NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
Donahue KE, Gartlehner G, Schulman ER, et al. Drug Therapy for Early Rheumatoid Arthritis: A Systematic Review Update [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2018 Jul. (Comparative Effectiveness Review, No. 211.)
Appendix Table E-1Disease activity, remission, radiographic outcomes, functional status, and harms (KQs 1–3)a
Intervention and Comparisons | Outcome | Study Design | Number of Studies; # of Subjects | Study Limitations | Consistency | Directness | Precision | Other limitations | Results | Strength of Evidence |
---|---|---|---|---|---|---|---|---|---|---|
Corticosteroid vs. csDMARDs | Disease activity | Trials | 5 RCTs: 2 double-blinded, 3 open label; N=1307 | High: open label design and high attrition | Inconsistent: direction of effect varies | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | Mixed for disease activity | Insufficient |
Remission | Trials | 5 RCTs: 2 double-blinded, 3 open label; N=1395 | Medium: open label design and high attrition | Consistent | Direct | Imprecise: not enough events to meet optimal information size | None | Higher remission in corticosteroid + MTX vs. MTX | Low | |
Radiographic changes | Trials | 4 RCTs: 2 double-blinded, 2 open label; N=1344 | Medium: open label design and high attrition | Inconsistent: direction of effect varies | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | Mixed results for radiographic changes | Insufficient | |
Functional capacity | Trials | 4 RCTs: 2 double-blinded, 2 open label; N=1344 | High: open label design and high attrition | Inconsistent: direction of effect varies | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | Mixed for functional capacity | Insufficient | |
D/C due to AEs | Trials | 4 RCTs: 2 double-blinded, 2 open label; N=1185 | Medium: open label design and high attrition | Consistent | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | No significant differences | Low | |
Serious AEs | Trials | 3 RCTs: 2 double-blinded, 1 open label; N =1085 | Medium: open label design and high attrition | Consistent | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | No significant differences | Moderate | |
High dose corticosteroid vs. IFX | Response | Trials | 2 double-blinded RCTs; N=156 | Medium: open label design | Consistent | Direct | Imprecise: large CIs cross appreciable benefits or harms, and optimal information size not met | None | No significant differences in ACR response | Insufficient |
Remission | Trials | 2 double-blinded RCTs; N=156 | Medium: open label design | Consistent | Direct | Imprecise: large CIs cross appreciable benefits or harms, and optimal information size not met | None | No significant differences in remission | Insufficient | |
Radiographic changes | Trial | 1 double-blinded RCT; N=112 | Medium: open label design | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, and optimal information size not met | None | No significant differences in SHS scores | Insufficient | |
Functional capacity | Trials | 2 double-blinded RCTs; N=156 | Medium: open label design | Inconsistent | Direct | Imprecise: not enough events to meet optimal information size | N/A | Mixed results for functional capacity | Insufficient | |
D/C due to AEs | Trials | 2 double-blinded RCTs; N=156 | Medium: open label design | Consistent | Direct | Imprecise: large CIs cross appreciable benefits or harms, and optimal information size not met | None | No differences in d/c due to AEs | Insufficient | |
Serious AEs | Trials | 2 double- blinded RCTs; N=156 | Medium: open label design | Inconsistent | Direct | Imprecise: large CIs cross appreciable benefits or harms, and optimal information size not met | None | Higher SAE in IFX + MTX vs. Methyl-PNL + MTX | Insufficient | |
High dose corticosteroid vs. csDMARD monotherapy | Response | Trial | 1 double-blinded RCT; N=44 | Medium | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, and not enough events to meet optimal information size | None | No significant differences in ACR response | Insufficient |
Remission | Trial | 1 double-blinded RCT; N=44 | Medium | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, and not enough events to meet optimal information size | None | No significant differences in remission | Insufficient | |
Functional capacity | Trial | 1 double-blinded RCT; N=44 | Medium | Unknown | Direct | Imprecise: : large CIs cross appreciable benefits or harms, and not enough events to meet optimal information size | None | Greater improvement in functional capacity in IV methyl-PNL + MTX vs. MTX | Insufficient | |
