A Study of RoActemra/Actemra and, if Initially Inadequately Responded to RoActemra/Actemra, Followed by MabThera/Rituxan in Patients With Rheumatoid Arthritis

The DAS28 was calculated as [0.28 x the square root of number of swollen joints] + [0.56 x the square root of number of tender joints] + [0.7 x the natural log of ESR] + [0.014 x VAS patient global assessment of disease activity]. VAS assessments involved a 10-cm horizontal scale from 'no disease activity' to 'maximum disease activity.' DAS28 scores ranged from 0 to 10, with higher scores indicating increased disease activity. Remission was defined as a DAS28 score <2.6 at the assessment visit.

The DAS28 was calculated as [0.28 x the square root of number of swollen joints] + [0.56 x the square root of number of tender joints] + [0.7 x the natural log of ESR] + [0.014 x VAS patient global assessment of disease activity]. VAS assessments involved a 10-cm horizontal scale from 'no disease activity' to 'maximum disease activity.' DAS28 scores ranged from 0 to 10, with higher scores indicating increased disease activity. Remission was defined as a DAS28 score <2.6 at the assessment visit.

The DAS28 was calculated as [0.28 x the square root of number of swollen joints] + [0.56 x the square root of number of tender joints] + [0.7 x the natural log of ESR] + [0.014 x VAS patient global assessment of disease activity]. VAS assessments involved a 10-cm horizontal scale from 'no disease activity' to 'maximum disease activity.' DAS28 scores ranged from 0 to 10, with higher scores indicating increased disease activity. Remission was defined as a DAS28 score <2.6 at the assessment visit.

The DAS28 was calculated as [0.28 x the square root of number of swollen joints] + [0.56 x the square root of number of tender joints] + [0.7 x the natural log of ESR] + [0.014 x VAS patient global assessment of disease activity]. VAS assessments involved a 10-cm horizontal scale from 'no disease activity' to 'maximum disease activity.' DAS28 scores ranged from 0 to 10, with higher scores indicating increased disease activity. Remission was defined as a DAS28 score <2.6 at the assessment visit.

The DAS28 was calculated as [0.28 x the square root of number of swollen joints] + [0.56 x the square root of number of tender joints] + [0.7 x the natural log of ESR] + [0.014 x the patient global assessment of disease activity using a VAS]. VAS assessments involved a 10-cm horizontal scale from 'no disease activity' to 'maximum disease activity.' DAS28 scores ranged from 0 to 10, with higher scores indicating increased disease activity. LDAS was defined as a DAS28 score <3.2 at the assessment visit.

The DAS28 was calculated as [0.28 x the square root of number of swollen joints] + [0.56 x the square root of number of tender joints] + [0.7 x the natural log of ESR] + [0.014 x the patient global assessment of disease activity using a VAS]. VAS assessments involved a 10-cm horizontal scale from 'no disease activity' to 'maximum disease activity.' DAS28 scores ranged from 0 to 10, with higher scores indicating increased disease activity. LDAS was defined as a DAS28 score <3.2 at the assessment visit.

The DAS28 was calculated as [0.28 x the square root of number of swollen joints] + [0.56 x the square root of number of tender joints] + [0.7 x the natural log of ESR] + [0.014 x VAS patient global assessment of disease activity]. VAS assessments involved a 10-cm horizontal scale from 'no disease activity' to 'maximum disease activity.' DAS28 scores ranged from 0 to 10, with higher scores indicating increased disease activity. Reductions >1.2 points from Baseline to the assessment visit were considered clinically relevant.

The DAS28 was calculated as [0.28 x the square root of number of swollen joints] + [0.56 x the square root of number of tender joints] + [0.7 x the natural log of ESR] + [0.014 x VAS patient global assessment of disease activity]. VAS assessments involved a 10-cm horizontal scale from 'no disease activity' to 'maximum disease activity.' DAS28 scores ranged from 0 to 10, with higher scores indicating increased disease activity. Reductions >1.2 points from Baseline to the assessment visit were considered clinically relevant.

The DAS28 was calculated as [0.28 x the square root of number of swollen joints] + [0.56 x the square root of number of tender joints] + [0.7 x the natural log of ESR] + [0.014 x VAS patient global assessment of disease activity]. VAS assessments involved a 10-cm horizontal scale from 'no disease activity' to 'maximum disease activity.' DAS28 scores ranged from 0 to 10, with higher scores indicating increased disease activity. Reductions >1.2 points from the reference visit (Week 16) to the assessment visit were considered clinically relevant.

