ACR20 responders are subjects with at least 20% improvement from Baseline for tender joint count (TJC), swollen joint count (SJC), and at least 3 of the 5 remaining core set measures: 1) Health Assessment Questionnaire-Disability Index (HAQ-DI), 2) C-reactive Protein (CRP), 3) Patient's Assessment of Arthritis Pain-Visual Analog Scale, 4) Patient's Global Assessment of Disease Activity-Visual Analog Scale, 5) Physician's Global Assessment of Disease Activity-Visual Analog Scale. Missing values were imputed using Non-Responder Imputation (NRI)

Of the 208 subjects in the Full Analysis Set (FAS) 208 subjects (69 400 mg CZP, 70 200 mg CZP, 69 placebo) are included in this analysis which uses Non-Response Imputation (NRI).

ACR20 responders are subjects with at least 20% improvement from Baseline for tender joint count (TJC), swollen joint count (SJC), and at least 3 of the 5 remaining core set measures: 1) Health Assessment Questionnaire-Disability Index (HAQ-DI), 2) C-reactive Protein (CRP), 3) Patient's Assessment of Arthritis Pain-Visual Analog Scale, 4) Patient's Global Assessment of Disease Activity-Visual Analog Scale, 5) Physician's Global Assessment of Disease Activity-Visual Analog Scale. Missing values were imputed using Non-Responder Imputation (NRI)

Since Non-Response Imputation (NRI) was used, all 333 subjects in the run-in period are included in this analysis

ACR50 responders are subjects with at least 50% improvement from Baseline for tender joint count (TJC), swollen joint count (SJC), and at least 3 of the 5 remaining core set measures: 1) Health Assessment Questionnaire-Disability Index (HAQ-DI), 2) C-reactive Protein (CRP), 3) Patient's Assessment of Arthritis Pain-Visual Analog Scale, 4) Patient's Global Assessment of Disease Activity-Visual Analog Scale, 5) Physician's Global Assessment of Disease Activity-Visual Analog Scale. Missing values were imputed using Non-Responder Imputation (NRI)

Since Non-Response Imputation (NRI) was used, all 333 subjects in the run-in period are included in this analysis

ACR70 responders are subjects with at least 70% improvement from Baseline for tender joint count (TJC), swollen joint count (SJC), and at least 3 of the 5 remaining core set measures: 1) Health Assessment Questionnaire-Disability Index (HAQ-DI), 2) C-reactive Protein (CRP), 3) Patient's Assessment of Arthritis Pain-Visual Analog Scale, 4) Patient's Global Assessment of Disease Activity-Visual Analog Scale, 5) Physician's Global Assessment of Disease Activity-Visual Analog Scale. Missing values were imputed using Non-Responder Imputation (NRI)

Since Non-Response Imputation (NRI) was used, all 333 subjects in the run-in period are included in this analysis

DAS28-ESR is calculated using the tender joint count (TJC), swollen joint count (SJC) erythrocyte sedimentation rate (ESR in mm/hour), and the Patient's Global Assessment of Disease Activity - Visual Analog Scale (VAS in mm) using the following formula: 0.56 x √(TJC) + 0.28 x √(SJC) + 0.70 x lognat (ESR) + 0.014 x Global Assessment of Arthritis where 28 joints are examined and a lower score indicates less disease activity. This analysis was carried out using the Last Observation Carried Forward (LOCF) method.

< 2.6 Remission,

> = 2.6 - < =3.2 Low, > 3.2 - < = 5.1 Moderate, > 5.1 High

Of the 333 subjects in the Run-in period, 325 are included in this analysis using the Last Observation Carried Forward (LOCF) method.

SDAI is calculated as the sum of tender joint count (TJC), swollen joint count (SJC), C-reactive protein (CRP in mg/dL), Patient's Global Assessment of Disease Activity - Visual Analog Scale (VAS in cm), and Investigator's Global Assessment of Disease Activity - Visual Analog Scale (VAS in cm). 28 joints are examined where a lower score indicates less disease activity. This analysis was carried out using the Last Observation Carried Forward (LOCF) method.

<= 3.3 Remission, > 3.3 - <= 11 Low, > 11 - <= 26 Moderate, > 26 High

Of the 333 subjects in the Run-in period, 326 are included in this analysis using the Last Observation Carried Forward (LOCF) method.

