This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
アプレミラスト 2.7.3 臨床的有効性(乾癬性関節炎) セルジーン株式会社
Confidential and Proprietary 71
6.7. ACR コンポーネント
Figure 26: Median Percent Change From Baseline in ACR Component Scores at Week 52 in Subjects Initially Randomized to Placebo (Study PSA-002; AAR Population; Data as Observed)
AAR = apremilast subjects as initially randomized/re-randomized; ACR = American College of Rheumatology; BID = twice daily; CRP = C-reactive protein; EE = early escape (re-randomized to apremilast at Week 16); EGA = evaluator’s (physician’s) global assessment of disease activity; HAQ-DI = Health Assessment Questionnaire – Disability Index; PBO = placebo; PGA = patient’s (subject’s) global assessment of disease activity; SJC = swollen joint count; TJC = tender joint count; XO = cross-over (re-randomized to apremilast at Week 24). Source: PSA-002 治験総括報告書 Table 14.2.46, Table 14.2.47, Table 14.2.48.1, Table 14.2.49, Table 14.2.50, Table 14.2.51.1, Table 14.2.52.1.
-64.3
45
-59.1
20
-50.0
38
-78.1
23
-80.0
45
-85.4
20
-55.5
38
-77.8
23
-20.8
44
-36.4
19
-25.0
38
-30.8
23
-42.2
45
-8.7
20
-14.2
38
15.1
23
-27.7
45
-45.0
19
-31.5
37
-43.8
23
-28.6
45
-50.0
19
-16.7
37
-36.1
23
-74.6
43
-68.1
18
-38.5
37
-77.2
23N =
Treatment PBO/20 EE PBO/20 XO PBO/30 EE PBO/30 XO
Med
ian
Per
cent
Cha
nge
from
Bas
elin
e
-100
-90.0
-80.0
-70.0
-60.0
-50.0
-40.0
-30.0
-20.0
-10.0
0.0
10.0
20.0
TJC SJC HAQ-DI CRP Pain PGA EGA
アプレミラスト 2.7.3 臨床的有効性(乾癬性関節炎) セルジーン株式会社
Confidential and Proprietary 72
Figure 27: Median Percent Change From Baseline in ACR Component Scores at Week 52 in Subjects Initially Randomized to Placebo (Study PSA-003; AAR Population; Data as Observed)
AAR = apremilast subjects as initially randomized/re-randomized; ACR = American College of Rheumatology; BID = twice daily; CRP = C-reactive protein; EE = early escape (re-randomized to apremilast at Week 16); EGA = evaluator’s (physician’s) global assessment of disease activity; HAQ-DI = Health Assessment Questionnaire – Disability Index; PBO = placebo; PGA = patient’s (subject’s) global assessment of disease activity; SJC = swollen joint count; TJC = tender joint count; XO = cross-over (re-randomized to apremilast at Week 24). Source: PSA-003 治験総括報告書 Table 14.2.46, Table 14.2.47, Table 14.2.48.1, Table 14.2.49, Table 14.2.50, Table 14.2.51.1, Table 14.2.52.1.
Figure 28: Median Percent Change From Baseline in ACR Component Scores at Week 52 in Subjects Initially Randomized to Placebo (Study PSA-004; AAR Population; Data as Observed)
AAR = apremilast subjects as initially randomized/re-randomized; ACR = American College of Rheumatology; BID = twice daily; CRP = C-reactive protein; EE = early escape (re-randomized to apremilast at Week 16); EGA = evaluator’s (physician’s) global assessment of disease activity; HAQ-DI = Health Assessment Questionnaire – Disability Index; PBO = placebo; PGA = patient’s (subject’s) global assessment of disease activity; SJC = swollen joint count; TJC = tender joint count; XO = cross-over (re-randomized to apremilast at Week 24). Source: PSA-004 治験総括報告書 Table 14.2.46, Table 14.2.47, Table 14.2.48.1, Table 14.2.49, Table 14.2.50, Table 14.2.51.1, Table 14.2.52.1.
