Canakinumab for the treatment of acute flares in difficult-to-treat gouty arthritis: Results of a multicenter, phase II, dose-ranging study

Authors


  • ClinicalTrials.gov identifier: NCT00798369.

Abstract

Objective

To assess the efficacy and tolerability of canakinumab, a fully human anti–interleukin-1β monoclonal antibody, for the treatment of acute gouty arthritis.

Methods

In this 8-week, single-blind, double-dummy, dose-ranging study, patients with acute gouty arthritis whose disease was refractory to or who had contraindications to nonsteroidal antiinflammatory drugs and/or colchicine were randomized to receive a single subcutaneous dose of canakinumab (10, 25, 50, 90, or 150 mg; n = 143) or an intramuscular dose of triamcinolone acetonide (40 mg; n = 57). Patients assessed pain using a 100-mm visual analog scale.

Results

Seventy-two hours after treatment, a statistically significant dose response was observed for canakinumab. All canakinumab doses were associated with numerically less pain than triamcinolone acetonide; thus, a dose with equivalent efficacy to triamcinolone acetonide 72 hours after treatment could not be determined. The reduction from baseline in pain intensity with canakinumab 150 mg was greater than with triamcinolone acetonide 24, 48, and 72 hours after treatment (differences of −11.5 mm [P = 0.04], −18.2 mm [P = 0.002], and −19.2 mm [P < 0.001], respectively), and 4, 5, and 7 days after treatment (all P < 0.05). Canakinumab significantly reduced the risk of recurrent flares versus triamcinolone acetonide (P ≤ 0.01 for all doses) (relative risk reduction 94% for canakinumab 150 mg versus triamcinolone acetonide). The overall incidence of adverse events was similar for canakinumab (41%) and triamcinolone acetonide (42%); most were mild or moderate in severity.

Conclusion

Our findings indicate that canakinumab 150 mg provides rapid and sustained pain relief in patients with acute gouty arthritis, and significantly reduces the risk of recurrent flares compared with triamcinolone acetonide.

Gout is a metabolic disease caused by monosodium urate monohydrate (MSU) crystal deposition and is one of the most common forms of inflammatory arthritis in adults (1, 2). Its prevalence increases with age and is especially high in older men (2). Gouty arthritis commonly occurs in the lower extremities (3, 4), and flares are characterized by rapid onset and build-up of pain, warmth, swelling, decreased range of motion, and redness of the involved joints (3, 4). Initial flares last hours to weeks, whereas subsequent flares may be more prolonged. The frequency of flares often increases over time in patients who are inadequately treated (3, 4).

The goal of therapy in an acute gout flare is prompt and safe termination of pain and inflammation. Acute gouty arthritis is usually treated with nonsteroidal antiinflammatory drugs (NSAIDs), colchicine, or corticosteroids (5, 6). However, current treatments are not always effective (7), and many patients with gouty arthritis have underlying comorbidities, including renal function impairment, cardiovascular disease, diabetes mellitus, hypertension, and gastrointestinal disorders, that restrict treatment options because standard therapies may worsen these comorbidities (5, 6, 8, 9). There is thus a need for effective treatments for patients in whom standard antiinflammatory therapies are not advised.

MSU crystals induce secretion of interleukin-1β (IL-1β) via activation of the NALP3 inflammasome (10). IL-1β is a pivotal cytokine that mediates inflammation and joint damage. Several lines of evidence suggest that the initiation of inflammation in gout is similar to that occurring in hereditary autoinflammatory syndromes such as cryopyrin-associated periodic syndrome (CAPS) (1, 10, 11), for which IL-1β blockade is highly effective (12–14). Therefore, IL-1β may be an appropriate therapeutic target for acute gouty inflammation. This is supported by preliminary clinical data in gout (15) and pseudogout (16, 17) on anakinra, an antagonist of the IL-1β receptor, and rilonacept (18, 19), another selective inhibitor of IL-1β signaling.

Canakinumab (ACZ885) is a fully human monoclonal antibody with a long half-life (21–28 days) (20); it provides potent and selective blockade of IL-1β signaling (21) and induces rapid and sustained complete remissions in patients with CAPS (12). Herein, we describe the results of a multicenter, phase II, dose-ranging study performed to determine an appropriate dose of canakinumab for patients with acute gouty arthritis whose disease is refractory to or who have contraindications to NSAIDs and/or colchicine.

PATIENTS AND METHODS

Study design.

This was an adaptive, single-dose, single-blind, double-dummy, dose-ranging, active-controlled study that was approved by each local independent ethics committee. It was performed in accordance with the International Conference on Harmonisation Harmonised Tripartite Guidelines for Good Clinical Practice and the ethics principles of the Declaration of Helsinki. All patients provided written informed consent. This clinical trial was registered with www.clinicaltrials.gov (registration number NCT00798369). Members of the Canakinumab in Gout Study Group are listed in Appendix A.

Patients were screened at the time of their acute gout flare. Eligible patients were subsequently randomized and received canakinumab at 1 of 5 doses (10, 25, 50, 90, or 150 mg) by subcutaneous injection and saline by intramuscular injection or the corticosteroid triamcinolone acetonide (40 mg) intramuscularly and a subcutaneous placebo injection on day 1. Randomization was carried out by means of an interactive voice response system.

Patients recorded pain intensity at prespecified time points and use of rescue medication during the first 7 days of the study in diaries. After the 6-hour-postdose pain assessment, patients could take rescue medication (consisting of prednisone up to a maximum dose of 30 mg once daily for up to 5 days and/or acetaminophen up to a maximum of 1 gm/dose or 3 gm/day) as needed during the first 7 days but not in the 4 hours before a pain assessment. Patients returned to the study center on days 4 (72 hours after study drug administration), 8, 29, and 57 for efficacy and safety assessments. Patients were blinded with regard to treatment.

