Radiographic joint damage in early rheumatoid arthritis patients: comparing tocilizumab- and methotrexate-based treat-to-target strategies

Radiographic joint damage in early rheumatoid arthritis patients: comparing tocilizumab- and... Abstract Objective To evaluate the progression of erosions and joint space narrowing (JSN) in feet and hands in the U-Act-Early trial. Methods In this trial, 317 newly diagnosed DMARD-naïve RA patients initiated randomly tocilizumab, or step-up MTX or a combination of the two. Radiographs were scored at baseline and after 52 and 104 weeks using the Sharp–van der Heijde erosion and JSN score. Between the strategy arms, changes from baseline and the proportions of patients without radiographic progression (change from baseline ≤0) were compared. Results Mean changes from baseline in erosion and JSN scores for the whole study population were after 52 weeks 0.59 and 0.18 and after 104 weeks 0.70 and 0.50, respectively. For JSN, at both time points no differences in progression were found between strategies (P ⩾ 0.09). For erosions, the progression was significantly lower at week 104 in both tocilizumab arms when compared with the MTX arm ((p≤0.023). Less progression of erosions in the feet was found after 104 weeks in both tocilizumab arms (P ⩽ 0.046); this was not significant for the hands (P ⩾ 0.11). The proportion of patients without progression in erosions was higher in the tocilizumab arms at week 52 (tocilizumab plus MTX: 87%, P = 0.038; tocilizumab: 81%, P = 0.29) and 104 (tocilizumab plus MTX: 85%, P = 0.001; tocilizumab: 77%, P = 0.028), compared with the MTX arm (74 and 60%, respectively). Conclusion In DMARD-naïve early RA patients, initiating a tocilizumab-based treat-to-target strategy inhibits the progression of erosions, especially in the feet, more compared with initiation of a step-up MTX strategy. Trial registration ClinicalTrials.gov, https://clinicaltrials.gov, NCT01034137. rheumatoid arthritis, tocilizumab, methotrexate, radiographic joint damage, Sharp/van der Heijde score Rheumatology key messages Fewer erosions were found in the tocilizumab-based strategies when compared with the methotrexate-based strategy. Especially in the feet there was less progression of erosions in RA patients initiating tocilizumab. No differences were found between strategies in joint space narrowing of the hands and feet. Introduction In RA, structural joint damage as a consequence of synovitis is commonly seen [1]. Typically bones are affected, resulting in erosions and loss of cartilage, leading to joint space narrowing (JSN), which eventually causes joint deformities and functional impairments if joint inflammation is not targeted adequately. With the development of conventional synthetic (csDMARDs) and biologic DMARDs (bDMARDs) and the implementation of treat-to-target approaches, aiming for clinical remission, a remarkable gain is made in preventing joint damage and therefore in improving the long-term outcome of RA patients [2]. As smaller joints are more frequently affected in early RA, radiographs of both hands and feet are the current standard to assess structural changes, and radiography is therefore considered to be a pivotal tool for assessing therapeutic efficacy. International recommendations/guidelines for the management of RA recommend that a csDMARD, preferably MTX, should be (part of) the first treatment strategy and that, if the treatment target is not achieved, adding a bDMARD may be considered [3, 4]. To date, MTX remains the anchor drug in treating newly diagnosed RA because of its beneficial efficacy–safety and cost-effectiveness ratios [3]. Nevertheless, in about 30% of the patients achieving the treatment target on this drug, progression of radiographic changes is still observed as a result of subclinical inflammation [5, 6]. Starting therapy with a bDMARD, with or without MTX, may improve the clinical outcome by more effectively reducing disease activity. Tapering and finally discontinuing the biological once remission is achieved might minimize health care costs and the number of patients who are over-treated, as some would have achieved the treatment target also with MTX therapy alone. In the 2-year, multicentre, phase 3, double-blind, placebo-controlled U-Act-Early strategy trial (ClinicalTrials.gov identifier NCT01034137), DMARD-naïve patients with early RA were randomized to start treatment with tocilizumab, step-up