CDAI is calculated as the sum of tender joint count (TJC), swollen joint count (SJC), Patient's Global Assessment of Disease Activity - Visual Analog Scale (VAS in cm), and Investigator's Global Assessment of Disease Activity - Visual Analog Scale (VAS in cm). 28 joints are examined. This analysis was carried out using the Last Observation Carried Forward (LOCF) method.

The range for the CDAI is 0 - 76 with a negative change in CDAI score indicating an improvement in disease activity and a positive change in score indicating a worsening of disease activity.

Of the 333 subjects in the Run-in period, 328 are included in this analysis using the Last Observation Carried Forward (LOCF) method.

DAS28-ESR is calculated using the tender joint count (TJC), swollen joint count (SJC) erythrocyte sedimentation rate (ESR in mm/hour), and the Patient's Global Assessment of Disease Activity - Visual Analog Scale (VAS in mm) using the following formula: 0.56 x √(TJC) + 0.28 x √(SJC) + 0.70 x lognat (ESR) + 0.014 x Global Assessment of Arthritis where 28 joints are examined and a lower score indicates less disease activity. Missing values were imputed using Non-Responder Imputation (NRI).

< 2.6 (Remission),

> = 2.6 - < =3.2 Low, > 3.2 - < = 5.1 Moderate, > 5.1 High

Since Non-Response Imputation (NRI) was used, all 333 subjects in the run-in period are included in this analysis

SDAI is calculated as the sum of tender joint count (TJC), swollen joint count (SJC), C-reactive protein (CRP in mg/dL), Patient's Global Assessment of Disease Activity - Visual Analog Scale (VAS in cm), and Investigator's Global Assessment of Disease Activity - Visual Analog Scale (VAS in cm). 28 joints are examined where a lower score indicates less disease activity. Missing values were imputed using Non-Responder Imputation (NRI).

<= 3.3 (Remission), > 3.3 - <= 11 Low, > 11 - <= 26 Moderate, > 26 High

Since Non-Response Imputation (NRI) was used, all 333 subjects in the run-in period are included in this analysis

CDAI is calculated as the sum of tender joint count (TJC), swollen joint count (SJC), Patient's Global Assessment of Disease Activity - Visual Analog Scale (VAS in cm), and Investigator's Global Assessment of Disease Activity - Visual Analog Scale (VAS in cm). 28 joints are examined where a lower score indicates less disease activity. Missing values were imputed using Non-Responder Imputation (NRI).

The range for the CDAI is 0 - 76 with a lower CDAI score indicating approvement in activity and a higher score indicating a decline activity.

Since Non-Response Imputation (NRI) was used, all 333 subjects in the run-in period are included in this analysis

Ratio is defined as the CRP value at Week 16 divided by the CRP value at Baseline. This analysis was carried out using the Last Observation Carried Forward (LOCF) method.

Of the 333 subjects in the Run-in period, 330 are included in this analysis using the Last Observation Carried Forward (LOCF) method.

HAQ-DI is derived based on the mean of individual scores in 8 categories of daily living actives (using 20 questions). Each question is scored 0-3 (0 = without any difficulty, 1 = with some difficulty, 2 = with much difficulty, and 3 = unable to do). Thus, the mean also has a range from 0-3. Change from baseline is computed as the value at Week 16 minus the baseline value. A negative value in change from baseline indicates an improvement. This analysis was carried out using the Last Observation Carried Forward (LOCF) method.

Of the 333 subjects in the Run-in period, 330 are included in this analysis using the Last Observation Carried Forward (LOCF) method.

ACR50 responders are subjects with at least 50% improvement from Baseline for tender joint count (TJC), swollen joint count (SJC), and at least 3 of the 5 remaining core set measures: 1) Health Assessment Questionnaire-Disability Index (HAQ-DI), 2) C-reactive Protein (CRP), 3) Patient's Assessment of Arthritis Pain-Visual Analog Scale, 4) Patient's Global Assessment of Disease Activity-Visual Analog Scale, 5) Physician's Global Assessment of Disease Activity-Visual Analog Scale. Missing values were imputed using Non-Responder Imputation (NRI)

Of the 208 subjects in the Full Analysis Set (FAS) 208 subjects (69 400 mg CZP, 70 200 mg CZP, 69 placebo) are included in this analysis which uses Non-Response Imputation (NRI).