The DAS28 was calculated as [0.28 x the square root of number of swollen joints] + [0.56 x the square root of number of tender joints] + [0.7 x the natural log of ESR] + [0.014 x VAS patient global assessment of disease activity]. VAS assessments involved a 10-cm horizontal scale from 'no disease activity' to 'maximum disease activity.' DAS28 scores ranged from 0 to 10, with higher scores indicating increased disease activity.

The DAS28 was calculated as [0.28 x the square root of number of swollen joints] + [0.56 x the square root of number of tender joints] + [0.7 x the natural log of ESR] + [0.014 x VAS patient global assessment of disease activity]. VAS assessments involved a 10-cm horizontal scale from 'no disease activity' to 'maximum disease activity.' DAS28 scores ranged from 0 to 10, with higher scores indicating increased disease activity.

The DAS28 was calculated as [0.28 x the square root of number of swollen joints] + [0.56 x the square root of number of tender joints] + [0.7 x the natural log of ESR] + [0.014 x VAS patient global assessment of disease activity]. VAS assessments involved a 10-cm horizontal scale from 'no disease activity' to 'maximum disease activity.' DAS28 scores ranged from 0 to 10, with higher scores indicating increased disease activity.

The DAS28 was calculated as [0.28 x the square root of number of swollen joints] + [0.56 x the square root of number of tender joints] + [0.7 x the natural log of ESR] + [0.014 x VAS patient global assessment of disease activity]. VAS assessments involved a 10-cm horizontal scale from 'no disease activity' to 'maximum disease activity.' DAS28 scores ranged from 0 to 10, with higher scores indicating increased disease activity.

Response was determined using EULAR criteria based upon DAS28 absolute scores at the assessment visit and the DAS28 reduction from Baseline. Participants with a score ≤3.2 and reduction of >1.2 points were assessed as having a 'good' response. Participants with a score >3.2 with reduction of >1.2 points, or a score ≤5.1 with reduction of >0.6 to ≤1.2 points, were assessed as having a 'moderate' response. Participants with a score >5.1 with reduction of >0.6 to ≤1.2 points, or any score with reduction ≤0.6 points, were assessed as nonresponders with response recorded as 'none.'

Response was determined using EULAR criteria based upon DAS28 absolute scores at the assessment visit and the DAS28 reduction from the reference visit (Week 16). Participants with a score ≤3.2 and reduction of >1.2 points were assessed as having a 'good' response. Participants with a score >3.2 with reduction of >1.2 points, or a score ≤5.1 with reduction of >0.6 to ≤1.2 points, were assessed as having a 'moderate' response. Participants with a score >5.1 with reduction of >0.6 to ≤1.2 points, or any score with reduction ≤0.6 points, were assessed as nonresponders with response recorded as 'none.'

Response was determined using EULAR criteria based upon DAS28 absolute scores at the assessment visit and the DAS28 reduction from Baseline. Participants with a score ≤3.2 and reduction of >1.2 points were assessed as having a 'good' response. Participants with a score >3.2 with reduction of >1.2 points, or a score ≤5.1 with reduction of >0.6 to ≤1.2 points, were assessed as having a 'moderate' response. Participants with a score >5.1 with reduction of >0.6 to ≤1.2 points, or any score with reduction ≤0.6 points, were assessed as nonresponders with response recorded as 'none.'

Response was determined using ACR criteria based upon assessment of 66 joints for swelling and 68 joints for tenderness; joints were classified dichotomously as swollen or not swollen and tender or not tender. Respectively, these assessments were used to generate a swollen joint count (SJC) ranging from 0 to 66 swollen joints and a tender joint count (TJC) ranging from 0 to 68 tender joints. Response was defined as a reduction from Baseline of at least 20% for one of the following: VAS scores for patient-reported pain, patient global assessment of disease activity, or physician global assessment of disease activity, Health Assessment Questionnaire Disability Index (HAQ-DI), or C-reactive protein (CRP); plus a reduction in individual SJC and TJC of 20% (ACR20), 50% (ACR50), or 70% (ACR70). VAS assessments involved a 10-cm horizontal scale from 'no disease activity' to 'maximum disease activity.'

