A Study of Tocilizumab in Combination With Disease-Modifying Anti-Rheumatic Drugs (DMARDs) in Participants With Moderate to Severe Active Rheumatoid Arthritis With an Inadequate Response to DMARDs

Synovitis of the wrist was assessed at three sites including the radioulnar joint (RUJ), the radiocarpal joint (RCJ), and the intercarpal-carpometacarpal joints (IC-CMCJ). Global RAMRIS scores were assigned on a scale of 0 to 3 at each site, where 0 represented normal appearance with no synovial enhancement and each 1-point increase reflected one-third of the presumed maximum volume of enhancing tissue in the synovial compartment. The scores for all three sites were added to give an aggregated score of 0 to 9. Baseline absolute value (AV) and change from Baseline to Week 4 were averaged among all participants, where negative changes indicated improvement in synovitis.

RE was calculated as [S0 minus (-) S55] divided by (÷) S0, multiplied by (×) 100 percent (%), where S0 was defined as the signal noise ratio before contrast injection, and S55 was defined as the signal noise ratio 55 seconds after injection. Signal noise ratios were measured as the ratio between the signal in the region of interest and the standard deviation of background noise. Baseline AV and change from Baseline to Week 4 were averaged among all participants, where negative changes indicated improvement in synovitis.

REE per second was calculated as [S55 - S0] ÷ [S0 × 55 seconds] × 100%, where S0 was defined as the signal noise ratio before contrast injection, and S55 was defined as the signal noise ratio 55 seconds after injection. Signal noise ratios were measured as the ratio between the signal in the region of interest and the standard deviation of background noise. Baseline AV and change from Baseline to Week 4 were averaged among all participants, where negative changes indicated improvement in synovitis.

Synovitis of the wrist was assessed at three sites including the RUJ, the RCJ, and the IC-CMCJ. Global RAMRIS scores were assigned on a scale of 0 to 3 at each site, where 0 represented normal appearance with no synovial enhancement and each 1-point increase reflected one-third of the presumed maximum volume of enhancing tissue in the synovial compartment. The scores for all three sites were added to give an aggregated score of 0 to 9. The changes from Baseline to Weeks 2, 12, 24, and 48 were averaged among all participants, where negative changes indicated improvement in synovitis.

RE was calculated as [S0 - S55] ÷ S0 × 100%, where S0 was defined as the signal noise ratio before contrast injection, and S55 was defined as the signal noise ratio 55 seconds after injection. Signal noise ratios were measured as the ratio between the signal in the region of interest and the standard deviation of background noise. The changes from Baseline to Weeks 2, 12, 24, and 48 were averaged among all participants, where negative changes indicated improvement in synovitis.

REE per second was calculated as [S55 - S0] ÷ [S0 × 55 seconds] × 100%, where S0 was defined as the signal noise ratio before contrast injection, and S55 was defined as the signal noise ratio 55 seconds after injection. Signal noise ratios were measured as the ratio between the signal in the region of interest and the standard deviation of background noise. The changes from Baseline to Weeks 2, 12, 24, and 48 were averaged among all participants, where negative changes indicated improvement in synovitis.

Synovitis of the wrist was assessed at three sites including the RUJ, the RCJ, and the IC-CMCJ. Global RAMRIS scores were assigned on a scale of 0 to 3 at each site, where 0 represented normal appearance with no synovial enhancement and each 1-point increase reflected one-third of the presumed maximum volume of enhancing tissue in the synovial compartment. The scores for all three sites were added to give an aggregated score of 0 to 9. Synovitis of MCP joints was determined on the basis of Short Inversion Time Inversion Recovery (STIR) sequence evaluation with modification of the RAMRIS score. Four MCP joint compartments were each assessed 0 to 3, so the aggregated MCP joint score ranged from 0 to 12. Combined synovitis in wrist and MCP joints was determined on the basis of STIR sequences to produce overall score from 0 to 21. Baseline AV and changes from Baseline to Weeks 2, 4, 12, 24, 48 were averaged among all participants, where negative changes indicated improvement in synovitis.

Erosion was evaluated at 15 sites in the wrist and 8 sites in the MCP joints. Bone erosion was scored on a scale of 0 to 10, where 0 represented no bone erosion and each 1-point increase reflected up to a 10% increase in extent of erosion. The number of bones with erosion was taken as the count of joints with a bone erosion score greater than or equal to (≥) 1. Baseline AV and changes from Baseline to Weeks 2, 4, 12, 24, and 48 were averaged among all participants, where negative changes indicated improvement in erosion.