-61.5
33
-76.1
22
-66.7
44
-78.6
23
-63.6
33
-87.3
22
-71.0
44
-100
23
-30.0
31
-35.7
21
-33.3
42
-42.9
23
21.1
33
-5.0
21
-14.7
44
-29.7
23
-35.5
33
-56.2
22
-25.0
43
-69.2
23
-26.1
33
-50.7
22
-13.7
44
-43.1
23
-55.3
32
-74.2
21
-63.5
43
-62.7
23N =
Treatment PBO/20 EE PBO/20 XO PBO/30 EE PBO/30 XO
Med
ian
Per
cent
Cha
nge
from
Bas
elin
e
-100
-90.0
-80.0
-70.0
-60.0
-50.0
-40.0
-30.0
-20.0
-10.0
0.0
10.0
20.0
30.0
TJC SJC HAQ-DI CRP Pain PGA EGA
アプレミラスト 2.7.3 臨床的有効性(乾癬性関節炎) セルジーン株式会社
Confidential and Proprietary 80
Table 9: CDAI Score Categories at Baseline, Week 16, and Week 24 in Studies PSA-002, PSA-003, and PSA-004 (FAS)
Table 10: Summary of DAS28(CRP) Categories in All Subjects Exposed to Apremilast During the Apremilast-exposure Period (Weeks 0-52) (Study PSA-002; AAR Population; Data as Observed)
Table 10: Summary of DAS28(CRP) Categories in All Subjects Exposed to Apremilast During the Apremilast-exposure Period (Weeks 0-52) (Study PSA-002; AAR Population; Data as Observed)(つづき)
Number (%) of Subjectsa PBO/APR 20 BIDb PBO/APR 30 BIDb APR 20 BIDc APR 30 BIDc Visit
AAR = apremilast subjects as initially randomized/re-randomized; APR = apremilast; BID = twice daily; DAS28(CRP) = 28-joint Disease Activity Score using C-reactive protein as the acute-phase reactant; EE = early escape (for placebo subjects, re-randomized to apremilast at Week 16; for apremilast subjects, continue treatment as randomized); NEE = no early escape; PBO = placebo; XO = cross-over (re-randomized to apremilast at Week 24).
a Percentage calculated based on the number of subjects with a post-baseline value at each time point. b Subjects in the PBO/20 EE and PBO/30 EE treatment groups were re-randomized to receive apremilast at Week 16. Subjects in the PBO/20 XO and PBO/30 XO treatment
groups were re-randomized to receive apremilast at Week 24. c All subjects initially randomized to apremilast continued on the same treatment regimen irrespective of early escape. Source: PSA-002 治験総括報告書 Table 14.2.58.2.
アプレミラスト 2.7.3 臨床的有効性(乾癬性関節炎) セルジーン株式会社
Confidential and Proprietary 83
Table 11: Summary of DAS28(CRP) Categories in All Subjects Exposed to Apremilast During the Apremilast-exposure Period (Weeks 0-52) (Study PSA-003; AAR Population; Data as Observed)
Table 11: Summary of DAS28(CRP) Categories in All Subjects Exposed to Apremilast During the Apremilast-exposure Period (Weeks 0-52) (Study PSA-003; AAR Population; Data as Observed)(つづき)
AAR = apremilast subjects as initially randomized/re-randomized; APR = apremilast; BID = twice daily; DAS28(CRP) = 28-joint Disease Activity Score using C-reactive protein as the acute-phase reactant; EE = early escape (for placebo subjects, re-randomized to apremilast at Week 16; for apremilast subjects, continue treatment as randomized); NEE = no early escape; PBO = placebo; XO = cross-over (re-randomized to apremilast at Week 24).
a Percentage calculated based on the number of subjects with a post-baseline value at each time point. b Subjects in the PBO/20 EE and PBO/30 EE treatment groups were re-randomized to receive apremilast at Week 16. Subjects in the PBO/20 XO and PBO/30 XO treatment
groups were re-randomized to receive apremilast at Week 24. c All subjects initially randomized to apremilast continued on the same treatment regimen irrespective of early escape. Source: PSA-003 治験総括報告書 Table 14.2.58.2.
アプレミラスト 2.7.3 臨床的有効性(乾癬性関節炎) セルジーン株式会社
Confidential and Proprietary 85
Table 12: Summary of DAS28(CRP) Categories in All Subjects Exposed to Apremilast During the Apremilast-exposure Period (Weeks 0-52) (Study PSA-004; AAR Population; Data as Observed)
Table 12: Summary of DAS28(CRP) Categories in All Subjects Exposed to Apremilast During the Apremilast-exposure Period (Weeks 0-52) (Study PSA-004; AAR Population; Data as Observed)(つづき)
Number (%) of Subjectsa PBO/APR 20 BIDb PBO/APR 30 BIDb APR 20 BIDc APR 30 BIDc Visit
AAR = apremilast subjects as initially randomized/re-randomized; APR = apremilast; BID = twice daily; DAS28(CRP) = 28-joint Disease Activity Score using C-reactive protein as the acute-phase reactant; EE = early escape (for placebo subjects, re-randomized to apremilast at Week 16; for apremilast subjects, continue treatment as randomized); NEE = no early escape; PBO = placebo; XO = cross-over (re-randomized to apremilast at Week 24).
a Percentage calculated based on the number of subjects with a postbaseline value at each time point. b Subjects in the PBO/20 EE and PBO/30 EE treatment groups were re-randomized to receive apremilast at Week 16. Subjects in the PBO/20 XO and PBO/30 XO treatment
groups were re-randomized to receive apremilast at Week 24. c All subjects initially randomized to apremilast continued on the same treatment regimen irrespective of early escape. Source: PSA-004 治験総括報告書 Table 14.2.58.2.