ACR70 responders are subjects with at least 70% improvement from Baseline for tender joint count (TJC), swollen joint count (SJC), and at least 3 of the 5 remaining core set measures: 1) Health Assessment Questionnaire-Disability Index (HAQ-DI), 2) C-reactive Protein (CRP), 3) Patient's Assessment of Arthritis Pain-Visual Analog Scale, 4) Patient's Global Assessment of Disease Activity-Visual Analog Scale, 5) Physician's Global Assessment of Disease Activity-Visual Analog Scale. Missing values were imputed using Non-Responder Imputation (NRI)

Of the 208 subjects in the Full Analysis Set (FAS) 208 subjects (69 400 mg CZP, 70 200 mg CZP, 69 placebo) are included in this analysis which uses Non-Response Imputation (NRI).

DAS28-ESR is calculated using the tender joint count (TJC), swollen joint count (SJC) erythrocyte sedimentation rate (ESR in mm/hour), and the Patient's Global Assessment of Disease Activity - Visual Analog Scale (VAS in mm) using the following formula: 0.56 x √(TJC) + 0.28 x √(SJC) + 0.70 x lognat (ESR) + 0.014 x Global Assessment of Arthritis where 28 joints are examined and a lower score indicates less disease activity. This analysis was carried out using an ANCOVA model on Last Observation Carried Forward.

< 2.6 Remission,

> = 2.6 - < =3.2 Low, > 3.2 - < = 5.1 Moderate, > 5.1 High

Of the 208 subjects in the Full Analysis Set (FAS) 204 subjects (69 400 mg CZP, 68 200 mg CZP, 67 placebo) are included in this analysis using the Last Observation Carried Forward (LOCF) method.

SDAI is calculated as the sum of tender joint count (TJC), swollen joint count (SJC), C-reactive protein (CRP in mg/dL), Patient's Global Assessment of Disease Activity - Visual Analog Scale (VAS in cm), and Investigator's Global Assessment of Disease Activity - Visual Analog Scale (VAS in cm). 28 joints are examined where a lower score indicates less disease activity. This analysis was carried out using an ANCOVA model on Last Observation Carried Forward (LOCF) data with factors treatment and Baseline score.

<= 3.3 Remission, > 3.3 - <= 11 Low, > 11 - <= 26 Moderate, > 26 High

Of the 208 subjects in the Full Analysis Set (FAS) 205 subjects (68 400 mg CZP, 70 200 mg CZP, 67 placebo) are included in this analysis using the Last Observation Carried Forward (LOCF) method.

CDAI is calculated as the sum of tender joint count (TJC), swollen joint count (SJC), Patient's Global Assessment of Disease Activity - Visual Analog Scale (VAS in cm), and Investigator's Global Assessment of Disease Activity - Visual Analog Scale (VAS in cm). 28 joints are examined where a lower score indicates less disease activity. This analysis was carried out using an ANCOVA model on Last Observation Carried Forward (LOCF) data with factors treatment and Baseline score. Range for CDAI is 0-76 with a lower CDAI score reflects approvement in activity and a higher score reflects a decline.

Of the 208 subjects in the Full Analysis Set (FAS) 206 subjects (69 400 mg CZP, 70 200 mg CZP, 67 placebo) are included in this analysis using the Last Observation Carried Forward (LOCF) method.

DAS28-ESR is calculated using the tender joint count (TJC), swollen joint count (SJC) erythrocyte sedimentation rate (ESR in mm/hour), and the Patient's Global Assessment of Disease Activity - Visual Analog Scale (VAS in mm) using the following formula: 0.56 x √(TJC) + 0.28 x √(SJC) + 0.70 x lognat (ESR) + 0.014 x Global Assessment of Arthritis where 28 joints are examined and a lower score indicates less disease activity. Missing values were imputed using Non-Responder Imputation (NRI).

< 2.6 (Remission),

> = 2.6 - < =3.2 Low, > 3.2 - < = 5.1 Moderate, > 5.1 High

Of the 208 subjects in the Full Analysis Set (FAS) 208 subjects (69 400 mg CZP, 70 200 mg CZP, 69 placebo) are included in this analysis which uses Non-Response Imputation (NRI).

SDAI is calculated as the sum of tender joint count (TJC), swollen joint count (SJC), C-reactive protein (CRP in mg/dL), Patient's Global Assessment of Disease Activity - Visual Analog Scale (VAS in cm), and Investigator's Global Assessment of Disease Activity - Visual Analog Scale (VAS in cm). 28 joints are examined where a lower score indicates less disease activity. Missing values were imputed using Non-Responder Imputation (NRI).

<= 3.3 (Remission), > 3.3 - <= 11 Low, > 11 - <= 26 Moderate, > 26 High

Of the 208 subjects in the Full Analysis Set (FAS) 208 subjects (69 400 mg CZP, 70 200 mg CZP, 69 placebo) are included in this analysis which uses Non-Response Imputation (NRI).