D/C due to AEs | Trial | 1 double-blinded RCT; N=44 | Medium | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, and not enough events to meet optimal information size | None | No significant differences in d/c due to AEs | Insufficient | |
Serious AEs | Trial | 1 double-blinded RCT; N=44 | Medium | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, and not enough events to meet optimal information size | None | No significant differences in SAEs | Insufficient | |
csDMARD monotherapy vs. csDMARD monotherapy | Disease activity | Trial | 1 double-blinded RCT; N=245 | High: high attrition, and large baseline differences between groups | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | No significant differences in disease activity in PNL + SSZ vs. PNL + MTX | Insufficient |
Disease activity | Cohort | 1 Cohort; N=1102 | High: confounding by indication | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, and optimal information size not met | None | No significant difference in disease activity in SSZ vs. MTX | Insufficient | |
Remission | Trial | 1 double-blinded RCT; N=245 | High: high attrition and direction of effect varies | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, and optimal information size not met | None | No significant differences in remission PNL + SSZ vs. PNL + MTX | Insufficient | |
Radiographic changes | Trial | 1 double-blinded RCT; N=245 | High: high attrition and large baseline differences between groups | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, and optimal information size not met | None | No significant differences in Larsen score in PNL + SSZ vs. PNL + MTX | Insufficient | |
Functional capacity | Trial | 1 double-blinded RCT; N=245 | High: high attrition and large baseline differences between groups | Unknown | Direct | Imprecise: not enough events to meet optimal information size | N/A | No significant differences in functional capacity in PNL + SSZ vs. PNL + MTX | Insufficient | |
Functional capacity | Cohort | 1 Cohort; N=1102 | High: confounding by indication | Unknown | Direct | Precise | None | No significant difference in functional capacity in SSZ vs. MTX | Insufficient | |
D/C due to AEs | Trial | 1 double-blinded RCT; N=245 | High: high attrition | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, and optimal information size not met | None | Higher d/c in SSZ + PNL vs. MTX + PNL | Insufficient | |
D/C due to AEs | Cohort | 1 Cohort; N=1102 | High: confounding by indication | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, and optimal information size not met | None | Higher d/c with SSZ vs. MTX | Insufficient | |
csDMARD combination therapy vs. csDMARD monotherapy | Disease activity | Trials | 5 double-blinded RCTs; N=1183 | Medium: open label design and high attrition | Consistent | Direct | Imprecise: large CIs cross appreciable benefits or harms, and optimal information size not met | None | No significant differences in disease activity (DAS, ACR response) for comparisons of MTX + SSZ vs. MTX | Low |
Disease activity | Cohort | 1 Cohort; N=230 | High: high attrition | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, and optimal information size not met | None | No significant difference in disease activity for MTX + SSZ vs. MTX | Insufficient | |
Radiographic changes | Trials | 5 double-blinded RCTs; N=1242 | Medium: high attrition | Inconsistent | Direct | Imprecise: large CIs cross appreciable benefits or harms, and optimal information size not met | None | Mixed results for radiographic changes | Insufficient | |
Functional capacity | Trials | 6 double-blinded RCTs; N=1347 | Medium: open label design and high attrition | Consistent | Direct | Imprecise: large CIs cross appreciable benefits or harms | N/A | No significant differences in functional capacity for comparisons of MTX + SSZ vs. MTX at 1 year or 5 years. No difference in functional capacity for comparisons of PNL + MTX + SSZ + HCQ vs. MTX or SSZ | Low | |
D/C due to AEs | Trials | 6 double-blinded RCTs; N=1347 | Medium: open label design and high attrition | Consistent | Direct | Imprecise | None | No significant differences | Low | |
D/C due to AEs | Cohort | 1 Cohort; N=230 | High: high attrition and high risk of selection bias for treatment discontinuation and confounding by indication | Unknown | Direct | Imprecise: not enough events to meet optimal information size | None | No significant differences | Insufficient | |
Serious AEs | Trials | 6 double-blinded RCTs; N =1347 | Medium: open label design, and high attrition | Consistent | Direct | Imprecise | None | No significant differences | Low | |
