What is the best biological treatment for rheumatoid arthritis? A systematic review of effectiveness

Table 1 Search strategies

PubMed

(((Arthritis, Rheumatoid[Text Word] or "Arthritis, Rheumatoid"[Mesh]) and (((((((((((rituximab[Text Word] or Mabthera[Text Word]) or Rituxan[Text Word]) or IDEC-C2B8 antibody[Text Word]) or "rituximab"[Supplementary Concept]) or (((((((TNFR-Fc fusion protein[Text Word] or TNR 001[Text Word]) or TNR-001[Text Word]) or TNF receptor type II-IgG fusion protein[Text Word]) or recombinant human dimeric TNF receptor type II-IgG fusion protein[Text Word]) or Enbrel[Text Word]) or etanercept[Text Word]) or "TNFR-Fc fusion protein"[Supplementary Concept])) or ((((infliximab[Text Word] or monoclonal antibody cA2[Text Word]) or MAb cA2[Text Word]) or Remicade[Text Word]) or "infliximab"[Sup plementary Concept])) or ((adalimumab[Text Word] or Humira[Text Word]) or "adalimumab"[Supplementary Concept])) or (((((certolizumab[Text Word] or CDP870[Text Word]) or CDP 870[Text Word]) or Cimzia[Text Word]) or certolizumab pegol[Text Word]) or "certolizumab pegol"[Supplementary Concept])) or ((((((((((((abatacept[Text Word] or BMS 188667[Text Word]) or BMS-188667[Text Word]) or nulojix[Text Word]) or CTLA-4-Ig[Text Word]) or cytotoxic T lymphocyte-associated antigen 4-immunoglobulin[Text Word]) or CTLA4-Fc[Text Word]) or CTLA4-Ig[Text Word]) or LEA29Y[Text Word]) or Orencia[Text Word]) or BELATACEPT[Text Word]) or BMS-224818[Text Word]) or "abatacept" [Supplementary Concept])) or (((tocilizumab[Text Word] or atlizumab[Text Word]) or Actemra[Text Word]) or "tocilizumab"[Supplementary Concept])) or ("golimumab"[Supplementary Concept] or (Simponi[Text Word] or golimumab[Text Word]))))) and (("Cohort Studies"[Mesh]) or (((cohort*[Text Word]) or controlled clinical trial[Publication Type]) or epidemiologic methods))

EMBASE

"golimumab"/exp and [embase]/lim or ("cnto$148" and [embase]/lim) or ("simponi" and [embase]/lim) or ("tocilizumab"/exp and [embase]/ lim) or ("actemra" and [embase]/lim) or ("actemra 200" and [embase]/lim) or ("atlizumab" and [embase]/lim) or ("r$1569" and [embase]/lim) or ("roactemra" and [embase]/lim) or ("abatacept"/exp and [embase]/lim) or ("bms$188667" and [embase]/lim) or ("ctla4$ig" and [embase]/ lim) or ("ctla4 immunoglobulin" and [embase]/lim) or ("ctla4 immunoglobulin g" and [embase]/lim) or ("orencia" and [embase]/lim) or ("certolizumab pegol"/exp and [embase]/lim) or ("cdp$870" and [embase]/lim) or ("cimzia" and [embase]/lim) or ("pha$738144" and [embase]/ lim) or ("adalimumab"/exp and [embase]/lim) or ("humira"/exp and [embase]/lim) or ("monoclonal antibody d2e7" and [embase]/lim) or ("trudexa" and [embase]/lim) or ("infliximab"/exp and [embase]/lim) or ("avakine" and [embase]/lim) or ("inflectra" and [embase]/lim) or ("remicade" and [embase]/lim) or ("remsima" and [embase]/lim) or ("revellex" and [embase]/lim) or ("etanercept"/exp and [embase]/lim) or ("embrel" and [embase]/lim) or ("enbrel" and [embase]/lim) or ("recombinant tumor necrosis factor receptor fc fusion protein" and [embase]/lim) or ("tnr$001" and [embase]/lim) or ("tumor necrosis factor receptor fc fusion protein" and [embase]/lim) or ("rituximab"/exp and [embase]/lim) or ("idec c2b8" and [embase]/lim) or ("mabthera" and [embase]/lim) or ("monoclonal antibody idec c2b8" and [embase]/lim) or ("reditux" and [embase]/lim) or ("rituxan" and [embase]/lim) or ("rituxin" and [embase]/lim) and ("rheumatoid arthritis"/exp and [embase]/lim or ("arthritis, rheumatoid" and [embase]/lim)) and ("cohort analysis"/exp and [embase]/lim or ("longitudinal study"/exp and [embase]/lim) or ("prospective study"/exp and [embase]/lim) or ("follow up"/exp and [embase]/lim) or ("cohort$" and [embase]/lim))