CDAI is calculated as the sum of tender joint count (TJC), swollen joint count (SJC), Patient's Global Assessment of Disease Activity - Visual Analog Scale (VAS in cm), and Investigator's Global Assessment of Disease Activity - Visual Analog Scale (VAS in cm). 28 joints are examined where a lower score indicates less disease activity. Missing values were imputed using Non-Responder Imputation (NRI).

The range for the CDAI is 0 - 76 with a lower CDAI score indicating approvement in activity and a higher score indicating a decline activity.

Of the 208 subjects in the Full Analysis Set (FAS) 208 subjects (69 400 mg CZP, 70 200 mg CZP, 69 placebo) are included in this analysis which uses Non-Response Imputation (NRI).

Ratio is defined as the CRP value at Week 34 divided by the CRP value at Baseline. This analysis was carried out using the Last Observation Carried Forward (LOCF) method with an ANCOVA model on observed log transformed data with factors treatment and log transformed Baseline CRP level. The number presented is the geometric least squares mean with it's 95% confidence interval.

Of the 208 subjects in the Full Analysis Set (FAS) 207 subjects (69 400 mg CZP, 70 200 mg CZP, 68 placebo) are included in this analysis using the Last Observation Carried Forward (LOCF) method.

HAQ-DI is derived based on the mean of individual scores in 8 categories of daily living actives (using 20 questions). Each question is scored 0-3 (0 = without any difficulty, 1 = with some difficulty, 2 = with much difficulty, and 3 = unable to do). Thus, the mean also has a range from 0-3. Change from baseline is computed as the value at Week 34 minus the baseline value. A negative value in change from baseline indicates an improvement. This analysis was carried out using an ANCOVA model on Last Observation Carried Forward (LOCF) data with factors treatment and Baseline score.

Of the 208 subjects in the Full Analysis Set (FAS) 207 subjects (69 400 mg CZP, 70 200 mg CZP, 68 placebo) are included in this analysis using the Last Observation Carried Forward (LOCF) method.

Change from Baseline in Fatigue Assessment scale (0 to 10, 0 is "No Fatigue" and 10 is "Fatigue as bad as you can imagine") is computed as the value at Week 34 minus the Baseline value. A negative value in change from baseline indicates an improvement. This analysis was carried out using an ANCOVA model on Last Observation Carried Forward (LOCF) data with factors treatment and Baseline score.

Of the 208 subjects in the Full Analysis Set (FAS) 205 subjects (68 400 mg CZP, 70 200 mg CZP, 67 placebo) are included in this analysis using the Last Observation Carried Forward (LOCF) method.

There are 8 SF-36 domain scores: Physical Functioning, Role Physical, Bodily Pain, General Health, Vitality, Social Functioning, Role Emotional and Mental Health, each ranging from 0 to 100, with higher scores indicating better health. A larger positive value in change from Baseline indicates an improvement. This analysis was carried out using an ANCOVA model on Last Observation Carried Forward (LOCF) data with factors treatment and Baseline score.

Of the 208 subjects in the Full Analysis Set (FAS) 207 subjects (69 400 mg CZP, 70 200 mg CZP, 68 placebo) are included in this analysis using the Last Observation Carried Forward (LOCF) method.

There are 8 SF-36 domain scores: Physical Functioning, Role Physical, Bodily Pain, General Health, Vitality, Social Functioning, Role Emotional and Mental Health, each ranging from 0 to 100, with higher scores indicating better health. A larger positive value in change from Baseline indicates an improvement. This analysis was carried out using an ANCOVA model on Last Observation Carried Forward (LOCF) data with factors treatment and Baseline score.

Of the 208 subjects in the Full Analysis Set (FAS) 206 subjects (69 400 mg CZP, 69 200 mg CZP, 68 placebo) are included in this analysis using the Last Observation Carried Forward (LOCF) method.

There are 8 SF-36 domain scores: Physical Functioning, Role Physical, Bodily Pain, General Health, Vitality, Social Functioning, Role Emotional and Mental Health, each ranging from 0 to 100, with higher scores indicating better health. A larger positive value in change from Baseline indicates an improvement. This analysis was carried out using an ANCOVA model on Last Observation Carried Forward (LOCF) data with factors treatment and Baseline score.

Of the 208 subjects in the Full Analysis Set (FAS) 206 subjects (69 400 mg CZP, 69 200 mg CZP, 68 placebo) are included in this analysis using the Last Observation Carried Forward (LOCF) method.