csDMARD plus TNF biologic vs. TNF biologic | ||||||||||
ADA + MTX vs. ADA or ADA vs. MTX | Response | Trial | 1 double-blinded RCT; N=799 | Medium: high attrition | Unknown | Direct | Precise | N/A | Higher ACR50 response for comparison of ADA + MTX vs. ADA | Moderate |
Remission | Trial | 1 double-blinded RCT; N=799 | Medium: high attrition | Unknown | Direct | Precise | None | Higher remission for ADA + MTX vs. ADA | Moderate | |
Radiographic changes | Trial | 1 double-blinded RCT; N=799 | Medium: high attrition | Unknown | Direct | Precise | None | Lower modified Sharp Score change for ADA + MTX vs. ADA | Moderate | |
Functional capacity | Trial | 1 double-blinded RCT; N=799 | Medium: high attrition | Unknown | Direct | Precise | N/A | Greater improvement in functional capacity in ADA + MTX vs. ADA | Moderate | |
D/C due to AEs | Trial | 1 double-blinded RCT; N=799 | Medium: high attrition | Unknown | Direct | Precise | None | No significant differences | Moderate | |
Serious AEs | Trial | 1 double-blinded RCT; N=799 | Medium: high attrition | Unknown | Direct | Precise | None | No significant differences | Moderate | |
csDMARD plus Non-TNF biologic vs. Non-TNF biologic | ||||||||||
ABA + MTX vs. ABA or ABA vs. MTX | Response | Trial | 1 double-blinded RCT; N=351 | Medium: high attrition | Unknown | Direct | Precise | None | No significant differences | Low |
Remission | Trial | 1 double-blinded RCT; N=351 | Medium: high attrition | Unknown | Direct | Precise | None | No significant differences | Low | |
Functional capacity | Trial | 1 double-blinded RCT; N=351 | Medium: high attrition | Unknown | Direct | Precise | None | No significant differences | Low | |
D/C due to AEs | Trial | 1 double-blinded RCT; N=351 | Medium: high attrition | Unknown | Direct | Precise | None | No significant differences | Low | |
Serious AEs | Trial | 1 double-blinded RCT; N=351 | Medium: high attrition | Unknown | Direct | Precise | None | No significant differences | Low | |
TCZ + MTX vs. TCZ | Disease activity | Trials | 2 double-blinded RCTs; N=1479 | Medium | Inconsistent: direction of effect varies | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | Mixed results for disease activity (DAS) for TCZ + MTX vs. TCZ | Insufficient |
Remission | Trials | 2 double-blinded RCTs; N=1479 | Medium | Consistent | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | Higher remission for TCZ + MTX vs. TCZ | Low | |
Radiographic changes | Trials | 2 double-blinded RCTs; N=1479 | Medium | Consistent | Direct | Precise | None | Lower Sharp score changes in TCZ + MTX vs. TCZ | Moderate | |
Functional capacity | Trials | 2 double-blinded RCTs; N=1479 | Medium | Inconsistent: direction of effect varies | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | Mixed results for functional capacity at 52 weeks for TCZ + MTX vs. TCZ | Insufficient | |
D/C due to AEs | Trials | 2 double-blinded RCTs; N=1479 | Medium | Consistent | Direct | Precise | None | No significant differences | Moderate | |
Serious AEs | Trials | 2 double-blinded RCTs; N=1479 | Medium | Consistent | Direct | Precise | None | No significant differences | Moderate | |
csDMARDs vs. tsDMARDs | Disease activity | Trial | 1 double-blinded RCT; N=108 | Medium: high attrition | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, and not enough events to meet optimal information size | None | Higher DAS and ACR50 response for TOF + MTX vs. MTX | Insufficient |
Remission | Trial | 1 double-blinded RCT; N=108 | Medium: high attrition | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, and not enough events to meet optimal information size | None | Higher remission for TOF + MTX vs. TOF or MTX | Insufficient | |
Radiographic changes | Trial | 1 double-blinded RCT; N=108 | Medium: high attrition | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, and not enough events to meet optimal information size | None | Lower Sharp score changes with TOF compared with TOF + MTX or MTX | Insufficient | |
csDMARDs vs. tsDMARDs | Functional capacity | Trial | 1 double-blinded RCT; N=108 | Medium | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, and not enough events to meet optimal information size | N/A | No difference in functional capacity between TOF + MTX vs. MTX | Insufficient |
D/C due to AEs | Trial | 1 double-blinded RCT; N=108 | Medium: high attrition | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, and not enough events to meet optimal information size | None | No significant differences | Insufficient | |
