patients receiving recombinant interleukin-1 receptor antagonist (IL-1Ra) treatment only (mono Tx) overlap. Arrows indicate the time points at which an individual patient needed concomitant therapy (n = 7). After 30 days of treatment, 80% of the patients achieved an adapted ACR Pedi 90 response while receiving recombinant IL-1Ra monotherapy.

Download figure to PowerPoint

Maintenance of the high initial rate of response to first-line therapy with recombinant IL-1Ra in the vast majority of patients during >2 years of followup

At time point 12 months, 90% of patients (18 of 20) showed at least an adapted ACR Pedi 70 response, and 85% of patients (17 of 20) showed an adapted ACR Pedi 90 response (Figure 2). These 17 patients also met the criteria for clinically inactive disease. Two-thirds of included patients (13 of 20) achieved clinically inactive disease status while receiving monotherapy with recombinant IL-1Ra; in 11 of these patients, tapering and stopping medication were possible within the first year (Figure 1).

At the 2-year time point, 86% of the patients (12 of 14) achieved an adapted ACR Pedi 90 response and also met the criteria for clinical remission (either on medication [n = 4] or off medication [n = 8]). At the 3-year followup 10 of 11 patients (91%) achieved an adapted ACR Pedi 90 response and also met the criteria for clinical remission either on medication (n = 2) or off medication (n = 8). Taken together, these data showed high response rates in patients with systemic JIA who received recombinant IL-1Ra therapy early in the disease course, with high rates of sustained remission at 2 and 3 years of followup.

Additional treatment in patients with an incomplete response to first-line therapy with recombinant IL-1Ra

Table 3 summarizes the treatment history of 7 patients who failed to have a complete response to monotherapy with recombinant IL-1Ra during followup. Interestingly, in all of the patients who required additional therapy besides recombinant IL-1Ra, the requirement occurred within the first year of followup. After the start of recombinant IL-1Ra, only 2 of the 20 patients showed persistent fever and systemic inflammation on day 3. In these patients, the daily dose of recombinant IL-1Ra was increased (up to 4 mg/kg). However, because of persistent inflammation and/or fever, both patients required treatment with systemic steroids (1 mg/kg/day) after day 7, in addition to recombinant IL-1Ra. Steroid treatment was stopped in both patients within 6 months. In one of these patients, recombinant IL-1Ra treatment could be stopped 2 months later; however, she experienced a relapse more than 1 year after the initial start of recombinant IL-1Ra. Interestingly, she did not experience a response to long-acting IL-1 blockade with canakinumab either, but combination treatment with systemic steroids (1 mg/kg) and MTX resulted in improvement. She is now in clinical remission while being treated with MTX alone. The other patient, who had persistent fever at time point day 3, is now in remission while receiving recombinant IL-1Ra only on an alternate-day regimen (steroids were tapered and then stopped).

Table 3. Additional treatment in patients with an incomplete response to first-line therapy with recombinant IL-1Ra*
PatientTime point of flareSteroids started following recombinant IL-1Ra (daily dose)Other DMARD or biologic agentResponse at last followup
  1. IL-1Ra = interleukin-1 receptor antagonist; DMARD = disease-modifying antirheumatic drug; MTX = methotrexate; ACR Pedi 90 = American College of Rheumatology 90% improvement criteria for juvenile rheumatoid arthritis; MAS = macrophage activation syndrome.

8Month 9NoMTXRemission while not receiving medication
10Month 6Yes (0.3 mg/kg)CanakinumabAdapted ACR Pedi 90 response while receiving canakinumab
11Day 3Yes (1 mg/kg)NoRemission while receiving recombinant IL-1Ra every other day
12Month 2Yes (1 mg/kg)Tocilizumab (failed) and canakinumab (failed)Died of MAS and pulmonary hypertension
13Month 1Yes (0.3 mg/kg)NoRemission while not receiving medication
14Month 1Yes (1 mg/kg)Canakinumab (failed), subsequently MTXRemission while receiving MTX
17Month 1Yes (1 mg/kg)CanakinumabAdapted ACR Pedi 90 response while receiving canakinumab

In addition, we observed 4 relapses in the first 6 months after the start of recombinant IL-1Ra and 1 relapse (mild arthritis only) after discontinuation of recombinant IL-1Ra (9 months after inclusion in the study). All of these patients showed at least a partial response to recombinant IL-1Ra during the first months of treatment, and 3 of them showed persistent disease activity while receiving combination therapy with low-dose steroids and recombinant IL-1Ra. These 3 patients were switched to treatment with other biologic agents, and 2 of them experienced a good response to canakinumab. However, the third patient had only a partial initial response to tocilizumab after 1 year of treatment and a moderate initial response to canakinumab after 3 months of treatment. MAS developed in this patient after a viral infection and pulmonary hypertension of unknown origin, and this patient died of MAS 2 years after inclusion in our study ([41]). Thus, 6 of 7 patients with systemic JIA in whom additional therapy besides recombinant IL-1Ra had to be started showed at least adapted ACR Pedi 90 responses at the time of the last followup (mean followup 32 months).