There are 8 SF-36 domain scores: Physical Functioning, Role Physical, Bodily Pain, General Health, Vitality, Social Functioning, Role Emotional and Mental Health, each ranging from 0 to 100, with higher scores indicating better health. A larger positive value in change from Baseline indicates an improvement. This analysis was carried out using an ANCOVA model on Last Observation Carried Forward (LOCF) data with factors treatment and Baseline score.

Of the 208 subjects in the Full Analysis Set (FAS) 205 subjects (68 400 mg CZP, 69 200 mg CZP, 68 placebo) are included in this analysis using the Last Observation Carried Forward (LOCF) method.

There are 8 SF-36 domain scores: Physical Functioning, Role Physical, Bodily Pain, General Health, Vitality, Social Functioning, Role Emotional and Mental Health, each ranging from 0 to 100, with higher scores indicating better health. A larger positive value in change from Baseline indicates an improvement. This analysis was carried out using an ANCOVA model on Last Observation Carried Forward (LOCF) data with factors treatment and Baseline score.

Of the 208 subjects in the Full Analysis Set (FAS) 205 subjects (68 400 mg CZP, 69 200 mg CZP, 68 placebo) are included in this analysis using the Last Observation Carried Forward (LOCF) method.

There are 8 SF-36 domain scores: Physical Functioning, Role Physical, Bodily Pain, General Health, Vitality, Social Functioning, Role Emotional and Mental Health, each ranging from 0 to 100, with higher scores indicating better health. A larger positive value in change from Baseline indicates an improvement. This analysis was carried out using an ANCOVA model on Last Observation Carried Forward (LOCF) data with factors treatment and Baseline score.

Of the 208 subjects in the Full Analysis Set (FAS) 207 subjects (69 400 mg CZP, 70 200 mg CZP, 68 placebo) are included in this analysis using the Last Observation Carried Forward (LOCF) method.

There are 8 SF-36 domain scores: Physical Functioning, Role Physical, Bodily Pain, General Health, Vitality, Social Functioning, Role Emotional and Mental Health, each ranging from 0 to 100, with higher scores indicating better health. A larger positive value in change from Baseline indicates an improvement. This analysis was carried out using an ANCOVA model on Last Observation Carried Forward (LOCF) data with factors treatment and Baseline score.

Of the 208 subjects in the Full Analysis Set (FAS) 204 subjects (67 400 mg CZP, 69 200 mg CZP, 68 placebo) are included in this analysis using the Last Observation Carried Forward (LOCF) method.

There are 8 SF-36 domain scores: Physical Functioning, Role Physical, Bodily Pain, General Health, Vitality, Social Functioning, Role Emotional and Mental Health, each ranging from 0 to 100, with higher scores indicating better health. A larger positive value in change from Baseline indicates an improvement. This analysis was carried out using an ANCOVA model on Last Observation Carried Forward (LOCF) data with factors treatment and Baseline score.

Of the 208 subjects in the Full Analysis Set (FAS) 205 subjects (68 400 mg CZP, 69 200 mg CZP, 68 placebo) are included in this analysis using the Last Observation Carried Forward (LOCF) method.

PCS norm-based scores are calculated based upon the following 8 domain scores, Physical Functioning, Role Physical, Bodily Pain, General Health, Vitality, Social Functioning, Role Emotional and Mental Health, and range from 1 to 81, where 50 represents the normative value. A larger positive value in change from Baseline indicates an improvement. This analysis was carried out using an ANCOVA model on Last Observation Carried Forward (LOCF) data with factors treatment and Baseline score.

Of the 208 subjects in the Full Analysis Set (FAS) 202 subjects (66 400 mg CZP, 68 200 mg CZP, 68 placebo) are included in this analysis using the Last Observation Carried Forward (LOCF) method.

MCS norm-based scores are calculated based upon the following 8 domain scores, Physical Functioning, Role Physical, Bodily Pain, General Health, Vitality, Social Functioning, Role Emotional and Mental Health, and range from -9 to 82, where 50 represents the normative value. A larger positive value in change from Baseline indicates an improvement. This analysis was carried out using an ANCOVA model on Last Observation Carried Forward (LOCF) data with factors treatment and Baseline score.

Of the 208 subjects in the Full Analysis Set (FAS) 202 subjects (66 400 mg CZP, 68 200 mg CZP, 68 placebo) are included in this analysis using the Last Observation Carried Forward (LOCF) method.