Serious AEs | Trial | 1 double-blinded RCT; N=108 | Medium: high attrition | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, and not enough events to meet optimal information size | None | No significant differences | Insufficient | |
TNF biologic plus csDMARD vs. csDMARD | ||||||||||
ADA + MTX vs. MTX | Disease activity | Trials | 5 RCTs: 3 double-blinded, 2 open label; N=2485 | Medium: high attrition | Consistent | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | Higher ACR50 response with ADA + MTX vs. MTX | Low |
Remission | Trials | 5 RCTs: 3 double-blinded, 2 open label; N=2485 | Medium: high attrition | Consistent | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | Higher remission with ADA + MTX vs. MTX | Low | |
Radiographic changes | Trials | 5 RCTs: 3 double-blinded, 2 open label; N=2485 | Medium: high attrition | Consistent | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | Lower Sharp score changes for ADA + MTX vs. MTX | Low | |
Functional capacity | Trials | 5 RCTs: 3 double-blinded, 2 open label N=2485 | Medium: high attrition | Consistent | Direct | Precise | None | Greater improvement in functional capacity for ADA + MTX vs. MTX | Moderate | |
D/C due to AEs | Trials | 5 RCTs: 3 double-blinded, 2 open label; N=2485 | Medium: high attrition | Inconsistent: direction of effect varies | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | No differences | Low | |
Serious AEs | Trials | 5 RCTs: 3 double-blinded, 2 open label; N=2485 | Medium: high attrition | Inconsistent: direction of effect varies | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | No differences | Low | |
CZP + MTX vs. MTX | Disease activtiy | Trial | 1 double blined RCT; N=879 | Medium: high attrition | Unknown | Direct | Imprecise: optimal information size not met, and large CIs | None | Higher ACR50 response at 52 wks for CZP + MTX vs. MTX | Low |
Remission | Trials | 2 double-blinded RCT; N=1195 | Medium: high attrition | Unknown | Direct | Imprecise: optimal information size not met, and large CIs | None | Higher DAS remission for CZP + MTX vs. MTX | Low | |
Radiographic changes | Trials | 2 double-blinded RCT; N=1195 | Medium: high attrition | Unknown | Direct | Imprecise: optimal information size not met, and large CIs | None | Lower mTSS change for CZP + MTX vs. MTX | Low | |
Functional capacity | Trials | 2 double-blinded RCT; N=1195 | Medium: high attrition | Consistent | Direct | Imprecise: optimal information size not met, and large CIs | None | Greater improvement in HAQ-DI in CZP + MTX vs. MTX group at 52 weeks | Low | |
D/C due to AEs | Trials | 2 double-blinded RCT; N=1195 | Medium: high attrition | Unknown | Direct | Imprecise: optimal information size not met, and large CIs | None | No differences | Low | |
Serious AEs | Trials | 2 double-blinded RCT; N=1195 | Medium: high attrition | Unknown | Direct | Imprecise: optimal information size not met, and large CIs | None | No differences | Low | |
ETN + MTX vs. MTX and ETN vs. MTX | Disease activity | Trials | 3 double-blinded RCTs; N=2000 | Medium | Consistent | Direct | Precise | None | Higher ACR20 response for ETN + MTX and ETN vs. MTX | Moderate |
Remission | Trial | 1 double-blinded RCT; N=542 | Medium | Unknown | Direct | Imprecise: not enough events to meet optimal information size | None | Higher remission for ETN + MTX and ETN vs. MTX | Low | |
Radiographic changes | Trials | 2 double-blinded RCTs; N=1174 | Medium | Consistent | Direct | Precise | None | Lower Sharp score change for ETN + MTX and ETN vs. MTX | Moderate | |
Functional capacity | Trials | 3 double-blinded RCTs; N=2000 | Medium | Inconsistent: direction of effect varies | Direct | Imprecise: large CIs | None | Mixed results for functional capacity | Low | |
D/C due to AEs | Trials | 3 double-blinded RCTs; N=2000 | Medium | Consistent | Direct | Imprecise: not enough events to meet optimal information size | None | No differences | Low | |
Serious AEs | Trials | 2 double-blinded RCTs; N=2000 | Medium | Consistent | Direct | Imprecise: not enough events to meet optimal information size | None | No differences | Low | |
IFX + MTX vs. MTX | Disease activity | Trials | 3 double-blinded RCTs; N=1113 | Medium | Inconsistent: direction of effect varies | Direct | Imprecise: large CIs cross appreciable benefits or harms, not enough events to meet optimal information size | None | Mixed results for ACR response | Insufficient |
Remission | Trials | 3 double-blinded RCTs; N = 1113 | Medium | Consistent | Direct | Precise | None | Higher remission for IFX + MTX vs. MTX | Low | |