Maintenance of clinically inactive disease/disease remission after discontinuation of recombinant IL-1Ra therapy

Fifteen (75%) of the 20 patients showed at least an adapted ACR Pedi 90 response to recombinant IL-1Ra alone at time point 3 months (Figure 1). These patients then entered a tapering protocol for recombinant IL-1Ra, in which recombinant IL-1Ra was administered according to an alternate-day regimen for 4 weeks and stopped thereafter when disease was still in remission (Figure 1). In 7 of 15 patients, this tapering protocol was successful when it was started after 3 months. Eight of 15 patients experienced a relapse of fever and/or inflammation or arthritis symptoms during the tapering phase. A daily regimen of recombinant IL-1Ra was restarted, and a subsequent attempt was made to taper and stop this treatment. Within 1 year of starting treatment, recombinant IL-1Ra could be tapered and stopped in another 4 patients. Thus, 11 (73%) of 15 patients treated with recombinant IL-1Ra who achieved inactive disease status at time point 3 months could stop receiving this treatment within the first year.

Interestingly, patients in whom treatment could be tapered and stopped successfully after 3 months showed lower levels of IL-18, S100A12, and S100A8/9 at time point 3 months, prior to tapering, compared with patients in whom treatment could not be successfully tapered at time point 3 months (the difference for S100A12 was significant [P < 0.01], and a trend for significance was observed for IL-18 and S100A8/9). There was no significant association between any marker of inflammation at the time of diagnosis and the clinical response at 3 months. Although the numbers in this study are small, these data suggest that failure to successfully taper recombinant IL-1Ra might be predicted by persistently elevated levels of IL-18 and S100A proteins in the first months after the start of this treatment.

Side effects and (severe) adverse events

During followup, no serious side effects were observed during treatment with recombinant IL-1Ra. Local skin reactions were reported in 13 of 20 patients. No serious invasive infection was reported, although mild cutaneous or upper airway infection and reactivation of infection with herpes simplex virus type 1 were reported in several patients, none of whom required hospitalization or intravenous antibiotic treatment.

DISCUSSION

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

The discovery that the serum of patients with systemic JIA can induce an IL-1 signature in the cells of healthy individuals and the subsequent successful intervention with recombinant IL-1Ra in these patients were major breakthroughs in the understanding of the immune pathogenesis of systemic JIA ([7]). These findings underscored the concept that systemic JIA is an acquired, IL-1β–mediated autoinflammatory disease. Since the publication of this key finding, 2 randomized controlled clinical trials and several case series have shown beneficial effects of short-acting (recombinant IL-1Ra) ([7, 9, 30-33]) and long-acting (canakinumab) ([12, 42]) IL-1–blocking agents.

However, not all patients with systemic JIA respond to IL-1 blockade ([9, 30-32]). A possible explanation for this heterogeneous response to IL-1 blockade is that patients with systemic JIA constitute a heterogeneous patient population ([9]). Another explanation for the different responses to recombinant IL-1Ra therapy reported so far could be the timing of therapy. Nearly all of the patients with systemic JIA in these studies had longstanding disease and were still receiving systemic steroids when treatment with IL-1–blocking agents was initiated. Indeed, Nigrovic et al retrospectively described a small number of patients (n = 10) with systemic JIA in whom recombinant IL-1Ra was started as first-line therapy ([33]). The response rate in these 10 patients appeared to be higher than that in patients in their series with longstanding disease who had previously received steroids and/or DMARDs or other biologic agents.