Patients.

The study enrolled patients ages 18–80 years with a history of at least 1 previous gout flare who met the American College of Rheumatology 1977 preliminary criteria for acute gout (22). Patients were also required to have had an acute gout flare for ≤5 days, have a baseline pain intensity of ≥50 mm on a visual analog scale (VAS) ranging from no pain (0 mm) to unbearable pain (100 mm), have disease that was refractory to or have contraindications to NSAIDs and/or colchicine according to their treating physician, and have a body mass index (BMI) of ≤40 kg/m2. Patients receiving urate-lowering therapy were required to be on a stable dose regimen and were expected to remain on this regimen throughout the study.

Key exclusion criteria included the use of any of the following medications within specified periods before screening: ibuprofen, acetaminophen, aspirin, diclofenac, naproxen, cyclooxygenase 2 inhibitors, other NSAIDs, systemic or intraarticular corticosteroids, anakinra, rilonacept, any tumor necrosis factor inhibitor, or use of >1 dose of 0.6 mg colchicine in the 24 hours before screening, if not taking a stable dose. Patients were excluded if they had rheumatoid, infectious/septic, or other inflammatory arthritis; severe renal function impairment; drug allergies; idiopathic thrombocytopenic purpura; contraindication to intramuscular injection; donation or loss of ≥400 ml of blood in the 8 weeks before dosing; live vaccination in the 3 months before the start of the study; active or recurrent infection at enrollment; active pulmonary disease; requirement for antibiotics against latent tuberculosis; risk factors for tuberculosis; any surgical or underlying hepatic, hematologic, pulmonary, infectious, or gastrointestinal condition that compromised the immune system and/or would place the patient at unacceptable risk if they received immunomodulatory therapy; or long QT syndrome or QTc > 450 msec for men and > 470 msec for women. Women of childbearing age were required to be using an acceptable method of contraception.

Assessment and definition of response.

Patients assessed pain at baseline and each subsequent scheduled visit using the 100-mm VAS and a Likert scale ranging from 0–4, where 0 = no pain and 4 = extreme pain. Pain intensity scores were also recorded by patients in their diaries 6, 12, 24, and 48 hours after treatment and 4, 5, and 6 days after treatment. Subsequent flares were identified from patient-reported signs and symptoms of gouty arthritis. The following assessments were also made at each subsequent scheduled visit: physician's global assessment of response to treatment (rated as very good, good, fair, poor, or very poor), patient's global assessment of response to treatment (rated as excellent, good, acceptable, slight, or poor), and use of rescue medication. Blood samples were collected for assessment of blood chemistry (including C-reactive protein [CRP] and serum amyloid A protein [SAA] levels), and hematology. Adverse events (AEs) were reported throughout the study, and physicians assessed local tolerability at sites of subcutaneous and intramuscular injections at each scheduled visit.

The primary end point was determination of the canakinumab dose that produced equivalent efficacy to that achieved with triamcinolone acetonide 40 mg 72 hours after treatment, according to the patient's assessment of pain on a VAS. Secondary efficacy end points included time to 50% reduction in pain, time to recurrence of flare, reductions in CRP and SAA levels, use of rescue medication, and physician's and patient's global assessments of response to treatment. Multiple comparisons were not adjusted for when analyzing the secondary end points. Safety variables included the incidence of AEs and serious AEs (SAEs), and the incidence and severity of injection site reactions.

Sample size determination and statistical analysis.

A sample size of 200 patients was considered to be sufficient to give a half-width of the 95% confidence interval (95% CI) of the target dose of at most 40 mg. The expected precision for 200 subjects was calculated based on the following assumptions: the maximum mean effect for canakinumab was expected to correspond to a 60-mm decrease on the VAS, the mean effect for triamcinolone acetonide was expected to correspond to a decrease of 40 mm on the VAS, and the standard deviation for VAS measurements was expected to be 25 mm. The inclusion of the 2 covariates, baseline measurements and BMI, was not considered in the sample size calculation, on the assumption that their inclusion would decrease the variance of the residual error, and therefore increase the precision. The unconditional 95% CI accounting for model selection was wider than the conditional confidence interval that was based on the finally selected model. An accounting factor of 1.25 based on general simulations was taken into account.

All efficacy end points were analyzed using the full analysis set (i.e., all randomized patients who had ≥1 postbaseline VAS assessment), and safety assessments were based on the safety analysis set (i.e., all randomized patients who received the study drug and had ≥1 postbaseline safety assessment). The analysis for the primary end point was performed using MCPMod methodology, which combines multiple comparisons and model-based approaches (23). The null hypothesis of a constant dose-response curve for the primary efficacy end point was tested at a significance level of 5% against the alternative hypothesis of a nonconstant dose-response curve (23).

The secondary end points of change in pain intensity in the target joint (up to 7 days after treatment) on the 100-mm VAS and changes in CRP and SAA levels were analyzed by time point by analysis of covariance with treatment group, baseline value, and BMI at baseline as covariates. Patient's and physician's global assessments of response were analyzed by proportional odds regression. The number and proportion of patients who used rescue medication up to 7 days after treatment was presented by treatment group and analyzed using a logistic model. Between-group differences in time to ≥50% reduction from baseline in pain intensity and time to first rescue medication use were analyzed using the Cox proportional hazards regression model. Kaplan-Meier estimates of time to first acute gout flare recurrence after study drug administration were calculated, and the Cox proportional hazards regression model was applied.

RESULTS

Patient characteristics.