Radiographic changes | Trials | 2 double-blinded RCTs; N=1069 | Medium | Inconsistent: direction of effect varies | Direct | Imprecise: large CIs cross appreciable benefits or harms, not enough events to meet optimal information size | None | Mixed results for radiographic progression | Insufficient | |
Functional capacity | Trials | 3 double-blinded RCTs; N=1113 | Medium | Consistent | Direct | Precise | None | Greater functional capacity with IFX + MTX vs. MTX | Low | |
D/C due to AEs | Trials | 2 double-blinded RCTs; N = 1093 | Medium | Consistent | Direct | Precise | None | No differences | Low | |
Serious AEs | Trials | 2 double-blinded RCTs; N = 1093 | Medium | Consistent | Direct | Precise | None | No differences | Low | |
TNF biologic vs. csDMARD combination therapy | ||||||||||
ADA + MTX vs. MTX + PRED + HCQ + SSZ | Disease actvitiy | Trial | 1 double-blinded RCT; N=161 | High: high attrition | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, not enough events to meet optimal information size | None | No differences in DAS | Insufficient |
Remission | Trial | 1 double-blinded RCT; N=161 | High: high attrition | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, not enough events to meet optimal information size | None | No differences in remission | Insufficient | |
Radiographic changes | Trial | 1 double-blinded RCT; N=161 | High: high attrition | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, not enough events to meet optimal information size | None | No difference in radiographic score progression | Insufficient | |
Functional capacity | Trial | 1 double-blinded RCT; N=161 | High: high attrition | Unknown | Direct | Imprecise: not enough events to meet optimal information size | None | No difference in functional capacity | Insufficient | |
Serious AEs | Trial | 1 double-blinded RCT; N=161 | High: high attrition | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, not enough events to meet optimal information size | None | No differences | Insufficient | |
IFX + MTX vs. MTX + SSZ + HCQ | Disease activity | Trial | 1 double-blinded RCT; N=258 | Medium | Unknown | Direct | Precise | None | Increased ACR50 response for IFX + MTX vs. MTX + SSZ + HCQ | Low |
D/C due to AEs | Trial | 1 double-blinded RCT; N=258 | Medium | Unknown | Direct | Precise | None | No differences | Low | |
Serious AEs | Trial | 1 double-blinded RCT; N=258 | Medium | Unknown | Direct | Precise | None | No differences | Low | |
IFX + MTX + SSZ + HCQ + PRED vs. MTX + SSZ + HCQ + PRED | Disease activity | Trial | 1 double-blinded RCT; N=99 | Low | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, not enough events to meet optimal information size | None | No differences in ACR responses | Low |
Remission | Trial | 1 double-blinded RCT; N=99 | Low | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, not enough events to meet optimal information size | None | No differences in remission | Low | |
Radiographic changes | Trial | 1 double-blinded RCT; N=99 | Low | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, not enough events to meet optimal information size | None | No differences in radiographic score progression | Low | |
Functional capacity | Trial | 1 double-blinded RCT; N=99 | Low | Unknown | Direct | Imprecise: not enough events to meet optimal information size | None | No difference in functional capacity | Low | |
D/C due to Aes | Trial | 1 double-blinded RCT; N=99 | Low | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, not enough events to meet optimal information size | None | No differences | Low | |
Serious AEs | Trial | 1 double-blinded RCT; N=99 | Low | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, not enough events to meet optimal information size | None | No differences | Low | |
Non-TNF biologic vs. csDMARD monotherapy | ||||||||||
ABA + MTX vs. MTX | Disease activity | Trials | 2 double-blinded RCTs; N=860 | Medium: high attrition, and large baseline differences between groups | Consistent | Direct | Precise | None | Improved disease activity with ABA + MTX vs. MTX | Moderate |
Remission | Trials | 2 double-blinded RCTs; N = 860 | Medium: high attrition | Consistent | Direct | Precise | None | Higher remission rates for ABA + MTX vs. MTX | Moderate | |
Radiographic changes | Trials | 1 double-blinded RCT; N=509 | Medium: high attrition | Unknown | Direct | Precise | None | Lower Genant-modified Sharp scores in ABA + MTX vs. MTX | Low | |