アプレミラスト 2.7.3 臨床的有効性(乾癬性関節炎) セルジーン株式会社
Confidential and Proprietary 87
6.9. 併合解析
Table 13: Summary of Reduction of Signs and Symptoms of Active PsA Endpoints from the Pooled Analysis at Weeks 16 and 24 (FAS)
ACR 20/50/70 = American College of Rheumatology 20%/50%/70% response; ACR-N = American College of Rheumatology – N index; ANCOVA = analysis of covariance; APR = apremilast; BID = twice daily; BL = baseline; CMH = Cochran-Mantel-Haenszel; CRP = C-reactive protein; DMARD = disease-modifying antirheumatic drug; EGA = evaluator’s (physician’s) global assessment of disease activity; FAS = Full Analysis Set; HAQ-DI = Health Assessment Questionnaire Disability Index; LS = least squares; PGA = patient’s (subject’s) global assessment of disease activity; SE = standard error; SJC = swollen joint count; TJC = tender joint count; Trt. = treatment. a Change from Baseline is median percent change from baseline for TJC, SJC, subject’s assessment of pain, PGA, EGA, HAQ-DI (signs/symptoms), CRP, and LS mean change (SE) for HAQ-DI (physical function).
b Treatment effect for categorical endpoints (ACR 20/50/70) is based on adjusted difference in proportions for apremilast versus placebo (weighted average of the treatment differences across the strata of baseline disease-modifying antirheumatic drug (DMARD) use and study with the CMH weights). Treatment effect for ACR components (TJC, SJC, PGA, EGA, HAQ-DI [signs and symptoms], CRP, and ACR-N) is based on Hodges-Lehmann estimate of location shift. Treatment effect for other continuous endpoints (HAQ-DI [disease activity]) is the difference in LS means based on an analysis of covariance (ANCOVA) model for the change from baseline at the respective visit, with treatment group, baseline DMARD use, and study as factors, and baseline value as a covariate.
c For categorical endpoints, 2-sided P-value is based on Cochran-Mantel-Haenszel (CMH) test adjusting for baseline DMARD use and study. For ACR components (TJC, SJC, PGA, EGA, HAQ-DI [signs and symptoms], CRP, and ACR-N), P-value is based on an ANCOVA model for the van der Waerden normal scores of the percent change from baseline, with treatment group, baseline DMARD use, and study as factors, and the van der Waerden normal scores of the baseline value as a covariate. For other continuous endpoints, P-value is based on an ANCOVA model for the change from baseline at the respective visit, with treatment group, baseline DMARD use, and study as factors, and baseline value as a covariate. P-values in italics for the pooled analysis are ≤ 0.050 and considered nominally significant as there was no adjustment for multiplicity.
ANCOVA = analysis of covariance; APR = apremilast; BID = twice daily; BL = baseline; CMH = Cochran-Mantel-Haenszel; DMARD = disease-modifying antirheumatic drug; FACIT = Functional Assessment of Chronic Illness Therapy; FAS = Full Analysis Set; LOCF = last observation carried forward; LS = least squares; NRI = nonresponder imputation; PCS = physical component summary score; PF = physical functioning domain score; SE = standard error; SF-36v2 = 36-item Short Form Health Survey, version 2; Trt. = treatment. a Change from Baseline is LS mean change (SE).
b Treatment effect for categorical endpoints is based on adjusted difference in proportions for apremilast versus placebo (weighted average of the treatment differences across the strata of baseline disease-modifying antirheumatic drug (DMARD) use and study with the CMH weights). Treatment effect for continuous endpoints is the difference in LS means based on an analysis of covariance (ANCOVA) model for the change from baseline at Week 16, with treatment group, baseline DMARD use, and study as factors, and baseline value as a covariate.
c For categorical endpoints, 2-sided P-value is based on Cochran-Mantel-Haenszel (CMH) test adjusting for baseline DMARD use and study. For continuous endpoints, P-value is based on an ANCOVA model for the change from baseline at Week 16, with treatment group, baseline DMARD use, and study as factors, and baseline value as a covariate. P-values in italics for the pooled analysis are ≤ 0.050 and considered nominally significant as there was no adjustment for multiplicity.