Cochrane Controlled Trials Register

#1 MeSH descriptor: [Arthritis, Rheumatoid] explode all trees

#2 Rheumatoid Arthritis in Trials

#3 golimumab in Trials

#4 tocilizumab in Trials

#5 abatacept in Trials

#6 certolizumab pegol in Trials

#7 adalimumab in Trials

#8 infliximab in Trials

#9 etanercept in Trials

#10 rituximab in Trials

#11 #1 or #2 in Trials

#12 #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 in Trials

#13 #11 and #12

LILACS

(tw:((mh:(arthritis, rheumatoid)) or (tw:(artrite reumatoide)) or (tw:(artritis reumatoide)) )) and (tw:((tw:(adalimumab)) or (tw:(etanercept)) or (tw:(infliximab)) or (tw:(rituximab)) or (tw:(golimumab)) or (tw:(tocilizumab)) or (tw:(abatacept)) or (tw:(certolizumab pegol))))

Table 2 Characteristics of included studies

Nstudy	Ref.	Type of study	Time horizon	Patient	Intervention	Country conducting the study	Funding Sources	Duration of the study (mo)	Follow-up (mo)
1	Geborek et al[12]	Cohort	Prospective	Naive	ETA vs IFX vs LEF	Sweden	NR	24	12
2	Van Vollenhoven et al[13]	Registry	Prospective	NR	ETA vs ETA + MTX	Sweden	Mixed	NR	12
3	Cohen et al[14]	Cohort	Retrospective	Therapeutic failure	IFX vs ETA	France	NR	48	3
4	Finckh et al[15]	Registry	Prospective	Mixed	ADA vs ETA vs IFX	Switzerland	Mixed	80	12
5	Heiberg et al[16]	Cohort	Prospective	Mixed	ADA monotherapy vs ADA + MTX	Norway	Mixed	NR	12
6	Hyrich et al[17,18]	Registry	Prospective	NR	ETA monotherapy vs ETA + MTX vs ETA + DMARD and ADA monotherapy vs ADA + MTX vs ADA + DMARD	England	Pharmaceutical industry	NR	6
7	Kristensen et al[19]	Cohort	Prospective	Naive	ETA vs IFX	Sweden	Mixed	55	36
8	Bernal Rivera et al[20]	Cohort	Prospective	Naive	ADA vs ETA vs IFX	Spain	NR	24	12
9	Kristensen et al[19]	Cohort	Prospective	Naive	ETA vs IFX	Spain	NR	72	6
10	Radstake et al[23]	Cohort	Prospective	NR	IFX vs ADA	The Netherlands	Mixed	NR	6
12	Bazzani et al[24]	Registry	Prospective	Mixed	ADA vs ETA vs IFX	Italy	Pharmaceutical industry	25.29	36
13	Greenwood et al[47]	Cohort	Retrospective	NR	ADA vs ETA vs IFX	England	NR	NR	12
14	Laas et al[25]	Cohort	Prospective	Naive	ETA vs ADA	Finland	No pharmaceutical industry	36	3
15	Arenere Mendoza et al[26]	Cohort	Retrospective	Mixed	ADA vs ETA vs IFX	Spain	NR	80	12
16	Buch et al[46]	Cohort	Prospective	Therapeutic failure	RTX vs anti-TNF	England	NR	NR	6
17	Canhão et al[27]	Registry	Prospective	Naive	ADA vs ETA vs IFX	Portugal	Mixed	NR	12
18	Hetland et al[28]	Registry	Prospective	Naive	ADA vs ETA vs IFX	Denmark	Mixed	86	12
19	Blom et al[29]	Registry	Prospective	Therapeutic failure	RTX vs anti-TNF	The Netherlands	Mixed	NR	12
20	Chatzidionysiou et al[30]	Registry	Prospective	Mixed	RTX monotherapy vs RTX + MTX vs RTX + LEF	Europe	Pharmaceutical industry	NR	12
21	Gotenberg et al[45]	Registry	Prospective	Mixed	RTX vs ABAT	France	NR	NR	6
22	Iannone et al[32]	Registry	Prospective	NR	ADA vs ETA vs IFX	Italy	NR	NR	48
23	Leffers et al[33]	Registry	Prospective	Mixed	ABAT vs TOCI	Denmark	Mixed	NR	48
24	Martínez-Pérez et al[44]	Cohort	Retrospective	Mixed	RTX vs IFX	Spain	NR	NR	12
25	Wakabayashi et al[34]	Cohort	Retrospective	Therapeutic failure	TOCI vs ETA	Japan	No pharmaceutical industry	60	12
26	Finckh et al[36]	Cohort	Prospective	Therapeutic failure	RTX vs anti-TNF	Switzerland	Mixed	NR	24
27	Gomez-Reino et al[35]	Cohort	Prospective	Therapeutic failure	RTX vs anti-TNF	Spain	Pharmaceutical industry	36	12
28	Greenberg et al[37]	Registry	Prospective	Naive	ADA vs ETA vs IFX	Unied States	Mixed	74	24
29	Kekow et al[38]	Cohort	Retrospective	Therapeutic failure	RTX vs anti-TNF	Germany	Pharmaceutical industry	NR	6
30	Schabert et al[39]	Cohort	Retrospective	NR	ADA vs ETA vs IFX	Unied States	Mixed	15	12
31	Chatzidionysiou et al[40]	Registry	Prospective	Therapeutic failure	Anti-TNF vs ETA vs ADA	Stockholm	NR	NR	6
32	Keystone et al[43]	Cohort	Retrospective	Therapeutic failure	ABAT vs TOCI	Canada	NR	NR	12
33	Emery et al[41]	Cohort	Prospective	Therapeutic failure	RTX vs anti-TNF	Multicentre	Mixed	NR	12
34	Flouri et al[42]	Registry	Prospective	Mixed	ADA vs ETA vs IFX	Greece	Mixed	60	12
35	Harrold et al[31]	Registry	Prospective	Therapeutic failure	ABAT vs TOCI	Unied States	Mixed	NR	12