In total, 200 patients were enrolled, and 191 patients completed the study (Figure 1). Demographic and baseline disease characteristics were generally well balanced across treatment groups (Table 1). Most patients were male and had experienced frequent gout flares in the previous 12 months. There was a baseline imbalance in pain intensity (VAS assessment) between groups, with the lowest mean VAS score in the canakinumab 150 mg group (P < 0.005). Overall, 54% of the patients had mild renal dysfunction, defined as having an estimated glomerular filtration rate (GFR) of 60–89 ml/minute/1.73 m2, and 13% had moderate renal dysfunction, defined as having an estimated GFR of 30–59 ml/minute/1.73 m2. During the study, 34% of the patients were taking allopurinol (17–45% in the canakinumab 10–90 mg groups, 32% in the canakinumab 150 mg group, and 35% in the triamcinolone acetonide group).

Figure 1.

Disposition of the patients.

Table 1. Baseline characteristics of the patients with acute gouty arthritis*
 Canakinumab 10 mg (n = 28)Canakinumab 25 mg (n = 29)Canakinumab 50 mg (n = 29)Canakinumab 90 mg (n = 29)Canakinumab 150 mg (n = 28)Triamcinolone acetonide 40 mg (n = 57)
  • *

    BMI = body mass index; VAS = visual analog scale; NSAIDs = nonsteroidal antiinflammatory drugs; CRP = C-reactive protein; SAA = serum amyloid A protein; GFR = glomerular filtration rate.

  • P < 0.005 for test of comparability across all groups.

  • Data were missing for 1 patient.

  • §

    Data were missing for 5 patients.

  • Calculated using the Cockcroft-Gault formula.

  • #

    Calculated using the Modification of Diet in Renal Disease formula.

Male, no. (%)26 (92.9)26 (89.7)27 (93.1)24 (82.8)28 (100.0)55 (96.5)
Age, years      
 Mean ± SD49.9 ± 11.1250.5 ± 10.8954.9 ± 10.7552.2 ± 12.3850.6 ± 15.3852.4 ± 11.55
 Median (range)45.5 (34, 78)49.0 (32, 76)55.0 (35, 76)52.0 (28, 76)51.5 (20, 75)51.0 (20, 76)
Age groups, no. (%)      
 ≥18–40 years6 (21.4)4 (13.8)3 (10.3)5 (17.2)8 (28.6)7 (12.3)
 ≥41–64 years21 (75.0)21 (72.4)19 (65.5)20 (69.0)15 (53.6)39 (68.4)
 ≥65–74 years03 (10.3)6 (20.7)2 (6.9)3 (10.7)10 (17.5)
 ≥75 years1 (3.6)1 (3.4)1 (3.4)2 (6.9)2 (7.1)1 (1.8)
Race, no. (%)      
 White27 (96.4)27 (93.1)23 (79.3)25 (86.2)24 (85.7)54 (94.7)
 Black1 (3.6)1 (3.4)3 (10.3)4 (13.8)1 (3.6)3 (5.3)
 Asian01 (3.4)002 (7.1)0
 Other003 (10.3)01 (3.6)0
BMI, kg/m2      
 Mean ± SD32.13 ± 4.81430.55 ± 4.86732.02 ± 4.11131.28 ± 5.58530.24 ± 4.47830.91 ± 4.130
 Median (range)32.58 (23.1, 39.8)29.35 (23.1, 39.2)31.54 (21.1, 38.8)30.64 (23.6, 39.6)30.81 (20.1, 39.0)31.12 (20.9, 39.5)
Classification of gout, no. (%)      
 Acute monarticular16 (57.1)15 (51.7)18 (62.1)17 (58.6)19 (67.9)41 (71.9)
 Acute oligoarticular9 (32.1)10 (34.5)9 (31.0)9 (31.0)6 (21.4)13 (22.8)
 Acute polyarticular3 (10.7)4 (13.8)2 (6.9)3 (10.3)3 (10.7)3 (5.3)
Number of joints affected by acute gouty flares within the past 5 days, no. (%)      
 116 (57.1)17 (58.6)19 (65.5)17 (58.6)22 (78.6)43 (75.4)
 27 (25.0)2 (6.9)4 (13.8)6 (20.7)5 (17.9)9 (15.8)
 32 (7.1)4 (13.8)3 (10.3)2 (6.9)02 (3.5)
 41 (3.6)3 (10.3)2 (6.9)1 (3.4)01 (1.8)
 >42 (7.1)3 (10.3)1 (3.4)3 (10.3)1 (3.6)2 (3.5)
Assessment of pain, 100-mm VAS      
 Mean ± SD73.8 ± 11.5077.6 ± 12.9076.3 ± 13.3872.6 ± 10.6265.5 ± 9.1472.9 ± 12.78
 Median (range)73.0 (51, 94)76.0 (50, 98)77.0 (50, 99)74.0 (54, 94)68.0 (51, 81)71.0 (52, 100)
Likert assessment of pain, no. (%)      
 None000000
 Mild000004 (7.0)
 Moderate8 (28.6)6 (20.7)8 (27.6)7 (24.1)14 (50.0)15 (26.3)
 Severe19 (67.9)18 (62.1)17 (58.6)18 (62.1)14 (50.0)32 (56.1)
 Extreme1 (3.6)5 (17.2)4 (13.8)4 (13.8)06 (10.5)
Number of flares in previous year      
 Mean ± SD6.8 ± 8.146.3 ± 5.203.9 ± 2.595.9 ± 4.935.3 ± 4.966.5 ± 9.88
 Median (range)4.0 (1, 36)5.0 (1, 24)4.0 (0, 10)5.0 (1, 24)3.5 (1, 24)3.0 (0, 50)
Disease refractory to NSAIDs and/or colchicine, no. (%)27 (96.4)27 (93.1)26 (89.7)20 (69.0)24 (85.7)50 (87.7)
Patients for whom NSAIDs and/or colchicine were contraindicated, no. (%)4 (14.3)5 (17.2)7 (24.1)12 (41.4)4 (14.3)13 (22.8)
CRP, mg/liter      
 Mean ± SD23.25 ± 33.32425.07 ± 35.01032.84 ± 54.74929.35 ± 33.42421.84 ± 34.14618.58 ± 22.925
 Median (range)13.70 (1.2, 171.0)13.70 (0.0, 147.0)14.00 (0.7, 273.0)17.30 (0.3, 127.0)13.20 (1.0, 182.0)10.90 (0.0, 91.7)
SAA, mg/liter      
 Mean ± SD57.65 ± 157.05994.07 ± 170.713101.24 ± 205.03382.21 ± 132.49965.18 ± 109.27343.35 ± 76.510§
 Median (range)14.25 (1.2, 820.0)15.50 (1.8, 699.0)14.40 (2.6, 882.0)14.70 (1.3, 536.0)35.10 (2.0, 543.0)7.85 (1.3, 413.0)
Creatinine, μmoles/liter      
 Mean ± SD93.7 ± 19.690.3 ± 16.4100.0 ± 23.790.2 ± 21.293.6 ± 23.293.7 ± 17.9
 Median (range)94.0 (57, 137)90.0 (62, 135)93.0 (72, 178)88.0 (63, 171)88.0 (67, 164)90.0 (55, 149)
Creatinine clearance      
 Low01 (3.4)1 (3.4)01 (3.6)1 (1.8)
 Normal12 (42.9)10 (34.5)11 (37.9)13 (44.8)12 (42.9)26 (45.6)
 High16 (57.1)17 (58.6)16 (55.2)16 (55.2)14 (50.0)26 (45.6)
 Missing01 (3.4)1 (3.4)01 (3.6)4 (7.0)
Estimated GFR, no. (%)#      
 >90 ml/minute/1.73 m210 (35.7)10 (34.5)4 (13.8)11 (37.9)10 (35.7)16 (28.1)
 60–89 ml/minute/1.73 m215 (53.6)16 (55.2)18 (62.1)15 (51.7)15 (53.6)28 (49.1)
 30–59 ml/minute/1.73 m23 (10.7)2 (6.9)6 (20.7)3 (10.3)2 (7.1)9 (15.8)
 Missing data01 (3.4)1 (3.4)01 (3.6)4 (7.0)
Serum urate, mg/dl      
 Mean ± SD8.51 ± 1.717.75 ± 1.977.67 ± 2.107.53 ± 2.237.89 ± 1.577.83 ± 2.14
 Median (range)8.50 (4.99, 12.99)7.75 (3.90, 11.69)7.95 (3.90, 13.09)7.90 (3.29, 12.89)7.80 (4.50, 10.79)8.05 (2.40, 12.89)