Erosion was evaluated at 15 sites in the wrist and 8 sites in the MCP joints. Bone erosion was scored on a scale of 0 to 10, where 0 represented no bone erosion and each 1-point increase reflected up to 10% increase in extent of erosion. Global RAMRIS scores were calculated as the sum of all joint sites for the wrist (range, 0 to 150) and MCP joints (range, 0 to 80). Aggregate wrist and MCP joint scores could range from 0 to 230 points. Baseline AV and changes from Baseline to Weeks 2, 4, 12, 24, and 48 were averaged among all participants, where negative changes indicated improvement in erosion.

Edema was evaluated at 15 sites in the wrist and 8 sites in the MCP joints. Bone edema was scored on a scale of 0 to 3, where 0 represented no bone edema and each 1-point increase reflected one-third increase in extent of edema. The number of bones with edema was taken as the count of joints with a bone edema score greater than or equal to (≥) 1. Baseline AV and changes from Baseline to Weeks 2, 4, 12, 24, and 48 were averaged among all participants, where negative changes indicated improvement in edema.

Edema was evaluated at 15 sites in the wrist and 8 sites in the MCP joints. Bone edema was scored on a scale of 0 to 3, where 0 represented no bone edema and each 1-point increase reflected one-third increase in extent of edema. Global RAMRIS scores were calculated as the sum of all joint sites for the wrist (range, 0 to 45) and MCP joints (range, 0 to 24). Aggregate wrist and MCP joint scores could range from 0 to 69 points. Baseline AV and changes from Baseline to Weeks 2, 4, 12, 24, and 48 were averaged among all participants, where negative changes indicated improvement in edema.

The TMSS was calculated as the sum of ES and JSNS and ranged from 0 to 202. The ES was taken as the sum of joint scores collected for 14 joints in each hand (individually scored from 0 to 7) and ranged from 0 to 98 for both hands. The JSNS was the sum of joint scores collected for 13 joints in each hand (individually scored from 0 to 8) and ranged from 0 to 104 for both hands. Scores of 0 reflected no change, while higher scores reflected increased disease activity. Baseline AV and changes from Baseline to Weeks 24 and 48 were averaged among all participants, where negative changes indicated improvement in disease activity.

The Ritchie Articular Index was scored on a scale of 0 to 3, according to the grades of tenderness in each of 26 assessed joints. The total score was taken as the sum of joint scores and ranged from 0 to 78. Scores of 0 reflected no tenderness, while higher scores reflected increased tenderness. Baseline AV and changes from Baseline to Weeks 2, 4, 12, 24, and 48 were averaged among all participants, where negative changes indicated improvement in joint tenderness.

Perceived pain was assessed on a 0- to 100-millimeter (mm) VAS, where the distance from 0 mm represented the participant's self evaluation of pain (0 mm = no pain, 100 mm = maximum pain). Baseline AV and changes from Baseline to Weeks 2, 4, 12, 24, and 48 were averaged among all participants, where negative changes indicated a decrease in perceived pain.

Global assessment of disease activity was performed using a 0- to 100-mm VAS, where the distance from 0 mm represented the investigator's evaluation or the participant's self evaluation of disease activity (0 mm = no disease activity, 100 mm = maximum disease activity). Baseline AV and changes from Baseline to Weeks 2, 4, 12, 24, and 48 were averaged among all participants, where negative changes indicated improvement in disease activity.

The HAQ-DI assessed 20 items in eight functional activity domains including dressing, rising, eating, walking, hygiene, reach, grip, and usual activities. Each item was scored on a scale of 0 to 3, where 0 represented activities performed without difficulty and 3 represented inability to perform activities alone. The total score was calculated as an average of all item scores, and thus also ranged from 0 to 3. Baseline AV and changes from Baseline to Weeks 2, 4, 12, 24, and 48 were averaged among all participants, where negative changes indicated an increase in ability to perform activities independently.