ADA: Adalimumab; ETA: Etanercept; IFX: Infliximab; RTX: Rituximab; ABAT: Abatacept; TOCI: Tocililizumab; LEF: Leflunomid; MTX: Methotrexate; sDMARD: Synthetic disease-modifying antirheumatic drugs; NR: Not reported.

Table 3 Quality assessment of articles for Newcastle Ottawa scale

N study	Ref.	Selection				Comparability	Results			Total
		Representati-veness of the cases	Selection of controls	Ascertain-ment of exposure	Demonstration that outcome of interest was not present at start of study	Comparability of cohorts on the basis of the design or analysis	Assessment of outcome	Was follow-up long enough for outcomes to occur	Adequacy of follow up of cohorts
1	Geborek et al[12]	1	1	1	1	2	0	1 (12 mo)	1	8
2	Van Vollenhoven et al[13]	0	1	1	1	2	0	1 (24 mo)	0	6
3	Cohen et al[14]	1	1	1	1	1	0	1 (3 mo)	0	6
4	Finckh et al[15]	1	1	1	0	1	1	1 (12 mo)	1	7
5	Heiberg et al[16]	1	1	1	1	2	0	1 (6 mo)	1	8
6	Hyrich et al[17,18]	1	1	1	1	2	0	1 (6 mo)	1	8
7	Kristensen et al[19]	1	1	1	0	2	0	1 (36 mo)	0	6
8	Bernal Rivera et al[20]	1	1	1	0	2	0	1 (12 mo )	1	7
9	Fernández-Nebro et al[21]	1	1	1	1	2	0	1 (6 mo)	1	8
10	Radstake et al[23]	0	1	1	1	2	0	1 (6 mo)	0	6
11	Kievit et al[22]	1	1	1	1	2	0	1 (6 mo)	0	7
12	Bazzani et al[24]	1	1	1	0	2	1	1 (6 mo)	1	8
13	Greenwood et al[47]	0	1	1	1	2	0	1 (12 mo)	0	6
14	Laas et al[25]	1	1	1	1	2	1	1 (3 mo)	1	9
15	Arenere Mendoza et al[26]	1	1	1	1	2	0	1 (12 mo)	0	7
16	Buch et al[46]	1	1	1	0	2	0	1 (6 mo)	0	6
17	Canhão et al[27]	1	1	1	0	2	0	1 (12 mo)	1	7
18	Hetland et al[28]	1	1	1	1	2	0	1 (12 mo)	1	8
19	Blom et al[29]	1	1	1	1	2	1	1 (12 mo)	1	9
20	Chatzidionysiou et al[30]	1	1	1	1	2	0	1 (12 mo)	0	7
21	Gotenberg et al[45]	1	1	1	1	2	0	1 (6 mo)	1	8
22	Iannone et al[32]	1	1	1	0	2	0	1 (48 mo)	0	6
23	Leffers et al[33]	1	1	1	1	2	0	1 (12 mo)	0	7
24	Martínez-Pérez et al[44]	0	1	0	1	2	0	1 (12 mo)	0	5
25	Wakabayashi et al[34]	0	1	1	1	2	1	1 (12 mo)	1	8
26	Finckh et al[36]	1	1	1	1	1	1	1 (24 mo)	1	8
27	Gomez-Reino et al[35]	1	1	1	0	2	1	1 (12 mo)	1	8
28	Greenberg et al[37]	1	1	1	1	2	0	1 (24 mo)	1	8
29	Kekow et al[38]	1	1	1	1	2	0	1 (6 mo)	1	8
30	Schabert et al[39]	1	1	1	1	1	1	1 (12 mo)	1	8
31	Chatzidionysiou et al[40]	1	1	1	0	2	0	1 (6 mo)	0	6
32	Keystone et al[43]	0	1	1	0	2	0	1 (12 mo)	0	5
33	Emery et al[41]	1	1	1	0	2	0	1(12 mo)	0	6
34	Flouri et al[42]	1	1	1	1	2	0	1 (12 mo)	1	8
35	Harrold et al[31]	1	1	1	0	2	0	1 (12 mo)	0	6

Table 4 Meta-analysis of the outcomes for patients with treatment-naïve and therapeutic failure