ANCOVA = analysis of covariance; APR = apremilast; BID = twice daily; BL = baseline; CDAI = Clinical Disease Activity Index; CMH = Cochran-Mantel-Haenszel; CRP = C-reactive protein; DAS28(CRP) = 28-joint Disease Activity Score; DMARD = disease-modifying antirheumatic drug; EULAR = European League Against Rheumatism; FAS = Full Analysis Set; LS = least squares; MASES = Maastricht Ankylosing Spondylitis Enthesitis Score; NRI = nonresponder imputation; PASI = Psoriasis Area and Severity Index; PsARC = Psoriatic Arthritis Response Criteria; SE = standard error; Trt. = treatment.
a Change from Baseline is LS mean change (SE).
b Treatment effect for categorical endpoints is based on adjusted difference in proportions for apremilast versus placebo (weighted average of the treatment differences across the strata of baseline disease-modifying antirheumatic drug (DMARD) use and study with the CMH weights). Treatment effect for continuous endpoints is the difference in LS means based on an analysis of covariance (ANCOVA) model for the change from baseline at Week 16, with treatment group, baseline DMARD use, and study as factors, and baseline value as a covariate.
c For categorical endpoints, 2-sided P-value is based on Cochran-Mantel-Haenszel (CMH) test adjusting for baseline DMARD use and study. For continuous endpoints, P-value is based on an ANCOVA model for the change from baseline at Week 16, with treatment group, baseline DMARD use, and study as factors, and baseline value as a covariate. P-values in italics for the pooled analysis are ≤ 0.050 and considered nominally significant as there was no adjustment for multiplicity.
d PASI-75 responses were evaluated in subjects with Baseline psoriatic skin involvement ≥ 3% body surface area (N = 231 placebo, 248 APR 20 BID, and 249 APR 30 BID).
e Change from Baseline in MASES score evaluated in subjects with pre-existing enthesopathy (MASES > 0) at Baseline and at least 1 post-baseline value.
f Change from Baseline in dactylitis severity score evaluated in subjects with pre-existing dactylitis (dactylitis severity score > 0) at Baseline and at least 1 post-baseline value.
Note: For categorical analyses, subjects who did not have sufficient data (observed or imputed) for a definitive determination of response status at Week 16/24 are counted as non-responders. Source: 5.3.5.3.2 項 有効性併合解析(PsA)Table 2.4.1, Table 2.4.2, Table 2.7.1, Table 2.7.2, Table 2.7.3, Table 2.7.4, Table 2.7.5, Table 2.7.6, Table 2.8.1, Table 2.8.2, Table 2.8.3, Table 2.8.4, Table 2.8.5, Table 2.8.6, Table 2.9.1, Table 2.9.2, Table 2.10.1, Table 2.10.2, Table 2.11.1, Table 2.11.2, Table 2.13.3, Table 2.13.4, Table 2.14.1, Table 2.14.2.
アプレミラスト 2.7.3 臨床的有効性(乾癬性関節炎) セルジーン株式会社
Confidential and Proprietary 94
Figure 37: Proportion of All Subjects Exposed to Apremilast Achieving Modified ACR 70 Responses During the Apremilast-exposure Period up to Week 52 (AAR Population; Pooled Studies PSA-002, PSA-003 and PSA-004; Data as Observed)
AAR = apremilast subjects as initially randomized/re-randomized; ACR 50 = American College of Rheumatology 50% response criteria; APR = apremilast; BID = twice daily;
EE = early escape (re-randomized to apremilast at Week 16); n/m = number of responders/number of subjects with sufficient data for definitive determination of response status at each time point (including subjects who discontinued early between the preceding visit and the visit in question); PBO = placebo; XO = cross-over (re-randomized to apremilast at Week 24).
Note: To enhance visual clarity, the data points collected at Weeks 16, 24, 40, and 52 are displayed in a staggered fashion above the hash mark for the corresponding time point. Source: 5.3.5.3.2 項 有効性併合解析(PsA)Table 2.20.1.
アプレミラスト 2.7.3 臨床的有効性(乾癬性関節炎) セルジーン株式会社
Confidential and Proprietary 95
6.10. 部分集団解析
Figure 38: Modified ACR 20 Response in the APR 20 BID Treatment Group at Week 24 by Demographic Subpopulation (Pooled Analysis; FAS; NRI)
ACR 20 = 20% improvement in American College of Rheumatology response criteria; Act = active treatment; Adj = adjusted;
APR = apremilast; BID = twice daily; BMI = body mass index; CI = confidence interval; Diff = difference; FAS = full analysis set; NRI = non-responder imputation; vs. = versus.
*= unadjusted difference and CI. Note: Subjects who discontinued early prior to Week 24 and subjects who did not have sufficient data for a definitive
determination of response status at Week 24 are counted as non-responders. Source: 5.3.5.3.2 項 有効性併合解析(PsA)Table 2.1.2, Table 3.1.2.
