A Study to Evaluate the Safety and Efficacy of MabThera (Rituximab) in Combination With Methotrexate (MTX) in Participants With Active Rheumatoid Arthritis Who Failed on Anti-Tumor Necrosis Factor Alpha Therapy

American College of Rheumatology (ACR) 20 response is defined as \>= 20% improvement (reduction) in score compared with baseline for both tender joint count (TJC)-68 joints and swollen joint count (SJC)-66 joints, as well as for 3 of the additional 5 ACR core set variables: Patient's Assessment of Pain over the previous 24 hours using a Visual Analog Scale (VAS) ranging from score 0 (no pain) to 100 (unbearable pain); Patient's Global Assessment of Disease Activity and Physician's Global Assessment of Disease Activity over the previous 24 hours using a VAS ranging score 0 (no disease activity) to 100 (maximum disease activity); Health Assessment Questionnaire (HAQ):8 domains (dressing/grooming, arising, eating, walking, hygiene, reach, grip; common daily activities) rated on a 4-point scale (0=without any difficulty to 3=unable to do) for a total possible score of 0 (best) to 3 (worst); and acute-phase reactant, either C-reactive protein (CRP) or erythrocyte sedimentation rate (ESR).

Time frame: Week 24

Population: Intent to treat (ITT) population included all randomized participants who received any part of an infusion of study drug. Participants whom treatment allocation was unblinded and who received treatment prior to randomization were excluded.

The disease activity score (DAS28) is an evaluation index of rheumatoid arthritis. The DAS28 applies a mathematical formula based on the following parameters: 1. TJC-28 joints, 2. SJC -28 joints, 3. ESR or CRP measurement, 4. Participant's judgement on his own overall health (global health \[GH\]) status expressed by a VAS (0 \[no disease activity\] to 100 \[maximum disease activity\]). The mathematical formula is 0.56 × √28TJC + 0.28 × √28SJC + 0.7 x loge ESR + 0.014 × GH. The DAS28 scale ranges from score of 0 to 10, where lower scores indicate best disease control and higher scores indicate worsening of disease. Change from Baseline = difference between the score at Week 24 and the score at Baseline.

Time frame: From Baseline (Day 1) to Week 24

Population: The ITT population included all randomized participants who received any part of an infusion of study drug. Participants whom treatment allocation was unblinded and who received treatment prior to randomization were excluded.

The Genant-modified Sharp scoring system assesses structural damage due to rheumatoid arthritis in radiographs. A score for erosions of 0-3.5 (8 gradations) is assigned for 14 joints in each hand and wrist, and 6 joints in each foot. Joint space narrowing scores of 0-4 (9 gradations) are assigned to 13 joints in each hand and 6 joints in each foot. The maximum erosion score is 40 x 3.5 = 140. The maximum joint space narrowing score is 38 x 4.0 = 152. Both the erosion and joint space narrowing scores are normalized to 145 and are added together for a maximum total Genant-modified Sharp score of 290. For all the three radiograph assessment, the minimum score is 0. A higher score indicates more damage and a negative change score indicates improvement. The change in score is to be calculated as: Change from Baseline = difference between the score at Weeks 24, 56, or 104 and the score at Baseline.

Time frame: From Baseline (Day 1) to Weeks (W) 24, 56, and 104

Population: The ITT population included all randomized participants who received any part of an infusion of study drug. Participants whom treatment allocation was unblinded and who received treatment prior to randomization were excluded.

The Short Form (SF)-36 determined participants' overall quality of life by assessing 1) limitations in physical functioning due to health problems; 2) limitations in usual daily activities because of physical health problems; 3) bodily pain; 4) general health perceptions; 5) vitality; 6) limitations in social functioning because of physical or emotional problems since last month; 7) limitations in usual work (house hold and outside) due to emotional problems 8) current and 1 year past status of health 9) general mental health. Transforming and standardizing these domains leads to the calculation of the physical component summary and mental component summary measures. Scores on each item were summed and averaged (range 0 \[worst\] to 100 \[best\]); increase in score from baseline indicated improvement. Change from Baseline = difference between the score at Week 24 and the score at Baseline.