Functional capacity | Trials | 2 double-blinded RCTs; N = 860 | Medium: high attrition | Inconsistent: direction of effect varies | Direct | Imprecise: large CIs cross appreciable benefits or harms, not enough events to meet optimal information size | None | Mixed results for functional capacity between ABA + MTX vs. MTX | Low | |
D/C due to AEs | Trial | 1 double-blinded RCT; N=509 | Medium: high attrition | Unknown | Direct | Precise | None | No differences | Low | |
Serious AEs | Trial | 1 double-blinded RCT; N=509 | Medium: high attrition | Unknown | Direct | Precise | None | No differences | Low | |
RIT + MTX vs. MTX | Disease activity | Trial | 1 double-blinded RCT; N=755 | Low | Unknown | Direct | Imprecise: not enough events to meet optimal information size | None | Improved disease activity with RIT + MTX vs. MTX | Moderate |
Remission | Trial | 1 double-blinded RCT; N=755 | Low | Unknown | Direct | Imprecise: not enough events to meet optimal information size | None | Higher remission with RIT + MTX vs. MTX | Moderate | |
Radiographic changes | Trial | 1 double-blinded RCT; N=755 | Low | Unknown | Direct | Imprecise: not enough events to meet optimal information size | None | Lower Genant-modified Sharp scores in RIT + MTX vs. MTX | Moderate | |
Functional capacity | Trial | 1 double-blinded RCT; N=755 | Low | Unknown | Direct | Imprecise: not enough events to meet optimal information size | None | Greater improvement in functional capacity in RIT + MTX vs. MTX | Moderate | |
D/C due to AEs | Trial | 1 double-blinded RCT; N=755 | Low | Unknown | Direct | Imprecise: not enough events to meet optimal information size | None | No differences | Moderate | |
Serious AEs | Trial | 1 double-blinded RCT; N=755 | Low | Unknown | Direct | Imprecise: not enough events to meet optimal information size | None | No differences | Moderate | |
TCZ + MTX vs. MTX | Disease activity | Trials | 2 double-blinded RCTs; N=1479 | Medium | Inconsistent: direction of effect varies | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | Mixed results for disease activity (DAS) | Insufficient |
Remission | Trials | 2 double-blinded RCTs; N=1479 | Medium | Consistent | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | Higher remission for TCZ + MTX vs. MTX | Low | |
Radiographic changes | Trials | 2 double-blinded RCTs; N=1479 | Medium: large baseline differences between groups | Consistent | Direct | Precise | None | Lower Sharp score changes in TCZ + MTX vs. MTX | Moderate | |
Functional capacity | Trials | 2 double-blinded RCTs; N=1479 | Medium | Inconsistent: direction of effect varies | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | Mixed results for functional capacity at 52 weeks for TCZ + MTX vs. MTX | Insufficient | |
D/C due to AEs | Trials | 2 double-blinded RCTs; N=1479 | Medium | Consistent | Direct | Precise | None | No significant differences | Moderate | |
Serious AEs | Trials | 2 double-blinded RCTs; N=1479 | Medium | Consistent | Direct | Precise | None | No significant differences | Moderate | |
TNF vs. Non-TNF | Disease activity | Trial | 1 open label RCT; N=329 | High: no ITT analysis | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, not enough events to meet optimal information size | None | No significant differences in DAS for RIT vs. ADA or ETN | Insufficient |
Remission | Trial | 1 open label RCT; N=329 | High: no ITT analysis | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, not enough events to meet optimal information size | None | No significant differences in remission for RIT vs. ADA or ETN | Insufficient | |
Functional capacity | Trial | 1 open label RCT; N=329 | High: no ITT analysis | Unknown | Direct | Precise | N/a | Greater improvement in functional capacity in RIT vs. TNF biologic (ADA or ETN) | Low | |
D/C due to AEs | Trial | 1 open label RCT; N=329 | High: no ITT analysis | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, not enough events to meet optimal information size | None | No differences | Insufficient | |
Serious AEs | Trial | 1 open label RCT; N=329 | High: no ITT analysis | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, not enough events to meet optimal information size | None | Higher for RIT vs. ADA or ETN | Insufficient | |
Combination strategies | ||||||||||
1-Sequential monotherapy starting with MTX vs. 2- step-up combination therapy vs. 3-combination with high-dose tapered prednisone vs. 4-combination therapy with infliximab | Disease activity | Trial | 1 double-blinded RCT; N=508 | Low | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | Lower disease activity scores for 3 (combination therapy with high dose prednisone) and 4 (combination therapy with IFX) than 1 (sequential DMARD therapy) or 2 (step-up therapy) at one year but no differences at 4 yrs and 10 years. | Moderate |