Response was determined using ACR criteria based upon assessment of 66 joints for swelling and 68 joints for tenderness; joints were classified dichotomously as swollen or not swollen and tender or not tender. Respectively, these assessments were used to generate an SJC ranging from 0 to 66 swollen joints and a TJC ranging from 0 to 68 tender joints. Response was defined as a reduction from the reference visit (Week 16) of at least 20% for one of the following: VAS scores for patient-reported pain, patient global assessment of disease activity, or physician global assessment of disease activity, HAQ-DI, or CRP; plus a reduction in individual SJC and TJC of 20% (ACR20), 50% (ACR50), or 70% (ACR70). VAS assessments involved a 10-cm horizontal scale from 'no disease activity' to 'maximum disease activity.'

Response was determined using ACR criteria based upon assessment of 66 joints for swelling and 68 joints for tenderness; joints were classified dichotomously as swollen or not swollen and tender or not tender. Respectively, these assessments were used to generate an SJC ranging from 0 to 66 swollen joints and a TJC ranging from 0 to 68 tender joints. Response was defined as a reduction from Baseline of at least 20% for one of the following: VAS scores for patient-reported pain, patient global assessment of disease activity, or physician global assessment of disease activity, HAQ-DI, or CRP; plus a reduction in individual SJC and TJC of 20% (ACR20), 50% (ACR50), or 70% (ACR70). VAS assessments involved a 10-cm horizontal scale from 'no disease activity' to 'maximum disease activity.'

The CDAI was calculated as [SJC + TJC + VAS patient global assessment of disease activity + VAS physician global assessment of disease activity]. VAS assessments involved a 10-cm horizontal scale from 'no disease activity' to 'maximum disease activity.' CDAI scores ranged from 0 to 76, with higher scores indicating increased disease activity. The SDAI was determined by adding CRP level to the CDAI score. Scores ranged from 0 to 86, with higher scores also indicating increased disease activity. A reduction in either score at the assessment visit reflects improvement in disease.

The CDAI was calculated as [SJC + TJC + VAS patient global assessment of disease activity + VAS physician global assessment of disease activity]. VAS assessments involved a 10-cm horizontal scale from 'no disease activity' to 'maximum disease activity.' CDAI scores ranged from 0 to 76, with higher scores indicating increased disease activity. The SDAI was determined by adding CRP level to the CDAI score. Scores ranged from 0 to 86, with higher scores also indicating increased disease activity. A reduction in either score at the assessment visit reflects improvement in disease.

The CDAI was calculated as [SJC + TJC + VAS patient global assessment of disease activity + VAS physician global assessment of disease activity]. VAS assessments involved a 10-cm horizontal scale from 'no disease activity' to 'maximum disease activity.' CDAI scores ranged from 0 to 76, with higher scores indicating increased disease activity. The SDAI was determined by adding CRP level to the CDAI score. Scores ranged from 0 to 86, with higher scores also indicating increased disease activity. A reduction in either score at the assessment visit reflects improvement in disease.

Blood samples for laboratory assessments, including hemoglobin level, were collected prior to each dose of study medication. The mean hemoglobin level was determined at Baseline and for assessment visits by averaging the observed hemoglobin level among all participants providing evaluable blood samples. Change from Baseline was calculated as [mean hemoglobin at the assessment visit minus mean hemoglobin at Baseline] and expressed in grams per liter (g/L).

Blood samples for laboratory assessments, including CRP level, were collected prior to each dose of study medication. The mean CRP level was determined at Baseline and for assessment visits by averaging the observed CRP level among all participants providing evaluable blood samples. Change from Baseline was calculated as [mean CRP at the assessment visit minus mean CRP at Baseline] and expressed in milligrams per deciliter (mg/dL).

Blood samples for laboratory assessments, including ESR, were collected prior to each dose of study medication. The mean ESR was determined at Baseline and for assessment visits by averaging the observed ESR among all participants providing evaluable blood samples. Change from Baseline was calculated as [mean ESR at the assessment visit minus mean ESR at Baseline] and expressed in millimeters per hour (mm/h).

Blood samples for laboratory assessments, including hemoglobin level, were collected prior to each dose of study medication. The mean hemoglobin level was determined at Baseline and for assessment visits by averaging the observed hemoglobin level among all participants providing evaluable blood samples. Change was calculated as [mean hemoglobin at Week 32 minus mean hemoglobin at Week 16] and expressed in g/L.

Blood samples for laboratory assessments, including CRP level, were collected prior to each dose of study medication. The mean CRP level was determined at Baseline and for assessment visits by averaging the observed CRP level among all participants providing evaluable blood samples. Change was calculated as [mean CRP at Week 32 minus mean CRP at Week 16] and expressed in mg/dL.