Intervention	Outcomes	Studies (references)	Partici-pants	Relative risk (95%CI) or other mesure	I2(%)	P value
IFX vs ETA	EULAR good response	10 (18,19,21,22,24, 26,27, 28, 32,42)	7247	0.86 [0.72-1.02]	76	< 0.0001
	EULAR moderate response	9 (18,19,21,22,24, 26, 28, 32,42)	6791	0.98 [0.84-1.15]	78	< 0.0001
	EULAR no response	9 (18,19,21,22,24, 26, 28, 32,42)	6791	1.20 [1.05-1.38]	46	0.06
	DAS 28 remission	7 (21,26,27,28,32,37,42)	2868	0.70 [0.59-0.84]	0	0.51
	DAS 28	2 (21,26)	196	0.40 [-0.27- 1.07]	59	0.12
	DAS 28 reduction	2 (15,22)	1321	0.40 [0.04-0.77]	77	0.04
	CDAI remission	4 (27,28,37,42)	2293	0.90 [0.74-1.09]	0	0.89
	SDAI remission	2 (27,42)	840	0.87 [0.61-1.26]	0	0.9
	HAQ	3 (21,26,39)	495	0.14 [0.00-0.27]	0	0.51
	ACR 20	2 (19,37)	1309	0.95 [0.86-1.06]	0	0.47
	ACR50	3 (19,28,37)	2315	0.92 [0.81-1.03]	10	0.33
	ACR70	3 (19,28,37)	2315	0.88 [0.57-1.36]	79	0.009
ADA vs ETA	EULAR good response	8 (20,22,24,26,27,28,32,42)	2492	0.97 [0.79-1.20]	73	0.0005
	EULAR moderate response	7 (20,22,24,26,28,32,42)	2080	1.00 [0.89-1.12]	0	0.48
	EULAR no response	7 (20,22,24,26,28,32,42)	2080	0.90 [0.62-1.32]	76	0.0003
	DAS 28 remission	6 (26,27,28,32,37,42)	2412	0.93 [0.68-1.26]	80	0.0001
	DAS 28	2 (20,26)	180	-0.09 [-0.25-0.06]	0	0.73
	DAS 28 reduction	2 (15,22)	1392	0.17 [-0.19-0.52]	68	0.08
	CDAI remission	4 (27,28,37,42)	1883	1.16 [0.77-1.74]	70	0.02
	SDAI remission	2 (27,42)	641	1.40 [0.76-2.59]	55	0.13
	HAQ	2 (26,39)	339	-0.15 [-0.39-0.10]	49	0.16
	HAQ reduction	2 (22,25)	653	-0.07 [-0.16-0.03]	0	0.92
	ACR 20	2 (20,37)	445	0.89 [0.71-1.12]	0	0.68
	ACR 50	3 (20,28,37)	1217	1.09 [0.91-1.31]	18	0.3
	ACR 70	3 (20,28,37)	1436	1.15 [0.92-1.43]	0	0.82
IFX vs ADA	EULAR good response	8 (22,23,24,26,27,28,32,42)	3025	1.25 [1.06-1.47]	57	0.02
	EULAR moderate response	7 (22,23,24,26,28,32,42)	2657	0.91 [0.79-1.04]	31	0.19
	EULAR no response	7 (22,23,24,26,28,32,42)	2657	0.77 [0.56-1.05]	75	0.0006
	DAS 28 remission	6 (26,27,28,32,37,42)	2760	1.15 [0.91-1.46]	63	0.02
	DAS 28 reduction	2 (15,22)	1097	-0.24 [-0.96-0.48]	91	0.001
	CDAI remission	4 (27,28,37,42)	2332	1.30 [0.90-1.88]	68	0.02
	SDAI remission	2 (27,42)	765	1.66 [0.94-2.93]	61	0.11
	HAQ	2 (26,39)	182	-0.33 [-0.53-0.13]	0	0.92