Remission | Trial | 1 double-blinded RCT; N=508 | Low | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | No differences in remission at 4 yrs and 10 years | Moderate | |
Radiographic changes | Trial | 1 double-blinded RCT; N=508 | Low | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | Lower Sharp/van der Heijde radiographic changes in groups 3-combination therapy with high dose prednisone) and 4(combination therapy with IFX) than 1 (sequential DMARD therapy) or 2 (step up therapy) at 5 years but no differences at 10 years. | Moderate | |
Functional capacity | Trial | 1 double-blinded RCT; N=508 | Low | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms, not enough events to meet optimal information size | None | Greater functional capacity in groups 3 (combination therapy with high dose prednisone) and 4 (combination therapy with IFX) than 1 (sequential DMARD therapy) or 2 (step up therapy) at 12 months, but no significant difference at 2 years, 5 years or 10 years. | Low | |
Serious AEs | Trial | 1 double-blinded RCT; N=508 | Low | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | No significant differences | Low | |
1-immediate MTX + ETN vs. 2-immediate MTX + SSZ + HCQ vs. 3-step up MTX to combo MTX + ETN vs. 4-step up MTX to combo MTX + SSZ + HCQ | Disease activity | Trial | 1 double-blinded RCT; N=755 | High: high attrition and no ITT analysis | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | The 2 immediate groups (groups 1 and 2) had improved disease activity compared with step up (groups 3 and 4) at 6 months, but no differences at 2 yrs | Insufficient |
Remission | Trial | 1 double-blinded RCT; N=755 | High: high attrition and no ITT analysis | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | No significant changes in remission at 2 yrs | Insufficient | |
Radiographic changes | Trial | 1 double-blinded RCT; N=755 | High: high attrition and no ITT analysis | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | No significant changes in radiographic scores at 2 yrs | Insufficient | |
Functional capacity | Trial | 1 double-blinded RCT; N=755 | High: high attrition and no ITT analysis | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | No significant differences at 48 and 102 weeks | Insufficient | |
D/C due to AEs | Trial | 1 double-blinded RCT; N=755 | High: high attrition and no ITT analysis | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | No significant differences | Insufficient | |
Serious AEs | Trial | 1 double-blinded RCT; N=755 | High: high attrition and no ITT analysis | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | No significant differences | Insufficient | |
ADA + MTX adjusted based on DAS vs. MTX | Disease Activity | Trials | 2 double-blinded RCTs; N=245 | High: high attrition | Consistent | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | No differences in ACR response at 2 yrs | Insufficient |
Remission | Trials | 2 double-blinded RCTs; N=245 | High: high attrition | Consistent | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | No differences in remission at 2 yrs | Insufficient | |
Radiographic changes | Trials | 2 double-blinded RCTs; N=245 | High: high attrition | Consistent | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | No differences in radiographic changes at 2 yrs | Insufficient | |
Functional capacity | Trials | 2 double-blinded RCTs; N=245 | High: high attrition | Inconsistent | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | Mixed results for functional capacity at 1 yr | Insufficient | |
D/C due to AEs | Trial | 1 double-blinded RCT; N=180 | High: high attrition | Unknown | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | No significant differences | Insufficient | |
Serious AEs | Trials | 2 double-blinded RCTs; N=245 | High: high attrition | Consistent | Direct | Imprecise: large CIs cross appreciable benefits or harms | None | No significant differences | Insufficient |
- a
Consistent with network meta-analysis. For the SOE for effect estimates derived from indirect comparisons only (i.e., no head to head trials), the SOE for all estimates was low. We downgraded for indirectness and precision in all cases. The NWMA model included only studies with low or unclear risk of bias, therefore we did not downgrade for study limitations. Because of the single estimate derived from the NWMA, we also did not downgrade for inconsistency.