Blood samples for laboratory assessments, including ESR, were collected prior to each dose of study medication. The mean ESR was determined at Baseline and for assessment visits by averaging the observed ESR among all participants providing evaluable blood samples. Change was calculated as [mean ESR at Week 32 minus mean ESR at Week 16] and expressed in mm/h.

Blood samples for laboratory assessments, including hemoglobin level, were collected prior to each dose of study medication. The mean hemoglobin level was determined at Baseline and for assessment visits by averaging the observed hemoglobin level among all participants providing evaluable blood samples. Change from Baseline was calculated as [mean hemoglobin at the assessment visit minus mean hemoglobin at Baseline] and expressed in g/L.

Blood samples for laboratory assessments, including CRP level, were collected prior to each dose of study medication. The mean CRP level was determined at Baseline and for assessment visits by averaging the observed CRP level among all participants providing evaluable blood samples. Change from Baseline was calculated as [mean CRP at the assessment visit minus mean CRP at Baseline] and expressed in mg/dL.

Blood samples for laboratory assessments, including ESR, were collected prior to each dose of study medication. The mean ESR was determined at Baseline and for assessment visits by averaging the observed ESR among all participants providing evaluable blood samples. Change from Baseline was calculated as [mean ESR at the assessment visit minus mean ESR at Baseline] and expressed in mm/h.

Study discontinuation was documented by reason for each participant prematurely withdrawing from the study. The percentage of participants was calculated as the number withdrawing for insufficient therapeutic response divided by the total number of participants who began treatment.

Blood samples were collected to analyze total B-cell panel via immunophenotyping. Subpopulations were as follows: transitional (cluster of differentiation [CD] 19-positive, immunoglobulin (Ig) D-positive, CD38 medium, CD10-positive); naive (CD19-positive, IgD-positive, CD38 medium, CD27-negative); pre-switch memory (CD19-positive, CD27-positive, IgD-positive); post-switch memory (CD19-positive, CD27-positive, IgD-negative); IgG-positive class-switched (CD19-positive, IgG-positive); IgA-positive class-switched (CD19-positive, IgA-positive); double-negative memory (CD19-positive, IgD-negative, CD27-negative); and plasmablasts (CD19-positive, IgD-negative, CD38 high, CD27 high). Naive B-cell compartment was defined as the sum of transitional and naive B-cells. The sum of memory B-cell subsets with or without double-negative B-cells was also determined.

Blood samples were collected to analyze total B-cell panel via immunophenotyping. Subpopulations were as follows: transitional (cluster of differentiation [CD] 19-positive, immunoglobulin (Ig) D-positive, CD38 medium, CD10-positive); naive (CD19-positive, IgD-positive, CD38 medium, CD27-negative); pre-switch memory (CD19-positive, CD27-positive, IgD-positive); post-switch memory (CD19-positive, CD27-positive, IgD-negative); IgG-positive class-switched (CD19-positive, IgG-positive); IgA-positive class-switched (CD19-positive, IgA-positive); double-negative memory (CD19-positive, IgD-negative, CD27-negative); and plasmablasts (CD19-positive, IgD-negative, CD38 high, CD27 high). Naive B-cell compartment was defined as the sum of transitional and naive B-cells. The sum of memory B-cell subsets with or without double-negative B-cells was also determined.

Blood samples were collected to analyze total B-cell panel via immunophenotyping. Subpopulations were as follows: transitional (cluster of differentiation [CD] 19-positive, immunoglobulin (Ig) D-positive, CD38 medium, CD10-positive); naive (CD19-positive, IgD-positive, CD38 medium, CD27-negative); pre-switch memory (CD19-positive, CD27-positive, IgD-positive); post-switch memory (CD19-positive, CD27-positive, IgD-negative); IgG-positive class-switched (CD19-positive, IgG-positive); IgA-positive class-switched (CD19-positive, IgA-positive); double-negative memory (CD19-positive, IgD-negative, CD27-negative); and plasmablasts (CD19-positive, IgD-negative, CD38 high, CD27 high). Naive B-cell compartment was defined as the sum of transitional and naive B-cells. The sum of memory B-cell subsets with or without double-negative B-cells was also determined.