	ACR 50	2 (28,37)	1458	1.14 [0.71-1.84]	79	0.03
	ACR 70	2 (28,37)	1458	1.41 [0.81-2.44]	72	0.06

A value of I² > 40% indicates statistical heterogeneity between the studies. A value of P < 0.10 from the chi-square test indicates statistical heterogeneity between the studies. CI: Confidence interval; ADA: Adalimumab; ETA: Etanercept; IFX: Infliximab; EULAR: European League Against Rheumatism; DAS 28: Disease activity score; SDAI: Simplified disease activity index; CDAI: Clinical disease activity; HAQ: Health Assessment Questionnaire; ACR: American College Rheumatology.

Table 5 Meta-analysis of the outcomes for anti-tumor necrosis factor naïve patients

Intervention	Outcomes	Studies (references)	n	RR (95%CI)or other mesure	I2(%)	P value
IFX vs ETA	EULAR good response	5 (19, 21, 22, 27, 28)	2822	0.82 [0.62-1.09]	82	0.0001
	EULAR moderate response	4 (19, 21, 22, 28)	2366	0.90 [0.61-1.33]	90	< 0.00001
	EULAR no response	4 (19, 21, 22, 28)	2366	1.29 [1.09-1.53]	27	0.25
	DAS 28 remission	4 (21, 27, 28, 37)	1804	0.82 [0.70-0.95]	0	0.4
	ACR 20	2 (19, 37)	1309	0.95 [0.86-1.06]	0	0.47
	ACR50	3 (19, 28, 37)	2315	0.92 [0.81-1.03]	10	0.33
	ACR70	3 (19, 28, 37)	2315	0.88 [0.57-1.36]	79	0.009
	CDAI remission	3 (27, 28, 37)	1876	0.88 [0.72-1.08]	0	0.93
ADA vs ETA	EULAR good response	4 (20, 22, 27, 28)	1590	1.11 [1.00-1.23]	0	0.4
	EULAR moderate response	3 (20, 22, 28)	1178	1.01 [0.83-1.24]	19	0.29
	EULAR no response	3 (20, 22, 28)	1178	0.69 [0.53-0.89]	11	0.32
	DAS 28 remission	3 (27, 28, 37)	1380	1.03 [0.82-1.29]	37	0.21
	ACR 20	2 (20, 37)	445	0.89 [0.71-1.12]	0	0.68
	ACR 50	3 (20, 28, 37)	1217	1.09 [0.91-1.31]	18	0.3
	ACR 70	3 (20, 28, 37)	1436	1.15 [0.92-1.43]	0	0.82
	HAQ reduction	2 (22, 25)	653	-0.07 [-0.16-0.03]	0	0.92
	CDAI remission	3 (27, 28, 37)	1601	1.02 [0.67-1.56]	72	0.03
IFX vs ADA	EULAR good response	3 (22, 27, 28)	1706	1.42 [1.18-1.72]	42	0.18
	EULAR moderate response	2 (22, 28)	1338	0.96 [0.58-1.59]	80	0.03
	EULAR no response	2 (22, 28)	1338	0.56 [0.45-0.69]	0	0.88
	DAS 28 remission	3 (27, 28)	1648	1.23 [0.95-1.59]	48	0.15
	ACR 50	2 (28, 37)	1458	1.14 [0.71-1.84]	79	0.03
	ACR 70	2 (28, 37)	1458	1.41 [0.81-2.44]	72	0.06
	CDAI remission	3 (27, 28, 37)	1875	1.17 [0.75-1.82]	75	0.02