ABA = abatacept; ACR = American College of Rheumatology; ACR50 = American College of Rheumatology 50% improvement; ADA = adalimumab; AEs = adverse events; CI = confidence interval; csDMARD = conventional synthetic disease-modifying antirheumatic drug; CZP=certolizumab pegol; d = day; DAS = Disease Activity Score; D/C = discontinuation; DMARD = disease-modifying antirheumatic drug(s); ETN = etanercept; HAQ = Health Assessment Questionnaire; HCQ = hydroxychloroquine; IFX = infliximab; ITT = intent-to-treat; mTSS = modified Sharp/van der Heijde score; MTX = methotrexate; N = number of patients; NA = not applicable; NWMA = network meta-analysis; obs = observational; PNL = prednisolone; PRED = prednisone; RCT = randomized controlled trial; RIT = rituximab; SAE = serious adverse events; SHS = Sharp/van der Heijde score; SOE = strength of evidence; SSZ = sulfasalazine; TCZ = tocilizumab; TNF = tumor necrosis factor; tsDMARD = targeted synthetic disease-modifying antirheumatic drug; vs. = versus; yr = year.
Appendix Table E-2Subgroup analyses for benefit and harms outcomes (KQ 4)
Intervention and Comparisons | Outcome | Study Design | Number of Studies; N of Subjects | Study Limitations | Consistency | Directness | Precision | Reporting Bias | Other Limitations | Results | Strength of Evidence |
---|---|---|---|---|---|---|---|---|---|---|---|
TNF biologic plus csDMARD vs. csDMARD: ADA + MTX vs. MTX | Disease activity/radiographic change | Trial | 1 double-blinded RCT; N=171 | High: no test of interaction for subgroup analyses; results based on regression analyses | Unknown | Direct | Imprecise: study does not meet optimal information size for subgroup analyses | Undetected | None | Disease activity is significantly associated with radiographic change | Insufficient |
TNF biologic vs. csDMARD: ETN vs. MTX | Age/response | Trial | 1 double-blinded RCT; N=424 | High: no test of interaction for subgroup analyses; results based on regression analyses | Unknown | Direct | Imprecise: no test of interaction for subgroup analyses | Undetected | None | Lower ACR response rates for older (>65 years) compared with younger patients | Insufficient |
Age/SAE | Trial | 1 double-blinded RCT; N=424 | High: no test of interaction for subgroup analyses; results based on regression analyses | Unknown | Direct | Imprecise: no test of interaction for subgroup analyses | Undetected | None | Higher risk of serious adverse events for older (>65 years) compared with younger patients | Insufficient | |
TNF biologic plus csDMARD vs. csDMARD: IFX + MTX vs. MTX | Disease activity/radiographic change | Trial | 1 double-blinded RCT; N=1049 | High: no test of interaction for subgroup analyses; results based on regression analyses | Unknown | Direct | Imprecise: Study does not meet optimal information size for subgroup analyses | Undetected | None | Disease activity is significantly associated with radiographic change | Insufficient |
TNF biologic vs. csDMARD combo therapy: IFX + MTX vs. csDMARD combo | Obesity/remission | Trial | 1 open-label RCT; N=260 | High: no test of interaction for subgroup analyses; results based on regression analyses | Unknown | Direct | Imprecise: Study does not meet optimal information size for subgroup analyses | Undetected | None | Obesity is significantly associated with lower rates of remission | Insufficient |
Obesity/response | Trial | 1 open-label RCT; N=260 | High: no test of interaction for subgroup analyses; results based on regression analyses | Unknown | Direct | Imprecise: Study does not meet optimal information size for subgroup analyses | Undetected | None | Obesity is significantly associated with lower rates of response | Insufficient |
ACR = American College of Rheumatology; ADA = adalimumab; csDMARD = conventional synthetic disease-modifying antirheumatic drug(s); ETN = etanercept; IFX = infliximab; IV = intravenous; MTX = methotrexate; N = number of patients; NA = not applicable; obs = observational; RCT = randomized controlled trial; SAE = serious adverse events; TNF = tumor necrosis factor; vs. = versus.