Time frame: From Baseline (Day 1) to Week 24

Population: The ITT population included all randomized participants who received any part of an infusion of study drug. Participants with available data at the time of evaluation were analyzed.

ACR 70 response is defined as a \>= 70% improvement (reduction) in score compared with baseline for both TJC -68 joints and SJC -66 joints, as well as for 3 of the additional 5 ACR core set variables: Patient's Assessment of Pain over the previous 24 hours: using a VAS ranging from score '0'=no pain to score '100'=unbearable pain; Patient's Global Assessment of Disease Activity and Physician's Global Assessment of Disease Activity over the previous 24 hours using a VAS ranging from score '0'=no disease activity to score '100'=maximum disease activity; HAQ : Health Assessment Questionnaire (HAQ):which included 8 domains (dressing/grooming, arising, eating, walking, hygiene, reach, grip; common daily activities) rated on a 4-point scale (0=without any difficulty to 3=unable to do), where the sum of scores was divided by the number of domains with a score for a total possible score of 0 (best) to 3 (worst); and acute-phase reactant, either CRP or ESR.

Time frame: Week 24

Population: The ITT population included all randomized participants who received any part of an infusion of study drug. Participants whom treatment allocation was unblinded and who received treatment prior to randomization were excluded.

ACR 50 response is defined as a \>= 50% improvement (reduction) in score compared with baseline for both TJC -68 joints and SJC -66 joints, as well as for 3 of the additional 5 ACR core set variables: Patient's Assessment of Pain over the previous 24 hours: using a VAS ranging from score '0'=no pain to score '100'=unbearable pain; Patient's Global Assessment of Disease Activity and Physician's Global Assessment of Disease Activity over the previous 24 hours using a VAS ranging from score '0'=no disease activity to score '100'=maximum disease activity; HAQ : Health Assessment Questionnaire (HAQ):which included 8 domains (dressing/grooming, arising, eating, walking, hygiene, reach, grip; common daily activities) rated on a 4-point scale (0=without any difficulty to 3=unable to do), where the sum of scores was divided by the number of domains with a score for a total possible score of 0 (best) to 3 (worst); and acute-phase reactant, either CRP or ESR.

Time frame: Week 24

Population: The ITT population included all randomized participants who received any part of an infusion of study drug. Participants whom treatment allocation was unblinded and who received treatment prior to randomization were excluded.

The SF-36 determined participants' overall quality of life by assessing :1) limitations in physical functioning due to health problems; 2) limitations in usual daily activities because of physical health problems; 3) bodily pain; 4) general health perceptions; 5) vitality; 6) limitations in social functioning because of physical or emotional problems since last month; 7) limitations in usual work (house hold and outside) due to emotional problems 8) current and 1 year past status of health 9) general mental health. Scores on physical component were summed and averaged (range 0 \[worst\] to 100 \[best\]); If participants' had shown change from baseline in physical health component score \>5.42, it was considered as improved; score between -5.42 to 5.42 was considered as unchanged, and score \< -5.42 was considered as worsened. Change from Baseline = difference between the score of physical component at Week 24 and the score at Baseline.

Time frame: From Baseline (Day 1) to Week 24

Population: The ITT population included all randomized participants who received any part of an infusion of study drug. Participants with available data at the time of evaluation were analyzed.

The SF-36 determined participants' overall quality of life by assessing :1) limitations in physical functioning due to health problems; 2) limitations in usual daily activities because of physical health problems; 3) bodily pain; 4) general health perceptions; 5) vitality; 6) limitations in social functioning because of physical or emotional problems since last month; 7) limitations in usual work (house hold and outside) due to emotional problems 8) current and 1 year past status of health 9) general mental health. Scores on mental component were summed and averaged (range = 0 \[worst\]-100 \[best\]); increase from baseline indicated improvement. If participants' had shown change from baseline in mental health score \>6.33, it was considered as improved; scores between -6.33 to 6.33 was considered unchanged, and score \<-6.33 was considered as worsened. Change from Baseline = difference between the mental component score at Week 24 and the score at Baseline.