Primary objective.

The study aimed to determine the canakinumab dose showing equivalent efficacy to triamcinolone acetonide 40 mg 72 hours after treatment, according to the VAS assessment of pain. A statistically significant dose response for canakinumab was seen at 72 hours (Figure 2). All doses were associated with numerically less pain (according to the VAS score) 72 hours after treatment compared with triamcinolone acetonide; thus, because all canakinumab doses showed better efficacy than anticipated, it was not possible to determine an equivalent dose at this time point.

Figure 2.

Dose-response curve showing predicted pain intensity 72 hours after treatment, as determined using a 100-mm visual analog scale (VAS). The predicted response and 95% confidence interval (95% CI) are shown. The predictions were estimated from mean baseline VAS and baseline body mass index values.

A statistically significant, linear dose response was also seen at 48 hours. Analysis of pain 48 hours after treatment showed that a canakinumab dose of 23 mg (95% CI 3–96 mg) had equivalent efficacy to triamcinolone acetonide 40 mg at this time point.

Secondary objectives.

Time to pain reduction and to flare recurrence.

For all clinical secondary end points, responses to the canakinumab 150 mg dose were superior to responses to triamcinolone acetonide 40 mg. The reduction from baseline in pain intensity (VAS assessment) was numerically greater than with triamcinolone acetonide from 6 hours onward, and differences were statistically significant 24, 48, and 72 hours after treatment (differences of −11.5 mm [P = 0.04], −18.2 mm [P = 0.002], and −19.2 mm [P < 0.001], respectively) and on days 4, 5, and 7 after treatment (differences of −14.1 mm [P = 0.012], −14.2 mm [P = 0.007], and −10.5 [P = 0.04], respectively). The reduction in pain from baseline was also numerically greater for the lower canakinumab doses compared with triamcinolone acetonide at all time points (except at 12 hours for the canakinumab 25 mg group), but differences between the canakinumab groups and the triamcinolone acetonide group were not statistically significant (Figure 3A). Time to a 50% reduction in pain was significantly shorter with canakinumab 150 mg than with triamcinolone acetonide 40 mg (median 1.0 days versus 2.0 days; P < 0.001); statistical significance versus triamcinolone acetonide was not reached for the lower doses (median 2.9 days in the 10 mg group, 2.9 days in the 25 mg group, 1.0 days in the 50 mg group, and 1.0 days in the 90 mg group). Canakinumab at all doses reduced the risk of recurrent flares compared with triamcinolone acetonide.

Figure 3.

Pain relief, time to recurrent gout flare, and C-reactive protein (CRP) levels over the course of the study in patients with acute gouty arthritis treated with canakinumab (10–150 mg) or triamcinolone acetonide (40 mg). A, Least squares (LS) mean change from baseline in pain intensity (as assessed using a 100-mm visual analog scale) following administration of canakinumab 10–150 mg or triamcinolone acetonide 40 mg. Bars show the mean ± SEM. ∗ = P < 0.05 versus triamcinolone acetonide 40 mg. B, Time to first acute gout flare recurrence after treatment, according to the Kaplan-Meier estimate. Incidence of flare at 8 weeks was 3.6% in the 10 mg group (n = 1), 10.3% in the 25 mg group (n = 3), 3.6% in the 50 mg group (n = 1), 13.8% in the 90 mg group (n = 4), 3.7% in the 150 mg group (n = 1), and 45.4% in the triamcinolone acetonide 40 mg group (n = 25) (P ≤ 0.01 for all doses of canakinumab versus triamcinolone acetonide). C, Median CRP levels following administration of canakinumab 10–150 mg or triamcinolone acetonide 40 mg. The upper limit of normal (ULN) is 3.0 mg/liter.

Time to recurrence of flare was significantly longer (P ≤ 0.01 for all doses). By 8 weeks after treatment only 1 patient in the canakinumab 150 mg group had experienced a flare (3.7%) compared with 25 patients (45%) in the triamcinolone acetonide group, a relative risk reduction of 94% (see Figure 3B for results for all doses).

Markers of inflammation.

At baseline, median CRP and SAA levels were above the upper limit of normal (ULN) (3.0 mg/liter for CRP; 6.7 mg/liter for SAA) in all treatment groups and did not differ significantly between groups (Table 2). Median CRP levels normalized by day 7 in all canakinumab groups except the lowest dose group and remained below the ULN for the rest of the study (Figure 3C). In the triamcinolone acetonide group, median CRP levels remained above the ULN throughout the study. Median SAA levels normalized by day 7 in all canakinumab groups and remained below the ULN for the rest of the study (Table 2). In the triamcinolone acetonide group, the median SAA level was still above the ULN on day 7, but had decreased to below the ULN by day 28.

Table 2. Changes from baseline in pain intensity, markers of inflammation, and physician's and patient's global assessments 72 hours and 7 days after treatment, and use of rescue medication*
 Canakinumab 10 mg (n = 28)Canakinumab 25 mg (n = 29)Canakinumab 50 mg (n = 28)Canakinumab 90 mg (n = 29)Canakinumab 150 mg (n = 27)Triamcinolone acetonide 40 mg (n = 56)
Base- line72 hours7 daysBase- line72 hours7 daysBase- line72 hours7 daysBase- line72 hours7 daysBase- line72 hours7 daysBase- line72 hours7 days
  • *

    All P values are versus triamcinolone acetonide 40 mg.– = not applicable (see Table 1 for other definitions).

  • P for actual (not percentage) change in VAS from baseline, by analysis of covariance (ANCOVA) with treatment group, VAS at baseline, and BMI at baseline as covariates.

  • By ANCOVA, with treatment group, protein level at baseline, and BMI at baseline as covariates.

  • §

    By proportional odds regression, with treatment group and BMI at baseline as covariates.

  • By logistic regression, with treatment group, pain intensity (assessed by VAS) at baseline, and BMI at baseline as covariates.

% change from baseline  in VAS score            
 No.282427252625282727255343
 Mean ± SD−67.0 ± 33.6−78.9 ± 28.2−67.9 ± 25.2−78.8 ± 29.0−65.1 ± 37.8−84.4 ± 25.6−71.7 ± 26.2−82.6 ± 21.6−84.6 ± 20.7−92.7 ± 12.1−57.8 ± 39.2−74.8 ± 32.7
 P0.3310.7470.5520.6670.3400.1310.0820.285<0.0010.042
CRP, mg/liter            
 No.282728292829282628292828272527565355
 Median (range)13.70 (1.2, 171.0)7.17 (1.0, 111.0)3.52 (0.3, 60.5)13.70 (0, 147.0)3.99 (0, 33.1)1.26 (0, 12.3)14.00 (0.7, 273.0)5.13 (0.4, 78.8)1.52 (0, 51.7)17.30 (0.3, 127.0)6.32 (0, 93.9)2.03 (0.2, 87.2)13.20 (1.0, 182.0)3.85 (0.3, 58.3)1.49 (0.3, 15.1)10.90 (0, 91.7)4.61 (0.2, 103.0)4.05 (0, 148.0)
 Change from baseline,            
  median (range)−3.20 (−148.1, 93.0)−6.45 (−164.4, 54.5)–6.01 (–113.9, 19.0)–11.26 (–139.6, 0)–9.13 (–219.3, 0.1)–12.14 (–263.3, 48.5)–7.25 (–103.7, 49.7)–13.93 (–124.1, 1.6)–7.58 (–153.4, 24.4)–11.88 (–177.2, 12.8)–3.10 (–66.7, 91.5)–4.67 (–79.2, 136.5)
 P0.4340.1630.1820.0060.6970.0220.9920.0700.3340.016
SAA, mg/liter            
 No.282628282829282828292828272726525453
 Median (range)14.25 (1.2, 820.0)5.45 (1.5, 490.0)4.50 (0, 106.0)15.50 (1.8, 699.0)7.55 (1.1, 238.0)3.50 (1.0, 24.9)14.40 (2.6, 882.0)5.50 (1.2, 700.0)3.50 (0, 102.0)14.70 (1.3, 536.0)7.80 (1.2, 258.0)3.45 (0.9, 158.0)35.10 (2.0, 543.0)7.80 (0, 372.0)4.60 (0.9, 76.9)7.85 (1.3, 413.0)7.20 (0, 726.0)6.90 (0, 668.0)
 Change from baseline,            
  median (range)–6.75 (–622.0, 473.0)–7.60 (–797.4, 100.1)−3.20 (−497.0, 146.9)−9.05 (−674.1, 4.5)–4.90 (–757.0, 575.0)–10.25 (–870.6, 94.8)–4.65 (–431.0, 84.6)–11.80 (–520.6, 2.7)–7.70 (–517.9, 281.1)–33.55 (–540.1, 73.1)–1.80 (–363.7, 622.0)–2.00 (–349.1, 606.2)
 P0.9240.0370.3040.0140.4550.0480.3400.0370.2950.039
Physician's global assessment,  no. (%)            
 Very good9 (32.1)12 (42.9)10 (34.5)11 (37.9)11 (39.3)17 (60.7)9 (31.0)11 (37.9)13 (48.1)19 (70.4)15 (26.8)18 (32.1)
 Good12 (42.9)10 (35.7)8 (27.6)8 (27.6)11 (39.3)8 (28.6)13 (44.8)12 (41.4)12 (44.4)8 (29.6)19 (33.9)21 (37.5)
 Fair4 (14.3)6 (21.4)8 (27.6)9 (31.0)4 (14.3)3 (10.7)7 (24.1)5 (17.2)2 (7.4)012 (21.4)7 (12.5)
 Poor2 (7.1)02 (6.9)1 (3.4)1 (3.6)000007 (12.5)7 (12.5)
 Very poor1 (3.6)01 (3.4)01 (3.6)000003 (5.4)2 (3.6)
 No. of patients with  missing data00000001 (3.4)0001 (1.8)
 P§0.2410.1850.4960.7440.0720.0050.1580.2070.003<0.001
Patient's global assessment,  no. (%)            
 Excellent5 (17.9)11 (39.3)5 (17.2)9 (31.0)11 (39.3)12 (42.9)10 (34.5)9 (31.0)12 (44.4)16 (59.3)12 (21.4)13 (23.2)
 Good13 (46.4)9 (32.1)13 (44.8)10 (34.5)9 (32.1)8 (28.6)9 (31.0)13 (44.8)12 (44.4)9 (33.3)18 (32.1)18 (32.1)
 Acceptable6 (21.4)5 (17.9)3 (10.3)4 (13.8)4 (14.3)8 (28.6)6 (20.7)5 (17.2)2 (7.4)012 (21.4)13 (23.2)
 Slight2 (7.1)2 (7.1)5 (17.2)5 (17.2)2 (7.1)03 (10.3)001 (3.7)9 (16.1)7 (12.5)
 Poor2 (7.1)1 (3.6)3 (10.3)1 (3.4)2 (7.1)01 (3.4)1 (3.4)1 (3.7)1 (3.7)5 (8.9)4 (7.1)
 No. of patients with  missing data00000001 (3.4)0001 (1.8)
 P§0.5780.0760.8880.4410.0530.0270.1310.0640.002<0.001
Patients receiving rescue medication, no. (%)13 (46.4)16 (55.2)16 (57.1)14 (48.3)6 (22.2)31 (55.4)
 P0.3870.9290.9970.5310.013
Patients taking each type of rescue medication, no. (%)            
 Acetaminophen9 (32.1)12 (41.4)15 (53.6)12 (41.4)5 (18.5)23 (41.1)
 Codeine4 (14.3)6 (20.7)4 (14.3)4 (13.8)1 (3.7)9 (16.1)
 Prednisolone/   prednisone5 (17.9)9 (31.0)8 (28.6)6 (20.7)2 (7.4)16 (28.6)