アプレミラスト 2.7.3 臨床的有効性(乾癬性関節炎) セルジーン株式会社
Confidential and Proprietary 96
Figure 39: Modified ACR 20 Response in the APR 30 BID Treatment Group at Week 24 by Demographic Subpopulation (Pooled Analysis; FAS; NRI)
ACR 20 = 20% improvement in American College of Rheumatology response criteria; Act = active treatment; Adj = adjusted;
APR = apremilast; BID = twice daily; BMI = body mass index; CI = confidence interval; Diff = difference; FAS = full analysis set; NRI = non-responder imputation; vs. = versus.
*= unadjusted difference and CI. Note: Subjects who discontinued early prior to Week 24 and subjects who did not have sufficient data for a definitive
determination of response status at Week 24 are counted as non-responders. Source: 5.3.5.3.2 項 有効性併合解析(PsA)Table 2.1.2, Table 3.1.2.
アプレミラスト 2.7.3 臨床的有効性(乾癬性関節炎) セルジーン株式会社
Confidential and Proprietary 97
Figure 40: Modified ACR 20 Response at Week 24 by PsA Subtype, PsA Duration, and Baseline CRP (Pooled Analysis; FAS; NRI)
APR 20 BID
APR30 BID
ACR 20 = 20% improvement in American College of Rheumatology response criteria; Act = active treatment; Adj = adjusted;
*=unadjusted difference and CI. Note: Subjects who discontinued early prior to Week 24 and subjects who did not have sufficient data for a definitive
determination of response status at Week 24 are counted as non-responders. Source: 5.3.5.3.2 項 有効性併合解析(PsA)Table 2.1.2, Table 3.1.2.
アプレミラスト 2.7.3 臨床的有効性(乾癬性関節炎) セルジーン株式会社
Confidential and Proprietary 100
Figure 43: Modified ACR 20 Response at Week 24 by Prior Biologic DMARD Use (Pooled Analysis; FAS; NRI)
ACR 20 = 20% improvement in American College of Rheumatology response criteria; BID = twice daily; DMARD = disease-modifying antirheumatic drug; FAS = full analysis
set; NRI = non-responder imputation. Note: Error bars denote standard error. Note: Subjects who discontinued early prior to Week 24 and subjects who did not have sufficient data for a definitive determination of response status at Week 24 are counted as
Prior Biologic Use Prior Biologic Use (Excluding Biologic
Failures)
Biologic Failures
14.7
73/496
27.8
139/500
30.4
151/497
17.0
64/376
30.9
118/382
32.5
124/382
7.1
8/112
17.4
20/115
24.3
26/107
8.2
6/73
17.8
13/73
26.4
19/72
5.1
2/39
16.7
7/42
20.0
7/35n/N =
アプレミラスト 2.7.3 臨床的有効性(乾癬性関節炎) セルジーン株式会社
Confidential and Proprietary 101
Figure 44: Modified ACR 20 Response at Week 24 by Baseline Use of Small-Molecule DMARDs (Pooled Analysis; FAS; NRI)
ACR 20 = 20% improvement in American College of Rheumatology response criteria; BID = twice daily; DMARD = disease-modifying antirheumatic drug; FAS = full analysis
set; MTX = methotrexate; NRI = non-responder imputation. Note: Error bars denote standard error. Note: Subjects who discontinued early prior to Week 24 and subjects who did not have sufficient data for a definitive determination of response status at Week 24 are counted as
APR = apremilast; BID = twice daily; BMI = body mass index; FAS = full analysis set; min – max = minimum to maximum; SD = standard deviation. a Rest of world countries included Australia, New Zealand, Republic of Korea, Russian Federation, South Africa, and Taiwan, province of China. A summary of subjects by country can be found in Table 14.1.3 of PSA-002 CSR, PSA-003 CSR, and PSA-004 CSR.
APR = apremilast; BID = twice daily; DMARD = disease-modifying antirheumatic drug; FAS = full analysis set; min – max = minimum to maximum; MTX = methotrexate; PsA = psoriatic arthritis; SD = standard deviation. a Baseline DMARD use as recorded on case report form.
b Other DMARDs in Study PSA-002 included azathioprine, etanercept, hydroxychloroquine, and mesalazine. Other DMARDs in Study PSA-003 included chloroquine/hydroxychloroquine, tocanol, and ustekinumab. Other DMARDs in Study PSA-004 included azathioprine, etanercept, hydroxychloroquine, and mesalazine. Etanercept and ustekinumab are prior medications captured as baseline medications because the dates of last dose were incomplete and are therefore imputed as being used at baseline.
c Prednisone or equivalent (ie, normalized to prednisone).