Time frame: From Baseline (Day 1) to Week 24

Population: The ITT population included all randomized participants who received any part of an infusion of study drug. Participants with available data at the time of evaluation were analyzed.

European League Against Rheumatism (EULAR) response is defined based on the DAS28 score and the EULAR response criteria (Van Gestel et al, 1996 and 1999). The DAS28 scale ranges from score of 0 to 10, where lower scores indicate best disease control and higher scores indicate worsening of disease. At a given visit, participants with a DAS28 score of \< 3.2 are considered good responders if the change from baseline in their DAS28 score is \>1.2. Participants with a DAS28 score \>= 3.2 to 5.1 are considered moderate responders if the change from baseline in their DAS28 score is \<=1.2 to \>=0.6. Participants with DAS28 score \>5.1 are considered non-responders if the change from baseline in their DAS28 score is \<=1.2 to \>=0.6.

Time frame: Week 24

Population: The ITT population included all randomized participants who received any part of an infusion of study drug. Participants whom treatment allocation was unblinded and who received treatment prior to randomization were excluded.

Percentage change in the scores of the following parameters of ACR core set relative to respective baseline scores in both study arms was analyzed : SJC (28 and 66 joints) and TJC (28 and 66 joints), patient's global assessment and physician's global assessment based on disease activity (both are expressed by VAS \[0 = no disease activity to 100 = maximum disease activity\]), HAQ (based on HAQ disability index \[HAQDI\]) which included 8 domains (dressing/grooming, arising, eating, walking, hygiene, reach, grip; common daily activities) rated on a 4-point scale (0=without any difficulty to 3=unable to do), where the sum of scores was divided by the number of domains with a score for a total possible score of 0 (best) to 3 (worst), pain assessment using a VAS ranging from score 0 (no pain) to 100 (unbearable pain), CRP concentration, and ESR.

Time frame: From Baseline (Day 1) to Week 24

Population: The ITT population included all randomized participants who received any part of an infusion of study drug. Participants whom treatment allocation was unblinded and who received treatment prior to randomization were excluded.

The DAS28 is an evaluation index of RA. The DAS28 applies a mathematical formula based on the following parameters: 1. TJC- 28 joints, 2. SJC- 28 joints, 3. ESR or CRP measurement, 4. Participant's judgement on his own overall health status (GH) expressed by a visual analogue scale VAS (0 \[no disease activity\] to 100 \[maximum disease activity\]). The mathematical formula is 0.56 × √28TJC + 0.28 × √28SJC + 0.7 x loge ESR + 0.014 × GH. The DAS28 scale ranges from score of 0 to 10. A participant was categorized as having low disease activity, if participant's DAS28 score was \<= 3.2, and was categorized as having clinical remission if participant's DAS28 score was \< 2.6.

Time frame: Week 24

Population: The ITT population included all randomized participants who received any part of an infusion of study drug. Participants whom treatment allocation was unblinded and who received treatment prior to randomization were excluded.

The Genant-modified Sharp scoring system assesses structural damage due to rheumatoid arthritis in radiographs. A score for erosions of 0-3.5 (8 gradations) is assigned for 14 joints in each hand and wrist, and 6 joints in each foot. The maximum erosion score is 40 x 3.5 = 140 which is normalized to 145. The minimum score is 0 and the maximum score is 145. A higher score indicates more damage and negative change score indicates improvement.

Time frame: Up to Week 104

Population: ITT population included all randomized participants who received any part of an infusion of study medication. Participants with available data at the time of evaluation were analyzed.