MTX or a combination of the two and were treated to target until sustained remission (defined as DAS28 <2.6 with four or fewer swollen joints for ⩾24 weeks) was achieved. Clinical efficacy and safety data have been reported previously in which significantly less joint damage, as assessed by the Sharp–van der Heijde score (SHS) [7], was found in both tocilizumab arms when compared with the MTX arm [8]. This article reports in more detail the effect of the treatment strategies on the progression of erosions and JSN separately, according to the SHS method, in the feet and hands. In addition, we evaluated if there was an association between progression of joint damage and the three therapeutic strategies and whether this association was in part independent from suppression of disease activity, as reported in literature for other biologicals [9–12]. Methods Study design and population Patients with newly diagnosed RA visiting one of the 21 participating rheumatology departments in the Netherlands were invited to participate in the U-Act-Early strategy trial. To be eligible, patients had to be >18 years, be DMARD-naïve, meet the 1987 ACR or the 2010 ACR/EULAR classification criteria for RA, have active disease (DAS28 ⩾2.6), and have been diagnosed with RA within the previous 12 months before inclusion. Patients were randomized (1:1:1) in permuted blocks of six, stratified by centre and baseline DAS28 (i.e. DAS28 <5.1 or ⩾5.1) to initiate tocilizumab plus MTX, or tocilizumab plus placebo-MTX, or MTX plus placebo-tocilizumab therapy and were treated to target until sustained remission was achieved. Placebo/tocilizumab (8 mg/kg) was given every 4 weeks intravenously up to a maximum of 800 mg per dose. MTX (orally) was started at 10 mg/week and increased stepwise with 5 mg every 4 weeks to 30 mg/week until remission or the maximum tolerable dose was achieved; the same scheme was used for placebo-MTX. If remission was not achieved with the initial therapy, HCQ was added as a second step in the initial treatment strategies. Hereafter, if the target still was not reached, the initial strategies ended and the subsequent treatment strategy was started. Patients randomized to the tocilizumab or MTX arm then discontinued the placebo and switched to tocilizumab plus MTX combination therapy; those who initially started with the combination switched to the standard of care (e.g. MTX combined with a TNF inhibitor). If the treatment target, sustained remission, was achieved, medication was tapered stepwise and finally stopped. First placebo/MTX was tapered with 5 mg every 4 weeks down to 10 mg/week and then discontinued; thereafter the dose of placebo/tocilizumab was decreased to 4 mg/kg and stopped after 3 months. Sustained drug-free remission (sDFR) was achieved when patients were drug-free for ⩾12 weeks while in sustained remission (one visit with low disease activity (DAS28 ⩽3.2) was allowed). A detailed description of the study design, including treatment strategies, has been previously reported [8]. The medical ethics research committee of the University Medical Centre Utrecht approved the study for all participating hospitals. All patients signed informed consent before entering the study. Sharp–van der Heijde score The SHS was used to evaluate progression of radiographic joint damage. Radiographs of the hands (including wrists) and feet were taken at baseline and after 52 and 104 weeks. Sixteen and six areas were considered for assessing erosions and JSN in both hands and feet, respectively. The maximum erosion score of hands and wrists is 160 and of the feet it is 120 (maximum total erosion score 280). The maximum JSN score of hands and wrists is 120 and of the feet it is 48 (maximum total JSN score 168). The sum of the erosion and JSN scores is the total SHS (maximum 448). All radiographs were scored centrally in chronological order by a blinded (experienced and trained) professional reader. Automated joint space width quantification To assess the JSN of the hands with a more sensitive method, we used an automated method for measuring joint space width (JSW) as previously described by Huo et al. [13, 14]. Briefly, this computerized joint space quantification method measures in millimetres the JSW of PIP joints, MCP joints and joints of the wrist, consisting of the trapezium-trapezoides-navicular, carpo-navicular-lunate and the radiocarpal joints, bilaterally. After the joint