Female n = 256 n = 267 n = 274 n = 797 Mean ± SD 1.324 ± 0.5890 1.263 ± 0.6219 1.337 ± 0.6258 1.308 ± 0.6130
Median (min – max) 1.375 (0.00 – 2.75)
1.250 (0.00 – 2.88)
1.375 (0.00 – 2.88)
1.375 (0.00 – 2.88)
CRP (mg/dL)a
Male n = 240 n = 232 n = 222 n = 694 Mean ± SD 1.096 ± 1.4200 1.015 ± 1.6240 1.118 ± 1.6697 1.076 ± 1.5702
Median (min – max) 0.538 (0.02 – 8.86)
0.411 (0.02 – 13.73)
0.537 (0.02 – 13.22)
0.491 (0.02 – 13.73)
Female n = 256 n = 268 n = 275 n = 799 Mean ± SD 0.968 ± 1.4485 0.957 ± 1.8261 0.830 ± 1.1159 0.917 ± 1.4892
Median (min – max) 0.425 (0.02 – 11.49)
0.459 (0.02 – 23.99)
0.355 (0.02 – 8.13)
0.422 (0.02 – 23.99)
ACR = American College of Rheumatology; APR = apremilast; BID = twice daily; HAQ-DI = Health Assessment Questionnaire Disability Index; CRP = C-reactive protein; FAS = full analysis set; SD = standard deviation; VAS = visual analog scale. a Normal range for CRP was < 0.5 mg/dL.
Note: Baseline data available for all subjects unless otherwise noted. Source: 5.3.5.3.2 項 有効性併合解析(PsA)Table 1.1.2a, Table 1.1.2.b.
アプレミラスト 2.7.3 臨床的有効性(乾癬性関節炎) セルジーン株式会社
Confidential and Proprietary 117
Table 20: Analysis by Sex of Modified ACR 20/50/70 Response, HAQ-DI Improvement, PsARC Response, SF-36v2 Physical Functioning Domain Score Improvement, and PASI-75 (Pooled Analysis; FAS)
Male Female Placebo
N = 240a
APR 20 BID N = 232a
APR 30 BID N = 222a
Placebo N = 256a
APR 20 BID N = 268a
APR 30 BID N = 275a
Endpoint Visit n/m (%)
b n/m (%)b
Treatment Difference
c n/m (%)b
Treatment Differencec n/m (%)b n/m (%)b
Treatment Differencec n/m (%)b
Treatment Differencec
Modified ACR 20
d
Week 16 39/240 (16.3)
86/232 (37.1)
21.0 (13.3, 28.7)
95/222 (42.8)
26.5 (18.6, 34.5)
54/256 (21.1)
74/268 (27.6)
6.6 (-0.7, 13.8)
89/275 (32.4)
11.6 (4.2, 19.0)
Week 24 33/240 (13.8)
76/232 (32.8)
19.3 (12.0, 26.6)
75/222 (33.8)
20.0 (12.5, 27.6)
40/256 (15.6)
63/268 (23.5)
8.0 (1.4, 14.7)
76/275 (27.6)
12.5 (5.7, 19.2)
Modified ACR 50d
Week 16 10/240 (4.2)
40/232 (17.2)
13.0 (7.6, 18.4)
38/222 (17.1)
12.8 (7.3, 18.3)
22/256 (8.6)
31/268 (11.6)
3.0 (-2.1, 8.1)
31/275 (11.3)
3.1 (-2.0, 8.1)
Week 24 15/240 (6.3)
41/232 (17.7)
11.4 (5.7, 17.1)
39/222 (17.6)
11.3 (5.5, 17.1)
19/256 (7.4)
29/268 (10.8)
3.4 (-1.5, 8.3)
39/275 (14.2)
7.1 (1.9, 12.2)
Modified ACR 70d
Week 16 3/240 (1.3)
15/232 (6.5)
5.3 (1.9, 8.8)
8/222 (3.6)
2.3 (-0.5, 5.1)
4/256 (1.6)
9/268 (3.4)
1.9 (-0.7, 4.5)
7/275 (2.5)
1.1 (-1.3, 3.5)
Week 24 5/240 (2.1)
17/232 (7.3)
5.3 (1.5, 9.0)
16/222 (7.2)
5.2 (1.3, 9.0)
7/256 (2.7)
8/268 (3.0)
0.3 (-2.6, 3.1)
14/275 (5.1)
2.5 (-0.8, 5.8)
HAQ-DIe
Week 16 LS Mean (SE)
-0.089 (0.0286)
-0.231 (0.0292)
-0.142 (-0.221, -0.062)
-0.273 (0.0299)
-0.184 (-0.265, -0.104)
-0.047 (0.0281)
-0.101 (0.0274)
-0.055 (-0.130, 0.021)
-0.158 (0.0272)
-0.112 (-0.187, -0.036)