Herein we describe the results of early treatment with recombinant IL-1Ra in a prospectively studied cohort of 20 consecutive patients with new-onset systemic JIA at a tertiary pediatric rheumatology referral center. Our hypothesis was that by targeting more specifically some key immunologic features of systemic JIA (both the IL-1 pathway and the deficient IL-18/NK cell axis) very early in the disease course, we not only would achieve high rates of response to recombinant IL-1Ra but also would be able to stop treatment in a substantial subset of patients. We observed that steroid-naive patients with new-onset systemic JIA treated with recombinant IL-1Ra alone as first-line therapy showed a rapid clinical response, with fever ending within days and normalization of parameters of inflammation (CRP, ESR, S100A proteins, ferritin, soluble IL-1R), cytokine levels (mainly IL-18), and arthritis manifestations in 80% of patients within 30 days. These data extend the findings of Nigrovic et al ([33]), confirming high rates of response to recombinant IL-1Ra when therapy is started early in the disease course in steroid-naive patients. Our observed response rates appear to be higher than those observed in published trials of recombinant IL-1Ra in patients with longstanding systemic JIA ([30, 31, 33]) and in recently published double-blind randomized trials of long-acting IL-1 blockade in patients with relatively longstanding disease ([12, 42]).

Taken together, our results show that the strategy of using recombinant IL-1Ra as a first-line therapy in patients with systemic JIA and adding treatment only when full remission is not achieved results in very high and sustained rates of adapted ACR Pedi 90 responses or inactive disease. More than 80% of the patients in our cohort achieved persistent disease remission, either on or off medication, during a mean followup of 2 years and 8 months.

Because our cohort comprised 20 consecutively presenting patients with new-onset systemic JIA (with an insufficient response to indomethacin alone), and because both the clinical and immunologic data at disease onset showed severe inflammatory disease, we believe that our cohort was not biased by selection.

Importantly, our study tested a stop strategy in patients with disease in remission at time point 3 months after the start of recombinant IL-1Ra. The majority of these patients (11 of 15) ultimately could stop treatment with recombinant IL-1Ra in the first year, with sustained remission during followup. This finding, as well as the rather rapid clinical responses to recombinant IL-1Ra (in our experience, the response to recombinant IL-1Ra is faster than the response to systemic steroids, thus shortening the time spent in the hospital), do not support the claim that treatment with recombinant IL-1Ra is more expensive than treatment with nonbiologic agents such as steroids and MTX.

The fact that especially IL-18 levels (and also S100A12 and S100A8/9 levels) were still increased at time point 3 months (i.e., the moment that tapering of recombinant IL-1Ra was started) in 4 of 8 patients in whom tapering initially was not successful, whereas the IL-18 level was only slightly increased in 1 of 7 patients in whom tapering was successful, suggests that IL-18 and/or S100A12 or S100A8/9 might serve as biomarkers guiding the strategy for stopping recombinant IL-1Ra treatment in patients with systemic JIA. This possibility was previously suggested for S100A8/9 ([43]) and IL-18 ([42-46]) but should be prospectively validated in other (and preferably larger) patient cohorts.

Compared with historical (“pre-biological”) data from 2 other cohorts, our data are remarkably good. A retrospective Italian cohort study (n = 80 patients) published in 2000 showed that the course of systemic JIA was monocyclic in ∼15% of patients, persistent in ∼33% of patients, and intermittent in ∼50% of patients ([47]). Another Canadian cohort study (n = 111 patients) that was also published in 2000 showed persistent disease activity and the need for medication in >60% of patients at 6 months after disease onset ([48]).

In our cohort, treatment with recombinant IL-1Ra early in the disease course appeared to be rather safe; no severe adverse events attributable to recombinant IL-1Ra were observed during the followup period. However, our cohort comprised only 20 patients. Meta-analyses, such as that conducted by Swart et al ([32]), and particularly international register studies on the safety of (biologic) treatments are necessary to address this issue in a more in-depth manner.

Taken together, the results of our study support the concept of treating patients with systemic JIA with recombinant IL-1Ra early in the disease course, before treatment with systemic steroids. Currently, it is not possible to individualize treatment with, for example, IL-1–blocking agents, IL-6–blocking agents, or systemic steroids based on immunologic or clinical parameters/biomarkers at disease onset. However, our findings indicate that S100 proteins and/or IL-18 in plasma are candidate biomarkers to guide therapeutic regimens. To definitively answer the question of which drug should be used as first-line therapy in systemic JIA, a head-to-head multicenter randomized controlled trial (IL-1 blocking versus IL-6 blocking versus steroids) with a sufficient number of patients and including (promising) biomarkers is warranted.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Vastert had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Vastert, Kuis, Prakken, Wulffraat.

Acquisition of data. Vastert, de Jager, Noordman, Holzinger, Wulffraat.

Analysis and interpretation of data. Vastert, de Jager, Noordman, Holzinger, Kuis, Prakken, Wulffraat.

REFERENCES

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES