Variable | Canakinumab 25 mg n=55 | Canakinumab 50 mg n=54 | Canakinumab 100 mg n=54 | Canakinumab 200 mg n=54 | Canakinumab 300 mg n=53 | Canakinumab every 4 weeks n=54 | Colchicine 0.5 mg n=108 |
---|---|---|---|---|---|---|---|

Mean number of flares per patient | |||||||

ANCOVA (preplanned analysis)^{*} | |||||||

Least-squares mean (SE) | 0.51 (0.19) | 0.45 (0.19) | 0.23 (0.19) | 0.41 (0.19) | 0.23 (0.19) | 0.70 (0.19) | 0.75 (0.14) |

Estimate difference to colchicine (95% CI) | −0.24 (−0.67 to 0.19) | −0.30 (−0.73 to 0.13) | −0.52 (−0.95 to −0.08)^{†} | −0.34 (−0.77 to 0.10) | −0.52 (−0.95 to −0.08)^{†} | −0.05 (−0.48 to 0.38) | |

Negative binomial model (post hoc analysis)^{‡} | |||||||

Rate ratio estimate (95% CI) | 0.60 (0.34 to 1.06) | 0.34 (0.17 to 0.69)^{§} | 0.28 (0.13 to 0.60)^{§} | 0.37 (0.19 to 0.72)^{§} | 0.29 (0.14 to 0.60)^{§} | 0.38 (0.19 to 0.77)^{§} | NA |

Patients experiencing ≥1 flare | |||||||

n (%) | 15 (27.3) | 9 (16.7) | 8 (14.8) | 10 (18.5) | 8 (15.1) | 9 (16.7) | 48 (44.4) |

Estimated OR (95% CI)^{¶} | 0.47 (0.23 to 0.95)^{†} | 0.25 (0.11 to 0.56)^{†} | 0.22 (0.09 to 0.51)^{†} | 0.28 (0.13 to 0.61)^{†} | 0.22 (0.10 to 0.52)^{†} | 0.25 (0.11 to 0.56)^{†} | NA |

Time to first flare^{**} | |||||||

HR (95% CI) | 0.57 (0.32 to 1.02) | 0.32 (0.16 to 0.66)^{†} | 0.28 (0.13 to 0.58)^{†} | 0.36 (0.18 to 0.71)^{†} | 0.28 (0.13 to 0.60)^{†} | 0.32 (0.16 to 0.66)^{†} | NA |

Average duration of all flares | |||||||

No. of patients with flares | 15 | 9 | 8 | 9 | 8 | 9 | 48 |

Least-squares mean (SD)^{*} | 4.62 (0.97) | 3.66 (1.20) | 2.80 (1.35) | 3.59 (1.25) | 3.09 (1.28) | 3.33 (1.23) | 5.08 (0.62) |

Estimated difference (95% CI) | −0.46 (−2.59 to 1.67) | −1.42 (−4.00 to 1.16) | −2.28 (−5.02 to 0.47) | −1.49 (−4.08 to 1.10) | −1.98 (−4.70 to 0.73) | −1.75 (−4.34 to 0.84) | NA |

↵* ANCOVA with treatment group, allopurinol dose at baseline and BMI at baseline as covariates.

↵† p≤0.05 versus colchicine 0.5 mg group.

↵‡ Post hoc analysis: Rate ratio estimated from negative binomial model with treatment group, allopurinol dose at baseline and BMI at baseline as covariates and log (time on study) as an offset. Estimates represent the ratio of mean number of gouty arthritis flares per patient in comparison with the colchicine 0.5 mg group.

↵§ p≤0.0083 (Bonferroni correction 0.05/6) versus colchicine 0.5 mg group.

↵¶ Logistical regression with treatment group, allopurinol dose at baseline and BMI at baseline as covariates. An OR >1 indicates that a gouty arthritis flare is more likely to occur in comparison with the colchicine 0.5 mg group.

↵** Cox's proportional hazard regression model with treatment group, allopurinol dose at baseline and BMI at baseline as covariates.

ANCOVA, analysis of covariance; BMI, body mass index; CI, confidence interval; HR, hazard ratio; NA, not applicable; OR, odds ratio; SE, standard error.