Physician's and patient's global assessments and rescue medication use.

Seventy-two hours after treatment (the first time point analyzed), 93% of patients in the canakinumab 150 mg group had very good or good responses according to physician's global assessment, compared with 61% for triamcinolone acetonide 40 mg (Table 2), and the difference between these groups remained statistically significant throughout the study (P < 0.05). Similarly, at 72 hours, 89% of the patients in the canakinumab 150 mg group reported excellent/good responses, compared with 54% for triamcinolone acetonide (Table 2), and the difference between these groups remained statistically significant throughout the study (P ≤ 0.02). (See Table 2 for results for other canakinumab groups.)

During the 7 days following study drug administration, 31 patients (55%) in the triamcinolone acetonide group received rescue medication for pain relief, compared with 6 patients (22%) in the canakinumab 150 mg group (P = 0.01) (Table 2), and time to first rescue medication dose was significantly longer with canakinumab 150 mg than with triamcinolone acetonide (hazard ratio 0.36, P = 0.02). Patients in the lower canakinumab dose groups received more rescue medication than those in the canakinumab 150 mg group, and differences in rescue medication use (Table 2) and time to first use (data not shown) compared with triamcinolone acetonide were not statistically significant. Use of corticosteroid rescue medication was lowest in the canakinumab 150 mg group (7% of patients) and was similar across the other treatment groups (18–31%) (Table 2).

Safety and tolerability.

Safety data for the individual dose groups are summarized in Table 3. No patients died or discontinued the study due to AEs. Four SAEs were reported with canakinumab (appendicitis [1 patient in the 25 mg group and 1 patient in the 50 mg group], bronchitis [1 patient in the 25 mg group], and carotid artery stenosis [1 patient in the 50 mg group]), and 1 was reported with triamcinolone acetonide 40 mg (cerebrovascular disorder), but none of these was suspected by the investigator to be related to the study medication. The overall incidence of AEs was similar for canakinumab (41%) and triamcinolone acetonide (42%). All events, except a flare of gout in the canakinumab 10 mg group and an episode of acute spastic bronchitis in the canakinumab 25 mg group, were mild or moderate in severity. No apparent pattern of AEs was seen in any of the treatment groups; no specific AE was experienced by >3 patients in any of the canakinumab groups, and there was no evidence of a dose response for any AE. The incidence of infectious AEs was low (<11%) in all groups (Table 3). Few injection site reactions were reported, and all were mild in severity (Table 3).

Table 3. Adverse events (safety set)*
 Canakinumab 10 mg (n = 28)Canakinumab 25 mg (n = 29)Canakinumab 50 mg (n = 29)Canakinumab 90 mg (n = 29)Canakinumab 150 mg (n = 28)Canakinumab any dose (n = 143)Triamcinolone acetonide 40 mg (n = 57)
  • *

    Values are the number (%) of patients. AST = aspartate aminotransferase; ALT = alanine aminotransferase.