Blood samples were collected to analyze total B-cell panel via immunophenotyping. Subpopulations were as follows: transitional, naïve, pre-switch memory, post-switch memory, IgG-positive class-switched, IgA-positive class-switched, double-negative memory, and plasmablasts. Naive B-cell compartment was defined as the sum of transitional and naive B-cells. The sum of memory B-cell subsets with or without double-negative B-cells was also determined. Extent of disease response, using change from Baseline to Week 16 in DAS28 score, was correlated to the percentage of B-cells within each subpopulation at Baseline. Correlation is indicated by a correlation coefficient (r) >0.2, with greater values indicating a stronger correlation.

Blood samples were collected to analyze total B-cell panel via immunophenotyping. Subpopulations were as follows: transitional, naïve, pre-switch memory, post-switch memory, IgG-positive class-switched, IgA-positive class-switched, double-negative memory, and plasmablasts. Naive B-cell compartment was defined as the sum of transitional and naive B-cells. The sum of memory B-cell subsets with or without double-negative B-cells was also determined. Extent of disease response, using change from Baseline in DAS28 score, was correlated to the percentage of B-cells within each subpopulation at Baseline. Correlation is indicated by a correlation coefficient (r) >0.2, with greater values indicating a stronger correlation.

Blood samples were collected to analyze total B-cell panel via immunophenotyping. Subpopulations were as follows: transitional, naïve, pre-switch memory, post-switch memory, IgG-positive class-switched, IgA-positive class-switched, double-negative memory, and plasmablasts. Naive B-cell compartment was defined as the sum of transitional and naive B-cells. The sum of memory B-cell subsets with or without double-negative B-cells was also determined. Extent of disease response, using change from Baseline in DAS28 score, was correlated to the percentage of B-cells within each subpopulation at Baseline. Correlation is indicated by a correlation coefficient (r) >0.2, with greater values indicating a stronger correlation.

Work days missed were documented by reason (either rheumatoid arthritis [RA] or other reasons) for each participant over the preceding 7-day period. The mean number of work days missed was calculated by averaging the number of days missed per week among all participants.

The SF-36 evaluates participant-rated quality of life using 8 domains: physical and social functioning, physical and emotional role limitations, bodily pain, general health, vitality, and mental health. The score for each section is the average of the individual question scores, which are scaled from 0 to 100, with higher scores indicating better functioning. The mean score at each timepoint was determined by averaging the scores among all participants.

The SF-36 evaluates participant-rated quality of life using 8 domains: physical and social functioning, physical and emotional role limitations, bodily pain, general health, vitality, and mental health. The score for each section is the average of the individual question scores, which are scaled from 0 to 100, with higher scores indicating better functioning. The mean score at each timepoint was determined by averaging the scores among all participants, and the change in each domain score was calculated as [mean score at Week 16 minus mean score at Baseline].

The SF-36 evaluates participant-rated quality of life using 8 domains: physical and social functioning, physical and emotional role limitations, bodily pain, general health, vitality, and mental health. The score for each section is the average of the individual question scores, which are scaled from 0 to 100, with higher scores indicating better functioning. The mean score at each timepoint was determined by averaging the scores among all participants, and the change in each domain score was calculated as [mean score at Week 32 minus mean score at Week 16].

The SF-36 evaluates participant-rated quality of life using 8 domains: physical and social functioning, physical and emotional role limitations, bodily pain, general health, vitality, and mental health. The score for each section is the average of the individual question scores, which are scaled from 0 to 100, with higher scores indicating better functioning. The mean score at each timepoint was determined by averaging the scores among all participants, and the change in each domain score was calculated as [mean score at Week 32 minus mean score at Week 16].

The HAQ-DI evaluates participant-reported quality of life using 8 categories: dressing/grooming, arising, eating, walking, hygiene, reach, grip, and other common activities such as running errands and performing household chores. Each category contains multiple questions, which are answered using a 4-point scale from 0 to 3. The overall index score is taken as an average of the individual item responses and may range from 0 to 3, where higher scores indicate more difficulty in daily living activities. The mean index score at each timepoint was determined by averaging the scores among all participants.

The HAQ-DI evaluates participant-reported quality of life using 8 categories: dressing/grooming, arising, eating, walking, hygiene, reach, grip, and other common activities such as running errands and performing household chores. Each category contains multiple questions, which are answered using a 4-point scale from 0 to 3. The overall index score is taken as an average of the individual item responses and may range from 0 to 3, where higher scores indicate more difficulty in daily living activities. The mean index score at each timepoint was determined by averaging the scores among all participants, and the change in score was calculated as [mean score at Week 16 minus mean score at Baseline].