Week 24 LS Mean (SE)
-0.105 (0.0297)
-0.260 (0.0303)
-0.154 (-0.236, -0.072)
-0.271 (0.0311)
-0.165 (-0.249, -0.082)
-0.035 (0.0292)
-0.092 (0.0284)
-0.056 (-0.135, 0.022)
-0.174 (0.0281)
-0.139 (-0.217, -0.061)
アプレミラスト 2.7.3 臨床的有効性(乾癬性関節炎) セルジーン株式会社
Confidential and Proprietary 118
Table 20: Analysis by Sex of Modified ACR 20/50/70 Response, HAQ-DI Improvement, PsARC Response, SF-36v2 Physical Functioning Domain Score Improvement, and PASI-75 (Pooled Analysis; FAS)(つづき)
Male Female Placebo
N = 240a
APR 20 BID N = 232
a
APR 30 BID N = 222
a
Placebo N = 256
a
APR 20 BID N = 268
a
APR 30 BID N = 275
a
Endpoint Visit n/m (%)
b n/m (%)b
Treatment Difference
c n/m (%)b
Treatment Differencec n/m (%)b n/m (%)b
Treatment Differencec n/m (%)b
Treatment Differencec
PsARCd Week 16 62/240
(25.8) 103/232 (44.4)
18.7 (10.5, 27.0)
114/222 (51.4)
25.6 (17.1, 34.1)
87/256 (34.0)
104/268 (38.8)
4.9 (-3.3, 13.0)
130/275 (47.3)
13.6 (5.4, 21.8)
Week 24 41/240 (17.1)
85/232 (36.6)
19.8 (12.2, 27.3)
93/222 (41.9)
24.7 (16.7, 32.7)
68/256 (26.6)
86/268 (32.1)
5.6 (-2.0, 13.3)
105/275 (38.2)
12.3 (4.6, 20.1)
SF-36v2 PFe Week 16 LS Mean (SE)
1.62 (0.508)
4.06 (0.520)
2.44 (1.02, 3.85)
4.42 (0.534)
2.80 (1.37, 4.24)
0.90 (0.503)
1.41 (0.488)
0.51 (-0.84, 1.87)
2.81 (0.484)
1.91 (0.56, 3.26)
Week 24 LS Mean (SE)
1.53 (0.523)
4.48 (0.535)
2.95 (1.49, 4.40)
4.41 (0.549)
2.88 (1.40, 4.36)
1.05 (0.518)
1.76 (0.503)
0.70 (-0.69, 2.09)
3.47 (0.496)
2.42 (1.03, 3.80)
アプレミラスト 2.7.3 臨床的有効性(乾癬性関節炎) セルジーン株式会社
Confidential and Proprietary 119
Table 20: Analysis by Sex of Modified ACR 20/50/70 Response, HAQ-DI Improvement, PsARC Response, SF-36v2 Physical Functioning Domain Score Improvement, and PASI-75 (Pooled Analysis; FAS)(つづき)
Male Female Placebo
N = 240a
APR 20 BID N = 232
a
APR 30 BID N = 222
a
Placebo N = 256
a
APR 20 BID N = 268
a
APR 30 BID N = 275
a
Endpoint Visit n/m (%)
b n/m (%)b
Treatment Difference
c n/m (%)b
Treatment Differencec n/m (%)b n/m (%)b
Treatment Differencec n/m (%)b
Treatment Differencec
PASI-75f
Week 16 3/123 (2.4)
23/119 (19.3)
17.1 (9.4, 24.8)
20/133 (15.0)
12.7 (6.2, 19.3)
9/108 (8.3)
27/129 (20.9)
12.7 (4.0, 21.4)
35/116 (30.2)
21.5 11.9, 31.1)
Week 24 4/123 (3.3)
25/119 (21.0)
18.2 (10.2, 26.1)
26/133 (19.5)
17.1 (9.8, 24.4)
12/108 (11.1)
28/129 (21.7)
10.7 (1.5, 19.9)
35/116 (30.2)
19.0 (9.0, 29.1)
ACR 20/50/70 = American College of Rheumatology 20%/50%/70% response; APR = apremilast; BID = twice daily; FAS = full analysis set; HAQ-DI = Health Assessment Questionnaire Disability Index; LS Mean = least-squares mean; PsARC = Psoriatic Arthritis Response Criteria; SF-36v2 PF = 36-item Short Form Health Survey, version 2, physical functioning domain score; PASI-75 = Psoriasis Area and Severity Index score improvement ≥ 75%; SE = standard error.