  • Serious adverse events (SAEs) were appendicitis (in 1 patient in the canakinumab 25 mg group and 1 patient in the canakinumab 50 mg group), bronchitis (in 1 patient in the canakinumab 25 mg group), carotid artery stenosis (in 1 patient in the canakinumab 50 mg group), and cerebrovascular disorder (in 1 patient in the triamcinolone acetonide group).

Any SAE02 (6.9)2 (6.9)004 (2.8)1 (1.8)
Any AE10 (35.7)13 (44.8)15 (51.7)12 (41.4)9 (32.1)59 (41.3)24 (42.1)
Any severe AE1 (3.6)1 (3.4)0002 (1.4)0
Any infectious AE03 (10.3)3 (10.3)2 (6.9)2 (7.1)10 (7.0)4 (7.0)
Injection site reaction    
 Subcutaneous1 (3.6)01 (3.4)1 (3.4)1 (3.6)4 (2.8)2 (3.5)
 Intramuscular1 (3.6)00001 (0.7)1 (1.8)
AEs reported in >5%  of patients in any group    
 Headache1 (3.6)3 (10.3)1 (3.4)2 (6.9)1 (3.6)8 (5.6)4 (7.0)
 Nasopharyngitis1 (3.6)1 (3.4)2 (6.9)01 (3.6)5 (3.5)2 (3.5)
 Increased AST0002 (6.9)1 (3.6)3 (2.1)0
 Increased blood uric acid0002 (6.9)1 (3.6)3 (2.1)0
 Oropharyngeal pain0002 (6.9)1 (3.6)3 (2.1)0
 Increased ALT0002 (6.9)02 (1.4)0
 Bone pain002 (6.9)002 (1.4)0
 Hyperhidrosis02 (6.9)0002 (1.4)0
 Pain in extremity01 (3.4)1 (3.4)002 (1.4)3 (5.3)
 Urinary tract infection0001 (3.4)01 (0.7)3 (5.3)

DISCUSSION

Acute gouty arthritis usually responds to treatment with NSAIDs or colchicine. However, patients with gout frequently have multiple comorbidities (24) that may mean that treatment with NSAIDs or colchicine is contraindicated. Furthermore, these treatments are not always effective, and some patients tolerate them poorly. In this study, we wished to investigate the role of canakinumab in such difficult-to-treat patients with acute gouty arthritis. Indeed, two-thirds of patients had some degree of renal impairment and therefore had a relative or absolute contraindication to NSAID therapy, and almost 90% of patients had disease that was refractory to NSAIDs and/or colchicine. The results of this study, one of the largest to date in acute gout, indicate that canakinumab, a specific inhibitor of IL-1β, is an effective treatment option for acute gout flares in such patients, inducing rapid and sustained reductions in pain and significantly reducing the risk of recurrent flares.

The study aimed to determine the canakinumab dose that would produce equivalent pain relief 72 hours after treatment to that achieved with intramuscular triamcinolone acetonide 40 mg, a commonly used treatment for acute gouty arthritis. It was intended that the identified dose would then be investigated further in a phase III study. Modeling results estimated that the canakinumab dose equivalent to triamcinolone acetonide 40 mg 72 hours after treatment was lower than the range of doses tested and therefore could not be determined. However, the 150 mg dose demonstrated consistently superior efficacy to triamcinolone acetonide 40 mg across all assessments, including time to 50% reduction in pain, patient's and physician's global assessments, reductions in markers of inflammation, rescue medication use, and risk of recurrent flares. Moreover, this dose produced a clinically meaningful difference in pain reduction (10 mm on the VAS) by 24 hours compared with triamcinolone acetonide. In contrast, statistically significant differences compared with triamcinolone acetonide were not reached with the lower canakinumab doses for most efficacy parameters, suggesting that the canakinumab 150 mg dose would be an appropriate one for further study.

A significant dose response was observed across the 5 canakinumab doses for reduction in pain at 72 hours, but was less evident for other efficacy measures. This may reflect the fact that rescue medication use was ∼2-fold higher in the lower canakinumab dose groups than in the canakinumab 150 mg group. The results of an additional analysis that excluded patients who had used prednisolone or prednisone suggested that the increased use of corticosteroids as rescue medication with the lower canakinumab doses had an impact on the dose-response effect observed for efficacy measures. Rescue medication use in the triamcinolone acetonide group was also ∼2-fold higher than for the canakinumab 150 mg group. Thus, the use of rescue medication may have contributed to the observed efficacy for triamcinolone acetonide and for the lower canakinumab doses. This provides further rationale for the choice of canakinumab 150 mg as the dose for further study.

Patients with gout may experience frequent recurrent flares, as demonstrated by the high rate of flares in our patient population before the start of the study. Such patients should be offered urate-lowering therapy. However, a significant number of patients, especially those with comorbidities, are not given urate-lowering therapy or cannot receive adequate doses to enable serum urate levels of <6 mg/dl to be achieved (25), whereas others are only poorly compliant with urate-lowering therapy (26, 27). (Indeed, in our study, only a third of the patients were receiving urate-lowering therapy.)

In the present study, canakinumab of all doses significantly reduced the risk of flare, compared with triamcinolone acetonide. In particular, in the canakinumab 150 mg group, only 1 patient (4%) experienced a flare during the 8 weeks after treatment, compared with 25 patients (45%) in the triamcinolone acetonide group, a relative risk reduction of 94% 8 weeks after treatment (P = 0.006). The higher incidence of flares in the triamcinolone acetonide group is unlikely to have reflected a higher risk of rebound flares in this group because approximately half of the patients receiving triamcinolone acetonide did not experience flares in the 8-week period, and more flares occurred in the second month of the study. In addition, effective plasma levels of triamcinolone acetonide are maintained for 14–21 days following administration, and previous studies have demonstrated the absence of rebound flares with triamcinolone acetonide (28, 29). These results thus suggest that canakinumab may be a valuable adjunctive therapy for treatment or prevention of recurrent flares.