margins were automatically delineated, the average Euclidean distance between the upper and lower margin was calculated within the middle 60% of the joint span and the JSW was determined for each joint separately. The average JSW scores of all the joints (i.e. PIP, MCP and wrist joints) of both extremities were used to calculate the total JSW hand score. The automated joint space quantification method has been proven effective in measuring radiographic changes in the hands within DMARD-naïve patients with early RA [15]; JSW quantification of the foot joints with this method has to date not yet been developed. Statistical analyses Analyses were performed based on the intention-to-treat population: those who were randomized and received at least one dose of medication. Radiographic joint scores are described as mean (s.d.) and median (interquartile range, IQR). For between-group comparisons at follow-up, we calculated changes from baseline in total SHS, including erosion and JSN scores and both subscores separately in the hands and feet, and analysed these using a two-sided stratified non-parametric Mann–Whitney (Van Elteren) test, controlling for the two randomization stratification factors, baseline DAS28 (i.e. DAS28 <5.1 or ⩾5.1) and centre. The same analyses were performed for JSN scores of the hand as measured by the computerized JSW quantification method. In addition, the proportions of patients without radiographic progression (change from baseline ⩽0) were analysed using a Pearson chi-square test. To evaluate if the progression in radiographic joint damage within the strategy arms over time was associated with disease activity (as expected) and whether an effect on progression of joint damage of the treatment strategies remained after correction for disease activity, we performed longitudinal regression analysis (mixed model analysis using a random intercept). The dependent variable in this analysis was the delta (Δ) SHS over time (i.e. change from baseline to week 52 and change from week 52 to 104) and baseline DAS28, baseline SHS and treatment strategy were used as time-independent covariates and time (week after baseline) and the average disease activity (mean DAS28) during a progression period as time-varying covariates. These longitudinal regression analyses were also performed for the JSN and erosion scores separately. Furthermore, to evaluate the effect of tapering and discontinuing therapy as defined in the design of this study, we compared the change in SHS (as well as for erosion and JSN) in longitudinal regression analyses between patients achieving sDFR vs those achieving the treatment target, sustained remission (i.e. not able to discontinue medication). Missing radiograph scores were imputed using data from the withdrawal visit with a maximum time window of 180 days. All analyses were performed based on the entire treatment strategy, thus including the addition of HCQ and switching from placebo to active therapy (in the tocilizumab and MTX arms). Patients switching to the standard of care according to protocol were no longer followed within the study. A P < 0.05 (two tailed) was considered statistically significant; R version 3.4.0 was used for data analyses. Results Patient population Of the recruited patients, 317 were eligible to participate in the U-Act-Early study (tocilizumab plus MTX arm, 106; tocilizumab arm, 103; MTX arm, 108); only minor differences in the baseline characteristics between the three strategy arms were noted [8]. The majority of the study population was female (67%), was seropositive for RF (72%) or anti-CCP (70%), had on average high disease activity [mean (s.d.) DAS28 5.2 (1.1) at baseline with a median (IQR) symptom duration of 26 (16–43) days], and had none or minor radiographic joint damage [median (IQR) SHS 0 (0–1)]. In total, radiographs were obtained of 97 (92%), 92 (89%) and 91 (84%) patients at week 52 and of 76 (72%), 75 (73%) and 75 (69%) patients at week 104 in the tocilizumab plus MTX, tocilizumab and MTX arms, respectively. Patient disposition After 52 weeks of treatment, 95% of the patients in the tocilizumab plus MTX- and 92% in the tocilizumab-based treatment strategies were still on the initial treatment regimen; in the MTX arm this proportion was 61% (Fig. 1). The proportions receiving or having received HCQ in addition, as part of the initial treatment strategy, during the first 52 weeks of the study