The HAQ-DI evaluates participant-reported quality of life using 8 categories: dressing/grooming, arising, eating, walking, hygiene, reach, grip, and other common activities such as running errands and performing household chores. Each category contains multiple questions, which are answered using a 4-point scale from 0 to 3. The overall index score is taken as an average of the individual item responses and may range from 0 to 3, where higher scores indicate more difficulty in daily living activities. The mean index score at each timepoint was determined by averaging the scores among all participants, and the change in score was calculated as [mean score at Week 32 minus the mean score at Week 16].

The HAQ-DI evaluates participant-reported quality of life using 8 categories: dressing/grooming, arising, eating, walking, hygiene, reach, grip, and other common activities such as running errands and performing household chores. Each category contains multiple questions, which are answered using a 4-point scale from 0 to 3. The overall index score is taken as an average of the individual item responses and may range from 0 to 3, where higher scores indicate more difficulty in daily living activities. The mean index score at each timepoint was determined by averaging the scores among all participants, and the change in score was calculated as [mean score at Week 32 minus the mean score at Week 16].

The HAQ-DI evaluates participant-reported quality of life using 8 categories: dressing/grooming, arising, eating, walking, hygiene, reach, grip, and other common activities such as running errands and performing household chores. Each category contains multiple questions, which are answered using a 4-point scale from 0 to 3. The overall index score is taken as an average of the individual item responses and may range from 0 to 3, where higher scores indicate more difficulty in daily living activities. Response was defined as a change in index score >0.22 from Baseline to Week 16.

The FACIT-F evaluates quality of life using 5 categories: physical well-being (PWB), social/family well-being (SWB), emotional well-being (EWB), functional well-being (FWB), and fatigue (FS). Participants answer each item on a 5-point scale from 0 to 4. The total score is the sum of individual responses across all 5 categories and may range from 0 to 160. The FACIT-General (FACIT-G; range 0 to 108) is the sum of scores for PWB, SWB, EWB, and FWB; the FACIT-Fatigue (FACIT-F) trial outcome index (TOI; range 0 to 108) is the sum of scores for PWB, FWB, and FS; and the FACIT-F fatigue (range 0 to 52) is the sum of scores for the FS only. For derivations of the FACIT-F reported here, higher scores indicate better quality of life. The mean score at each timepoint was determined by averaging scores among all participants.

The FACIT-F evaluates quality of life using 5 categories: PWB, SWB, EWB, FWB, and FS. Participants answer each item on a 5-point scale from 0 to 4. The total score is the sum of individual responses across all 5 categories and may range from 0 to 160. The FACIT-G (range 0 to 108) is the sum of scores for PWB, SWB, EWB, and FWB; the FACIT-F TOI (range 0 to 108) is the sum of scores for PWB, FWB, and FS; and the FACIT-F fatigue (range 0 to 52) is the sum of scores for the FS only. For derivations of the FACIT-F reported here, higher scores indicate better quality of life. The mean score at each timepoint was determined by averaging scores among all participants, and the change in score was calculated as [mean score at Week 16 minus mean score at Baseline].

The FACIT-F evaluates quality of life using 5 categories: PWB, SWB, EWB, FWB, and FS. Participants answer each item on a 5-point scale from 0 to 4. The total score is the sum of individual responses across all 5 categories and may range from 0 to 160. The FACIT-G (range 0 to 108) is the sum of scores for PWB, SWB, EWB, and FWB; the FACIT-F TOI (range 0 to 108) is the sum of scores for PWB, FWB, and FS; and the FACIT-F fatigue (range 0 to 52) is the sum of scores for the FS only. For derivations of the FACIT-F reported here, higher scores indicate better quality of life. The mean score at each timepoint was determined by averaging scores among all participants, and the change in score was calculated as [mean score at Week 32 minus mean score at Week 16].

The FACIT-F evaluates quality of life using 5 categories: PWB, SWB, EWB, FWB, and FS. Participants answer each item on a 5-point scale from 0 to 4. The total score is the sum of individual responses across all 5 categories and may range from 0 to 160. The FACIT-G (range 0 to 108) is the sum of scores for PWB, SWB, EWB, and FWB; the FACIT-F TOI (range 0 to 108) is the sum of scores for PWB, FWB, and FS; and the FACIT-F fatigue (range 0 to 52) is the sum of scores for the FS only. For derivations of the FACIT-F reported here, higher scores indicate better quality of life. The mean score at each timepoint was determined by averaging scores among all participants, and the change in score was calculated as [mean score at Week 32 minus mean score at Week 16].