Change from Baseline in Patient's Assessment of Arthritis Pain-VAS (0 to 100 mm visual analog scale, 0 being no pain and 100 being most severe pain) is computed as the value at Week 34 minus the Baseline value. A negative value in change from Baseline indicates an improvement. This analysis was carried out using an ANCOVA model on Last Observation Carried Forward (LOCF) data with factors treatment and Baseline score.

Of the 208 subjects in the Full Analysis Set (FAS) 207 subjects (69 400 mg CZP, 70 200 mg CZP, 68 placebo) are included in this analysis using the Last Observation Carried Forward (LOCF) method.

Change from Baseline in Patient's Global Assessment of Disease Activity-VAS (0 to 100 mm visual analog scale, 0 being no symptoms and 100 being severe symptoms) is computed as the value at Week 34 minus the Baseline value. A negative value in change from Baseline indicates an improvement. This analysis was carried out using an ANCOVA model on Last Observation Carried Forward (LOCF) data with factors treatment and Baseline score.

Of the 208 subjects in the Full Analysis Set (FAS) 207 subjects (69 400 mg CZP, 70 200 mg CZP, 68 placebo) are included in this analysis using the Last Observation Carried Forward (LOCF) method.

ACR20 loss are subjects with <20% improvement from Baseline for tender joint count, swollen joint count, and at least 3/5 core set measures: 1) Health Assessment Questionnaire-Disability Index (HAQ-DI), 2) C-reactive Protein, 3) Patient's Assessment of Arthritis Pain-Visual Analog Scale, 4) Patient's Global Assessment of Disease Activity-Visual Analog Scale, 5) Physician's Global Assessment of Disease Activity-Visual Analog Scale at 2 consecutive visits. Subjects losing response for 2 consecutive visits are considered as having the event on the day of the visit where response was first lost.

Of the 208 subjects in the Full Analysis Set (FAS) 208 subjects (69 400 mg CZP, 70 200 mg CZP, 69 placebo) are included in this analysis.

Ratio is defined as the ESR value at Week 16 divided by the ESR value at Baseline. This analysis was carried out using the Last Observation Carried Forward (LOCF) method.

Of the 333 subjects in the Run-in period, 328 are included in this analysis using the Last Observation Carried Forward (LOCF) method.

Ratio is defined as the ESR value at Week 34 divided by the ESR value at Baseline. This analysis was carried out using the Last Observation Carried Forward (LOCF) method with an ANCOVA model on observed log transformed data with factors treatment and log transformed Baseline CRP level. The number presented is the geometric least squares mean with it's 95% confidence interval.

Of the 208 subjects in the Full Analysis Set (FAS) 205 subjects (69 400 mg CZP, 68 200 mg CZP, 68 placebo) are included in this analysis using the Last Observation Carried Forward (LOCF) method.

Change from Baseline in Physician's Global Assessment of Disease Activity-VAS (0 to 100 mm visual analog scale, 0 being no symptoms and 100 being severe symptoms) is computed as the value at Week 34 minus the Baseline value. A negative value in change from Baseline indicates an improvement. This analysis was carried out using an ANCOVA model on Last Observation Carried Forward (LOCF) data with factors treatment and Baseline score.

Of the 208 subjects in the Full Analysis Set (FAS) 206 subjects (69 400 mg CZP, 70 200 mg CZP, 67 placebo) are included in this analysis using the Last Observation Carried Forward (LOCF) method.

Change from Baseline in Tender Joint Count is computed as the value at Week 34 minus the Baseline value (28 joints were assessed at each visit). A negative value in change from Baseline indicates an improvement. This analysis was carried out using an ANCOVA model on Last Observation Carried Forward (LOCF) data with factors treatment and Baseline score.

Of the 208 subjects in the Full Analysis Set (FAS) 207 subjects (69 400 mg CZP, 70 200 mg CZP, 68 placebo) are included in this analysis using the Last Observation Carried Forward (LOCF) method.

Change from Baseline in Swollen Joint Count is computed as the value at Week 34 minus the Baseline value (28 joints were assessed at each visit). A negative value in change from Baseline indicates an improvement. This analysis was carried out using an ANCOVA model on Last Observation Carried Forward (LOCF) data with factors treatment and Baseline score.

Of the 208 subjects in the Full Analysis Set (FAS) 207 subjects (69 400 mg CZP, 70 200 mg CZP, 68 placebo) are included in this analysis using the Last Observation Carried Forward (LOCF) method.