were 30% in the tocilizumab plus MTX arm, 21% in the tocilizumab arm and 62% in the MTX arm. During the second year, the proportions of patients within the treatment strategies who received or had received HCQ further increased only marginally (tocilizumab plus MTX, 37%; tocilizumab, 22%; MTX, 66%). After 104 weeks, the large majority of the patients in the tocilizumab-based strategies still remained on the initial treatment regimen (tocilizumab plus MTX, 92%; tocilizumab, 87%), in contrast to the MTX strategy arm (54%). Fig. 1 View largeDownload slide Proportion (%) of patients remaining on the initial regimen during the entire study TCZ: tocilizumab. Fig. 1 View largeDownload slide Proportion (%) of patients remaining on the initial regimen during the entire study TCZ: tocilizumab. Radiographic joint damage The cumulative frequencies of changes from baseline to weeks 52 and 104 in the SHS are shown in Fig. 2. In the previously published results of U-Act-Early it was reported that changes from baseline in radiographic joint damage, as evaluated by the SHS, were significantly lower in the tocilizumab plus MTX arm compared with the MTX arm after 52 weeks (P = 0.016; Table 1) [8]. After 104 weeks, less progression in joint damage was noted in both tocilizumab arms (tocilizumab plus MTX, P = 0.021; tocilizumab, P = 0.038) when compared with the MTX arm. In the present study, when evaluating the changes of the JSN and erosion scores separately, no significant differences were found in the JSN scores at week 52 (P ⩾ 0.09) or 104 (P ⩾ 0.25) between the three treatment arms. For erosions, the changes from baseline to week 52 did not statistically differ between the treatment arms (P ⩾ 0.20), but after 104 weeks, significantly lower erosion progression scores were found in the tocilizumab plus MTX (P = 0.016) and tocilizumab arm (P = 0.023) when compared with the MTX arm. Figure 3 shows proportions of patients without radiographic progression (change from baseline ⩽0). In those assigned to initiate tocilizumab plus MTX, the proportions of patients without progression in SHS were significantly higher at weeks 52 (84% vs 67%, P = 0.009) and 104 (74% vs 52%, P = 0.006), compared with those assigned to the MTX strategy; for the tocilizumab strategy, the comparisons with the MTX strategy were not statistically significant (77%, P = 0.13 and 65%, P = 0.10, respectively). The proportion of patients without progression in erosions was higher in the tocilizumab-based strategies at week 52 (tocilizumab plus MTX: 87%, P = 0.038; tocilizumab: 81%, P = 0.29) and at week 104 (tocilizumab plus MTX: 85%, P < 0.001; tocilizumab: 77%, P = 0.028), compared with the MTX strategy (74% and 60%, respectively). No significant differences were found between the strategies in the proportion of patients without progression in JSN (P ⩾ 0.22). When analysing the changes from baseline in JSN and erosion scores of both the hands and feet separately, no significant differences were found between the strategies in both extremities in JSN scores during the trial period (P ⩾ 0.12; Table 2). Also, no significant differences were noted in the erosion score of hands during follow-up (P ⩾ 0.11). After 104 weeks, significantly less progression of erosions was found in the feet in both tocilizumab arms (tocilizumab plus MTX, P = 0.046; tocilizumab, P = 0.022) compared with in the MTX arm; this was not significant at week 52 (P ⩾ 0.36). Table 1 Change from baseline in radiographic joint damagea Joint damage  Tocilizumab plus MTX (n = 106)  Tocilizumab (n = 103)  MTX (n = 108)  Comparative tests; P-value  Total SHSb      Week 52        TCZ + MTX vs MTX; P = 0.016          Mean (s.d.)  0.50 (1.50)  0.79 (3.24)  0.96 (2.87)  TCZ vs MTX; P = NS          Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  TCZ + MTX vs TCZ; P = NS      Week 104        TCZ + MTX vs MTX; P = 0.021          Mean (s.d.)  1.18 (3.92)  1.45 (4.27)  1.53 (2.42)  TCZ vs MTX; P = 0.038          Median (IQR)  0.00 (0.00–1.00)  0.00 (0.00–2.00)  0.00 (0.00–2.56)  TCZ + MTX vs TCZ; P = NS  JSN score      Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.20          Mean (s.d.)  0.10 (0.47)  0.24 (1.19)  0.21 (0.73)            Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)        Week 104        TCZ + MTX vs TCZ vs MTX; P ≥ 0.25          Mean (s.d.)  0.18 (0.72)  0.77 (2.41)  0.53 (1.58)            Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  