a N reflects number of subjects randomized; the number of subjects evaluated may vary by endpoint and time point. b n/m = number of responders/number of subjects in the subgroup. Subjects with a missing subgroup factor are not included. c Treatment difference for categorical endpoints is the adjusted difference in proportions (weighted average of the treatment differences across the strata of baseline disease-
modifying antirheumatic drug (DMARD) use and study with the Cochran-Mantel-Haenszel weights) and corresponding 2-sided 95% confidence interval based on a normal approximation to the weighted average. Treatment difference for continuous endpoints is the LS mean difference (based on an analysis of covariance model for the change from baseline, with treatment group, baseline DMARD use [yes/no], study, sex, and the treatment-by-sex interaction as factors and the baseline value as a covariate) and corresponding standard error.
d For modified ACR 20/50/70 and PsARC response, subjects who discontinued early prior to the respective visits, subjects who entered early escape (EE) at Week 16 (for the Week 24 analyses), and subjects who did not have sufficient data for a definitive determination of response status at the respective visits were counted as nonresponders. Joints temporarily or permanently not assessable at baseline were excluded from joint count. For other unassessed joints at baseline, the joint assessment at the Screening visit, if assessed, was used as the Baseline assessment; otherwise, the joint was excluded from joint count. The last observed joint assessment (at baseline or postbaseline) was used for joints unassessed at the respective visits. There was no imputation for other missing ACR component scores.
e For HAQ-DI and SF-36v2 PF, for subjects who discontinued from the study prior to Week 16, the last available postbaseline value observed prior to discontinuation was carried forward to Weeks 16 and 24. For subjects who entered EE at Week 16 or who did not enter EE but discontinued from the study between Weeks 16 and 24, the last available postbaseline value observed prior to EE or discontinuation, respectively, was carried forward to Week 24. Missing values for subjects who did not discontinue or enter EE were imputed using the latest available postbaseline value prior to the visit in question.
f For PASI-75, for subjects who discontinued from the study prior to Week 16, the last available postbaseline value observed prior to discontinuation was carried forward to Weeks 16 and 24. For subjects who entered EE at Week 16 or who did not enter EE but discontinued from the study between Weeks 16 and 24, the last available postbaseline value observed prior to EE or discontinuation, respectively, was carried forward to Week 24. Missing values for subjects who did not discontinue or enter EE were imputed using the latest available postbaseline value prior to the visit in question. Subjects who did not have sufficient data (observed or imputed) for a determination of response status at the respective visits were counted as nonresponders.
Female 252 -0.067 (0.0812) 266 -0.110 (0.0791) -0.043 (-0.261, 0.176) 273 -0.224 (0.0777) -0.157 (-0.374, 0.061) ACR = American College of Rheumatology; APR = apremilast; BID = twice daily; CI = confidence interval; EGA= Evaluator (Physician) global Assessment of Disease Activity; FAS = full analysis set; HAQ-DI = Health Assessment Questionnaire – Disability Index; CRP = C-reactive protein; LS = least squares; Pain VAS = subject’s assessment of pain visual analog scale; PGA = Patient’s (Subject’s) Global Assessment of Disease Activity; SE = standard error; SJC = swollen joint count; TJC = tender joint count. a Subjects with a baseline value and at least one postbaseline value at or prior to Week 16 are included. Subjects with a missing subgroup factor are not included.
b Estimates are based on an analysis of covariance model for the change from baseline, with treatment group, baseline disease-modifying antirheumatic drug (DMARD) (yes/no), study, the subgroup factor, and the treatment-by-subgroup-factor interaction as factors and the baseline value as a covariate.
Note: For subjects who discontinued from the study prior to Week 16, the last available postbaseline value observed prior to discontinuation was carried forward to Weeks 16 and 24. For subjects who entered early escape (EE) at Week 16 or who did not enter EE but discontinued from the study between Weeks 16 and 24, the last available postbaseline value observed prior to EE or discontinuation, respectively, was carried forward to Week 24. Missing values for subjects who did not discontinue or enter EE were imputed using the latest available postbaseline value prior to the visit in question. Source: 5.3.5.3.2 項 有効性併合解析(PsA)Table 3.12.1.
アプレミラスト 2.7.3 臨床的有効性(乾癬性関節炎) セルジーン株式会社
Confidential and Proprietary 121
Figure 50: Mean Change from Baseline in HAQ-DI at Week 24 in the APR 20 BID Treatment Group by PsA Subtype, Prior DMARD Use, and Baseline (Concomitant) DMARD Use (Pooled Analysis; FAS; LOCF)
Figure 51: Mean Change from Baseline in HAQ-DI at Week 24 in the APR 30 BID Treatment Group by PsA Subtype, Prior DMARD Use, and Baseline (Concomitant) DMARD Use (Pooled Analysis; FAS; LOCF)