A value of I² > 40% indicates statistical heterogeneity between the studies. A value of P < 0.10 from the chi-square test indicates statistical heterogeneity between the studies. CI: Confidence interval; ADA: Adalimumab; ETA: Etanercept; IFX: Infliximab; EULAR: European League Against Rheumatism; DAS 28: Disease activity score; CDAI: Clinical disease activity; HAQ: Health assessment questionnaire; ACR: American College Rheumatology.

Table 6 Meta-analysis of the outcomes for patients with anti-tumor necrosis factor therapeutic failure

Intervention	Outcomes	Studies (references)	n	Relative risk (95%CI)or other mesure	I2(%)	P value
RTX vs anti-TNF	EULAR good response	4 (35, 38, 40, 46)	1608	0.96 [0.60-1.54]	74	0.009
	EULAR moderate response	5 (29, 35, 38, 40, 46)	1706	1.02 [0.79-1.32]	66	0.02
	EULAR no response	3 (35, 38, 40)	1406	1.00 [0.53-1.89]	85	0.001
	DAS 28 reduction	6 (35, 36, 38, 40, 41, 46)	1584	0.42 [-0.65--0.20]	62	0.02
ETA vs control	EULAR good response	2 (14, 40)	173	2.11 [1.23-3.62]	0	0.48
	IFX		38	1.60 [0.63-4.09]
	RTX		135	2.42 [1.25-4.68]
	DAS 28 reduction	2 (34, 40)	152	0.15 [-0.65-0.95]	77	0.04
	RTX		113	-0.22 [-0.64-0.20]
	TOCI		39	0.60 [-0.05-1.25]

A value of I² > 40% indicates statistical heterogeneity between the studies. A value of P < 0.10 from the chi-square test indicates statistical heterogeneity between the studies. CI: Confidence interval; RTX: Rituximab; ETA: Etanercept; IFX: Infliximab; TOCI: Tocilizumab; EULAR: European League Against Rheumatism; DAS 28: Disease activity score.

Table 7 Meta-analysis of the outcomes for patients in treatment with biological monotherapy vs biological in combination with methotrexate

Intervention	Outcomes	Studies (references)	Participants	Relative risk (95%CI) or other mesure	I2(%)	P value
bDMARD monotherapy vs bDMARD + MTX	EULAR good response	3 (16,17, 30)	3000	0.57 [0.34-0.95]	82	0.0008
	DAS 28	3 (17, 20, 30)	2913	0.25 [-0.02-0.52]	69	0.01
	HAQ	2 (165, 30)	655	0.13 [0.03-0.22]	0	0.43

A value of I² > 40% indicates statistical heterogeneity between the studies. A value of P < 0.10 from the χ² test indicates statistical heterogeneity between the studies. CI: Confidence interval; bDMARD: Biological disease-modifying antirheumatic drugs; MTX: Methotrexate; EULAR: European League Against Rheumatism; DAS 28: Disease activity score; HAQ: Health Assessment Questionnaire.