Erosion score      Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.20          Mean (s.d.)  0.40 (1.35)  0.57 (2.38)/  0.79 (3.00)            Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–1.00)      Week 104        TCZ + MTX vs MTX; P = 0.016          Mean (s.d.)  0.41 (1.25)  0.67 (2.58)/  1.01 (1.68)  TCZ vs MTX; P = 0.023          Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–1.00)  TCZ+MTX vs TCZ; P = NS  Joint damage  Tocilizumab plus MTX (n = 106)  Tocilizumab (n = 103)  MTX (n = 108)  Comparative tests; P-value  Total SHSb      Week 52        TCZ + MTX vs MTX; P = 0.016          Mean (s.d.)  0.50 (1.50)  0.79 (3.24)  0.96 (2.87)  TCZ vs MTX; P = NS          Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  TCZ + MTX vs TCZ; P = NS      Week 104        TCZ + MTX vs MTX; P = 0.021          Mean (s.d.)  1.18 (3.92)  1.45 (4.27)  1.53 (2.42)  TCZ vs MTX; P = 0.038          Median (IQR)  0.00 (0.00–1.00)  0.00 (0.00–2.00)  0.00 (0.00–2.56)  TCZ + MTX vs TCZ; P = NS  JSN score      Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.20          Mean (s.d.)  0.10 (0.47)  0.24 (1.19)  0.21 (0.73)            Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)        Week 104        TCZ + MTX vs TCZ vs MTX; P ≥ 0.25          Mean (s.d.)  0.18 (0.72)  0.77 (2.41)  0.53 (1.58)            Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  Erosion score      Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.20          Mean (s.d.)  0.40 (1.35)  0.57 (2.38)/  0.79 (3.00)            Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–1.00)      Week 104        TCZ + MTX vs MTX; P = 0.016          Mean (s.d.)  0.41 (1.25)  0.67 (2.58)/  1.01 (1.68)  TCZ vs MTX; P = 0.023          Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–1.00)  TCZ+MTX vs TCZ; P = NS  Joint damage assessed according to the Sharp–van der Heijde method. aData not normally distributed. bPreviously published data. JSN: joint space narrowing; NS: not significant; SHS: Sharp–van der Heijde score; TCZ: tocilizumab. Table 2 Change from baseline in erosion and joint space narrowinga Joint damage in hands/feet  Tocilizumab plus MTX (n = 106)  Tocilizumab (n = 103)  MTX (n = 108)  Comparative tests; P-value  JSN score   Hands          Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.16              Mean (s.d.)  0.04 (0.25)  0.17 (1.05)  0.08 (0.32)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)          Week 104        TCZ + MTX vs TCZ vs MTX; P ≥ 0.68              Mean (s.d.)  0.11 (0.48)  0.41 (1.74)  0.20 (0.81)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)   Feet          Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.12              Mean (s.d.)  0.06 (0.29)  0.07 (0.50)  0.16 (0.60)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)          Week 104        TCZ + MTX vs TCZ vs MTX; P ≥ 0.12              Mean (s.d.)  0.07 (0.35)  0.35 (1.33)  0.36 (1.24)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  Erosion score   Hands          Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.18              Mean (s.d.)  0.15 (0.96)  0.44 (2.25)  0.29 (1.34)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)          Week 104        TCZ + MTX vs TCZ vs MTX; P ≥ 0.11              Mean (s.d.)  0.19 (1.02)  0.45 (2.51)  0.47 (1.82)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)   Feet          Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.36              Mean (s.d.)  0.24 (0.89)  0.13 (0.48)  0.49 (1.78)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)          Week 104        TCZ + MTX vs MTX; P = 0.046              Mean (s.d.)  0.22 (0.63)  0.23 (0.67)  0.71 (1.38)  TCZ vs MTX; P = 0.022              Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–1.00)  TCZ + MTX vs TCZ; P = NS  Joint damage in hands/feet  Tocilizumab plus MTX (n = 106)  Tocilizumab (n = 103)  MTX (n = 108)  Comparative tests; P-value  JSN score   Hands          Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.16              Mean (s.d.)  0.04 (0.25)  0.17 (1.05)  0.08 (0.32)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)          Week 104        TCZ + MTX vs TCZ vs MTX; P ≥ 0.68              Mean (s.d.)  0.11 (0.48)  0.41 (1.74)  0.20 (0.81)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)   Feet          Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.12              Mean (s.d.)  0.06 (0.29)  0.07 (0.50)  