The patient population involved in this study was largely typical of patients with gouty arthritis whose disease is refractory to or who have contraindications to standard antiinflammatory therapy. However, most patients were white. There are currently no data to suggest that responses to IL-1β blockade are influenced by race. Hence, the results of our study are likely to be valid for other races.

Canakinumab was generally well tolerated. No organ toxicity was observed, and there were no specific dose-related effects. The 4 SAEs reported were not considered to be related to study medication. These results are further supported by findings in patients with CAPS, in whom treatment for up to 48 weeks was shown to be well tolerated (12).

This study further validates the concept of IL-1β as a key mediator of inflammation in acute gouty arthritis. This was first demonstrated in a case–cohort study in which anakinra demonstrated excellent efficacy in terms of pain relief in patients for whom standard antiinflammatory therapy was contraindicated (15). Rilonacept, another selective IL-1β inhibitor, has also been reported to produce a significant reduction in pain, and inflammation in a small open-label study in patients with chronic active gout (18). However, anakinra and rilonacept require frequent administration (once daily and once weekly, respectively, compared with once every 8 weeks for canakinumab in CAPS) (12) and are often associated with injection site reactions in patients with CAPS (13, 14). Canakinumab may therefore offer significant advantages over anakinra and rilonacept.

This study had a number of limitations. First, triamcinolone acetonide 40 mg was chosen as the active comparator based on the clinical experience of the investigators and because it is the standard dose used in most countries in which the study was performed. In a few countries, the standard initial dose is 60 mg, and the only 2 published studies investigating triamcinolone acetonide used the 60 mg dose. However, both were small studies (involving 27 and 31 patients) (28, 29), and there are no data to indicate a difference in efficacy between the 2 triamcinolone acetonide doses.

Second, this global study included patients whose disease was refractory to or who had contraindications for NSAIDs and/or colchicine, based on their physician's assessment. The absence of strict criteria for refractoriness or contraindication may mean that the patient population was not well defined, and hence data cannot easily be compared with those from other studies. However, in this phase II study, we chose to leave the definition of refractoriness and contraindications to the treating physicians, thus reflecting routine clinical practice, in order to determine the characteristics of patients regarded by clinicians as having disease that is refractory to and/or having contraindications for NSAIDs and/or colchicine. Results from this study will help to define the patient population for inclusion in a planned phase III study.

As the incidence and prevalence of gout are increasing (1, 30), there is a need for more effective treatments to provide rapid and sustained pain relief in acute gouty arthritis, especially because many patients cannot tolerate or have disease that is refractory to standard treatments, and/or have comorbidities that prevent their use. Despite the widespread use of corticosteroids in the treatment of gout flares, there are few randomized controlled trials of these therapies in acute gout (31). This randomized controlled study therefore provides important data on the efficacy of triamcinolone acetonide and canakinumab for acute gouty arthritis and suggests that canakinumab may be a valuable new treatment option. Additional studies are now warranted to further investigate the value of canakinumab in the treatment of gout.

AUTHOR CONTRIBUTIONS

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. So 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. So, Murphy, Arulmani, Sallstig, Schlesinger.

Acquisition of data. So, De Meulemeester, Pikhlak, Yücel, Richard, Murphy, Arulmani, Sallstig.

Analysis and interpretation of data. So, Pikhlak, Richard, Murphy, Arulmani, Sallstig, Schlesinger.

Acknowledgements

The authors thank the patients and investigators (the Canakinumab in Gout Study Group) who took part in this study. The authors thank Kirstin Stricker (Clinical Communication Leader, a member of the clinical development team, Novartis Pharma) for her contribution to the interpretation of the data, critical review of the paper, and for coordinating author discussions and writing of the manuscript, and Rowena Hughes and Annette Keith (of Oxford PharmaGenesis) for medical writing support, editorial assistance, and collation and incorporation of comments from all authors. This editorial help was supported by Novartis Pharma.

APPENDIX A

THE CANAKINUMAB IN GOUT STUDY GROUP

The following investigators were members of the Canakinumab in Gout Study Group: Bernardo Pons Estel (Argentina); Marc De Meulemeester (Belgium); John Li, Dennis O'Keefe, Proton Rahman, (Canada); Frederic Liote (France); Hischam Bouzo, Oana Förster, Uwe Gerbaulet, Veit Lorenz, René Martz, Ruediger Moericke, Ingo Senftleber (Germany); Maria Misterska-Skora (Poland); Olga Ershova, Anatoly Kuzin, Vadim Mazurov, Evgeny Nasonov, Marianna Petrova, Andrey Pikhlak, Olga Ryabitseva, Ludmila Suplotova (Russia); Claude Merlin, Burkhard Moeller, Alexander So, Alan Tyndall (Switzerland); Nurullah Akkoç, Ayhan Dinç, Ahmet Gül, Yasar Karaaslan, Ahmet Onat, Süleyman Ozbek, Taskin Sentürk, A Eftal Yücel (Turkey); Bhavesh Bodalia, Adrian Darrah (UK); Mohammad Ali, Andrew Baldassare, Billy Chacko, Vishala Chindalore, Clancy Cone, Michael Cox, Ara Dikranian, Paul Dura, Marina Fernandez, Chester Fisher, Wayne Gilbert, Alan Kivitz, Howard Knapp, William Knibbe, Brock McConnehey, Madura Rangaraj, Daniel Ripley, Lance Rudolph, Kenneth Saag, Mercedes Samson, Naomi Schlesinger, Elliott Schwartz, Talha Shamim, David Sikes, Joel Silverfield, Atul Singhal, Arnaldo Torres, Robert Trapp, Charles White (US).

Ancillary