The DAS28 was derived from assessments of C-reactive protein (CRP), tender joint count (TJC), swollen joint count (SJC), and general health according to 100-mm VAS. DAS28 scores were calculated as [0.56 × square root of TJC] plus (+) [0.28 × square root of SJC] + [0.36 × natural log (CRP + 1)] + [0.014 × VAS] + 0.96. TJC was defined as the number of painful joints and SJC was defined as the number of swollen joints, each assessed on 28 joints. CRP was measured in milligrams per deciliter (mg/dL). DAS28 scores could range from 0 to 10, where higher scores represented higher disease activity. Baseline AV and changes from Baseline to Weeks 2, 4, 12, 24, and 48 were averaged among all participants, where negative changes indicated an improvement in disease activity.

Level of VEGF was measured in picograms per milliliter (pg/mL). Baseline AV and changes from Baseline to Weeks 2, 4, 12, 24, and 48 were averaged among all participants.

ESR was measured in millimeters per hour (mm/h). Baseline AV and changes from Baseline to Weeks 2, 4, 12, 24, and 48 were averaged among all participants.

Level of hsCRP was measured in mg/dL. Baseline AV and changes from Baseline to Weeks 2, 4, 12, 24, and 48 were averaged among all participants.

Level of sIL6R was measured in pg/mL. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

Level of mRNA for IL-17 (2^ΔCt) was quantified by polymerase chain reaction (PCR). Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

Level of mRNA for IL-23 receptor (2^ΔCt) was quantified by PCR. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

Level of ROR-γT (2^ΔCt) was quantified by PCR. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

Level of mRNA for FOXP3 (2^ΔCt) was quantified by PCR. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The intensity of CD4-positive cell infiltration was expressed as the percentage of PBMCs. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The mean fluorescence intensity of CD4-positive cells was measured by flow cytometry. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The intensity of CD25-positive cell infiltration was expressed as the percentage of PBMCs. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The mean fluorescence intensity of CD25-positive cells was measured by flow cytometry. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The intensity of CD45RO-positive cell infiltration was expressed as the percentage of PBMCs. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The mean fluorescence intensity of CD45RO-positive cells was measured by flow cytometry. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The intensity of CCR6-positive cell infiltration was expressed as the percentage of PBMCs. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The mean fluorescence intensity of CCR6-positive cells was measured by flow cytometry. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The intensity of CCR4-positive cell infiltration was expressed as the percentage of PBMCs. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The mean fluorescence intensity of CCR4-positive cells was measured by flow cytometry. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The intensity of IL-23Rp19-positive cell infiltration was expressed as the percentage of PBMCs. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The mean fluorescence intensity of IL-23Rp19-positive cells was measured by flow cytometry. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The intensity of Treg cell infiltration was expressed as the percentage of PBMCs. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The intensity of Treg cell infiltration was expressed as the percentage of T cells. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The absolute number of Treg cells was expressed as cells per microliter (cells/mcL). Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The intensity of Th17 cell infiltration was expressed as the percentage of PBMCs. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The intensity of Th17 cell infiltration was expressed as the percentage of T cells. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The absolute number of Th17 cells was expressed as cells/mcL. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The intensity of CD19-positive cell infiltration was expressed as the percentage of PBMCs. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The mean fluorescence intensity of CD19-positive cells was measured by flow cytometry. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The intensity of CD24-positive cell infiltration was expressed as the percentage of PBMCs. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The mean fluorescence intensity of CD24-positive cells was measured by flow cytometry. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The intensity of CD27-positive cell infiltration was expressed as the percentage of PBMCs. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The mean fluorescence intensity of CD27-positive cells was measured by flow cytometry. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The intensity of CD38-positive cell infiltration was expressed as the percentage of PBMCs. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The mean fluorescence intensity of CD38-positive cells was measured by flow cytometry. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The intensity of IgM-positive cell infiltration was expressed as the percentage of PBMCs. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The mean fluorescence intensity of IgM-positive cells was measured by flow cytometry. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The intensity of mature B cell infiltration was expressed as the percentage of PBMCs. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The intensity of mature B cell infiltration was expressed as the percentage of B cells. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The absolute number of mature B cells was expressed as cells/mcL. Baseline AV and change from Baseline to Week 4 were averaged among all participants.

The intensity of memory B cell infiltration was expressed as the percentage of PBMCs. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The intensity of memory B cell infiltration was expressed as the percentage of B cells. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The absolute number of memory B cells was expressed as cells/mcL. Baseline AV and change from Baseline to Week 4 were averaged among all participants.