0.16 (0.60)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)          Week 104        TCZ + MTX vs TCZ vs MTX; P ≥ 0.12              Mean (s.d.)  0.07 (0.35)  0.35 (1.33)  0.36 (1.24)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  Erosion score   Hands          Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.18              Mean (s.d.)  0.15 (0.96)  0.44 (2.25)  0.29 (1.34)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)          Week 104        TCZ + MTX vs TCZ vs MTX; P ≥ 0.11              Mean (s.d.)  0.19 (1.02)  0.45 (2.51)  0.47 (1.82)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)   Feet          Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.36              Mean (s.d.)  0.24 (0.89)  0.13 (0.48)  0.49 (1.78)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)          Week 104        TCZ + MTX vs MTX; P = 0.046              Mean (s.d.)  0.22 (0.63)  0.23 (0.67)  0.71 (1.38)  TCZ vs MTX; P = 0.022              Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–1.00)  TCZ + MTX vs TCZ; P = NS  Joint damage measured according to the Sharp–van der Heijde method. aData not normally distributed. JSN: joint space narrowing; NS: not significant; TCZ: tocilizumab. Fig. 2 View largeDownload slide Cumulative probability (%) plots of change from baseline in Sharp–van der Heijde score (A) Change from baseline to week 52; (B) change from baseline to week 104. Asterisks indicate scores beyond the range of the y-axes. SHS: Sharp–van der Heijde score; TCZ: tocilizumab. Fig. 2 View largeDownload slide Cumulative probability (%) plots of change from baseline in Sharp–van der Heijde score (A) Change from baseline to week 52; (B) change from baseline to week 104. Asterisks indicate scores beyond the range of the y-axes. SHS: Sharp–van der Heijde score; TCZ: tocilizumab. Fig. 3 View largeDownload slide Proportion (%) of patients without radiographic progression Change from baseline to weeks 52 and 104. *P < 0.05 when compared with the MTX arm. All scores according to the Sharp–van der Heijde method; Erosion: erosion score; JSN: joint space narrowing score; TCZ: tocilizumab. Fig. 3 View largeDownload slide Proportion (%) of patients without radiographic progression Change from baseline to weeks 52 and 104. *P < 0.05 when compared with the MTX arm. All scores according to the Sharp–van der Heijde method; Erosion: erosion score; JSN: joint space narrowing score; TCZ: tocilizumab. Joint damage in sustained drug-free remission It was previously reported that a significantly higher proportion of patients who initiated tocilizumab plus MTX (37/106, 35%) or tocilizumab (28/103, 27%) therapy achieved sDFR when compared with those initiating MTX (12/108, 11%) therapy (tocilizumab plus MTX vs MTX: P < 0.001; tocilizumab vs MTX: P = 0.004) [8]. When performing longitudinal analyses to evaluate the radiographic progression in those achieving sDFR vs those achieving sustained remission (i.e. not able to discontinue medication), significantly lower SHS (mean −0.75, 95% CI: −1.38, −1.11; P = 0.022) was found in patients achieving sDFR indicating less progression. These differences were also significant for JSN (mean −0.38, 95% CI: −0.67, 0.08; P = 0.012 vs sustained remission) but not for erosion (mean −0.44, 95% CI: −1.05, 0.17; P = 0.16 vs sustained remission) scores. No treatment-dependent differences of this effect were observed (P ⩾ 0.08). Association of radiographic progression with disease activity In the longitudinal analysis we found less progression in joint damage over time in the tocilizumab strategies (tocilizumab plus MTX: mean SHS −0.71, 95% CI: −0.33, −1.36; P = 0.030; tocilizumab: mean SHS −0.53, 95% CI: −1.18, 0.12; P = 0.12) compared with MTX strategy, corrected for baseline DAS28 and SHS and time (week of visit). After correcting also for DAS28 score over time, the effect of the tocilizumab strategies on progression of joint damage decreased slightly (tocilizumab plus MTX: mean SHS −0.55, 95% CI: −1.22, 0.11; P = 0.10 vs MTX; tocilizumab: mean SHS −0.39, 95% CI: −1.05, 0.28; P = 0.26 vs MTX) but was no longer statistically significant. As expected, we found that higher disease activity over time was significantly related to higher progression of joint damage (mean DAS28 0.29, 95% CI: 0.05, 0.52; P = 0.015). As tocilizumab is known to affect acute-phase proteins [16, 17], we repeated the analysis using the clinical disease activity