Table 3

Head-to-head studies comparing bDMARDs to other bDMARDs

Population	Study	Risk of bias	Treatment	N	Primary endpoint	P value	ACR20 (%)	ACR50 (%)	ACR70 (%)	DAS28 <2.6 (%)	CDAI ≤2.8 (%)	ΔHAQ
MTX-IR	Burmester 2017 (MONARCH)63 137	Low	ADA 40 mg Q2W	185	ΔDAS28-ESR at week 24	<0.001	58	30	12	7	3	−0.43
	Burmester 2017 (MONARCH)63 137	Low	SLM 200 mg Q2W	184	ΔDAS28-ESR at week 24	<0.001	72	46	23	27	7	−0.61
	Smolen 2016 (EXXELERATE)4	Low	ADA 40 mg Q2W+MTX	454	ACR 20 (%) at week 12	0.532	71				22
	Smolen 2016 (EXXELERATE)4	Low	CZP 400/200 mg Q2W+MTX	454	ACR 20 (%) at week 12	0.532	69				25
	Taylor 2018 (SIRROUND-H)66	Low	ADA 40 mg Q2W	186	ACR 50 (%) + ΔDAS28-ESR at week 24	Reference	57	32	13	8		−0.52
			SKM 50 mg Q4W	186		0.306/0.013	54	27	12	13		−0.51
			SKM 100 mg Q2W	187		0.464/ <0.001	59	35	16	20		−0.53
	Genovese 2018b65	Low	ADA 40 mg Q2W+MTX	56	ACR 20 (%) at week 12	Reference	68	48	21	30	7	−0.6
			ABT-122 60 mg Q2W+MTX	55		0.863	62	35	22	22	7	−0.6
			ABT-122 120 mg Q2W+MTX	56		0.414	75	46	18	38	11	−0.6
			ABT-122 120 mg QW +MTX	55		0.196	80	47	36	42	11	−0.9
csDMARD-IR	Porter 2016 (ORBIT)62	High	Anti-CD20 (RTX)	140	ΔDAS28-ESR (non-inferiority) at week 52	0.24	66	49	23	23		−0.49
csDMARD-IR	Porter 2016 (ORBIT)62	High	TNFi (ETA/ADA)	134	ΔDAS28-ESR (non-inferiority) at week 52	0.24	71	45	26	21		−0.38
TNF-IR	Blanco 2017 (NURTURE 1)138	Low	Placebo +csDMARD	138	ACR 20 (%) at week 24	Reference	18	9	5			−0.3
			ABA 500/750/1000mg+csDMARD	138		<0.05	43	28	12			−0.6
			SEC 10 mg/kg i.v. +150 mg s.c. Q4W+csDMARD	137		0.031	31	17	10			−0.4
			SEC 10 mg/kg i.v. +75 mg s.c. Q4W+csDMARD	138		0.092	28	12	5			−0.3
Mixed cs/bDMARD-IR	Weinblatt 2018 (EARTH EXPLORER 2)*64	Low	GLM 50 mg Q4W	68	ACR 20/50/70%, DAS28-CRP <2.6, ΔHAQ>0.22 at week 24	0.666/0.293/0.156/0.108/0.208	66	43	26	29	18	−0.64
Mixed cs/bDMARD-IR	Weinblatt 2018 (EARTH EXPLORER 2)*64	Low	MVM 100 mg Q2W+MTX	70	ACR 20/50/70%, DAS28-CRP <2.6, ΔHAQ>0.22 at week 24	0.666/0.293/0.156/0.108/0.208	62	35	16	17	6	−0.44

Results of secondary efficacy outcomes are shown at the time point of the primary endpoint.
*Study not powered to formally compare the treatments. Detailed results of risk of bias analyses are shown in online supplementary table S2.2 in the supplementary appendix.
Δ, change from baseline; ABA, abatacept; ACR, American College of Rheumatology; ADA, adalimumab; bDMARDs, biological disease-modifying antirheumatic drugs; CDAI, clinical disease activity index; CRP, C-reactive protein; csDMARDs, conventional synthetic disease-modifying antirheumatic drugs; CZP, certolizumab pegol; DAS28, Disease Activity Score of 28 joints; ESR, erythrocyte sedimentation rate; ETA, etanercept; GLM, golimumab; HAQ, Health Assessment Questionnaire; i.v., intravenous; MTX, methotrexate; MVM, mavrilimumab; RTX, rituximab; SEC, secukinumab; SKM, sirukumab; SLM, sarilumab; TNFi, TNF inhibitor; TNF-IR, tumour necrosis factor-insufficient responder.