The intensity of transitional B cell infiltration was expressed as the percentage of PBMCs. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The intensity of transitional B cell infiltration was expressed as the percentage of B cells. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The absolute number of transitional B cells was expressed as cells/mcL. Baseline AV and change from Baseline to Week 4 were averaged among all participants.

The intensity of plasma B cell infiltration was expressed as the percentage of PBMCs. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The intensity of plasma B cell infiltration was expressed as the percentage of B cells. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

The absolute number of plasma B cells was expressed as cells/mcL. Baseline AV and change from Baseline to Week 4 were averaged among all participants.

Level of Th17CCL20 was measured in pg/mL. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

Level of Th17CCL20 was measured in pg/mL. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

Level of BCA was measured in pg/mL. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

Level of SDF1 was measured in pg/mL. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

Level of BAFF was measured in pg/mL. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

Level of APRIL was measured in pg/mL. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

Level of TNF-α was measured in pg/mL. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

Level of IL-1β was measured in pg/mL. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

Level of IL-17 was measured in pg/mL. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

Level of MCaP-1 was measured in pg/mL. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

Level of osteocalcin was measured in pg/mL. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

Level of Type I collagen N-propeptide was measured in pg/mL. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

Level of CTX-1 was measured in pg/mL. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

Level of ICTP was measured in pg/mL. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

Level of PIIANP was measured in pg/mL. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

Level of HELIX-II was measured in pg/mL. Baseline AV and changes from Baseline to Weeks 2 and 4 were averaged among all participants.

Level of Hb was measured in grams per liter (g/L). Baseline AV and changes from Baseline to Weeks 2, 4, 12, 24, and 48 were averaged among all participants.

Level of STR was measured in pg/mL. Baseline AV and changes from Baseline to Day 2 and Weeks 2 and 4 were averaged among all participants.

IRE was approximated by parametric mapping via DYNAMIKA software and expressed as change in relative signal intensity per second (ΔI/sec). The mean of three different slices was used in the determination of IRE. Each slice consisted of a two-dimensional (2D) sequence of images acquired from the same physical location at different time instances. Baseline AV and change from Baseline to Week 48 were averaged among all participants.

ME was approximated by parametric mapping via DYNAMIKA software and expressed as ratio of signal enhancement before and after contrast injection. The mean of three different slices was used in the determination of ME. Each slice consisted of a 2D sequence of images acquired from the same physical location at different time instances. Baseline AV and change from Baseline to Week 48 were averaged among all participants.

Ntotal was approximated using DYNAMIKA software. The sum of three different slices was used in the determination of Ntotal. Each slice consisted of a 2D sequence of images acquired from the same physical location at different time instances. Baseline AV and change from Baseline to Week 48 were averaged among all participants.

Npersistent was approximated using DYNAMIKA software. The sum of three different slices was used in the determination of Npersistent. Each slice consisted of a 2D sequence of images acquired from the same physical location at different time instances. Baseline AV and change from Baseline to Week 48 were averaged among all participants.

Nplateau was approximated using DYNAMIKA software. The sum of three different slices was used in the determination of Nplateau. Each slice consisted of a 2D sequence of images acquired from the same physical location at different time instances. Baseline AV and change from Baseline to Week 48 were averaged among all participants.

Nwashout was approximated using DYNAMIKA software. The sum of three different slices was used in the determination of Nwashout. Each slice consisted of a 2D sequence of images acquired from the same physical location at different time instances. Baseline AV and change from Baseline to Week 48 were averaged among all participants.

Ntotal and IRE were approximated using DYNAMIKA software. The sum of three different slices was used in the determination of Ntotal. The mean of three different slices was used in the determination of IRE. Each slice consisted of a 2D sequence of images acquired from the same physical location at different time instances. Function of Ntotal×IRE was expressed as voxels times change in relative intensity per second (v*ΔI/sec). Baseline AV and change from Baseline to Week 48 were averaged among all participants.

Ntotal and ME were approximated using DYNAMIKA software. The sum of three different slices was used in the determination of Ntotal. The mean of three different slices was used in the determination of ME. Each slice consisted of a 2D sequence of images acquired from the same physical location at different time instances. Function of NtotalME was expressed as voxels times ratio of signal intensity before and after contrast injection (v*ratio). Baseline AV and change from Baseline to Week 48 were averaged among all participants.