index (CDAI) as disease activity measure, which is a composite score not containing inflammatory markers, and found similar results (tocilizumab plus MTX: mean CDAI −0.61, 95% CI: −0.01, 0.63; P = 0.09 vs MTX; tocilizumab: mean CDAI −0.43, 95% CI: −1.14, 0.28; P = 0.23 vs MTX) confirming the (non-statistically significant) effect of tocilizumab on joint damage. As significant differences were found between the strategy arms during follow-up in erosions but not in JSN, these analyses were also applied to both subscores separately. For erosions, the progression over time, corrected for disease activity, was lower (not significant) in the tocilizumab strategies (tocilizumab plus MTX: mean −0.67, 95% CI: −1.36, 0.02; P = 0.059 vs MTX; tocilizumab: mean −0.50, 95% CI: −1.18, 0.18; P = 0.15 vs MTX); for JSN, the disease activity corrected mean (95% CI) progression over time was in the tocilizumab plus MTX arm −0.27 (−0.61, 0.08) and in the tocilizumab arm 0.01 (−0.33, 0.35), both not statistically significant (P ⩾ 0.13) when compared with the MTX arm. Joint space quantification The change from baseline in JSW, as measured by joint space quantification, after weeks 52 and 104 in the PIP, MCP, wrist and hand joints is shown in Table 3. No significant differences were found between the treatment strategies in the PIP (P ⩾ 0.21), MCP (P ⩾ 0.18), wrist (P ⩾ 0.53) and hand (P ⩾ 0.21) joints during follow-up, confirming the JSN results in the hands of the SHS. Table 3 Change from baseline in millimetres in joint space width of the hand Joint space quantification  Tocilizumab plus MTX (n = 106)  Tocilizumab (n = 103)  MTX (n = 108)  Comparative tests; P-values  Hand      Week 52  −0.019 (0.21)  0.022 (0.24)  −0.014 (0.22)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.22      Week 104  −0.063 (0.29)  −0.027 (0.22)  −0.056 (0.27)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.21  Wrist      Week 52  −0.008 (0.17)  0.019 (0.20)  0.002 (0.19)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.56      Week 104  −0.045 (0.18)  −0.027 (0.18)  −0.029 (0.24)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.53  MCP joints      Week 52  −0.005 (0.10)  −0.008 (0.07)  −0.013 (0.06)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.20      Week 104  −0.015 (0.14)  −0.010 (0.07)  −0.016 (0.07)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.18  PIP joints      Week 52  0.000 (0.05)  0.009 (0.05)  0.000 (0.06)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.23      Week 104  0.000 (0.08)  0.005 (0.07)  −0.003 (0.07)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.21  Joint space quantification  Tocilizumab plus MTX (n = 106)  Tocilizumab (n = 103)  MTX (n = 108)  Comparative tests; P-values  Hand      Week 52  −0.019 (0.21)  0.022 (0.24)  −0.014 (0.22)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.22      Week 104  −0.063 (0.29)  −0.027 (0.22)  −0.056 (0.27)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.21  Wrist      Week 52  −0.008 (0.17)  0.019 (0.20)  0.002 (0.19)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.56      Week 104  −0.045 (0.18)  −0.027 (0.18)  −0.029 (0.24)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.53  MCP joints      Week 52  −0.005 (0.10)  −0.008 (0.07)  −0.013 (0.06)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.20      Week 104  −0.015 (0.14)  −0.010 (0.07)  −0.016 (0.07)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.18  PIP joints      Week 52  0.000 (0.05)  0.009 (0.05)  0.000 (0.06)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.23      Week 104  0.000 (0.08)  0.005 (0.07)  −0.003 (0.07)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.21  Values given as mean (s.d.). Joint width measured with automated joint space quantification. TCZ: tocilizumab. Discussion To our knowledge, U-Act-Early is the first randomized controlled trial reporting on the 2-year progression of radiographic joint damage in DMARD-naïve patients with very early RA treated to target with a tocilizumab-based strategy. Previously published results show that initiation of tocilizumab therapy, with or without MTX, resulted in significantly less joint damage, as measured by the SHS, when compared with initiating step-up MTX therapy, which is the current standard for starting DMARD treatment in RA patients [8]. As an extension to the previous results, this study demonstrates an inhibiting effect of treat-to-target tocilizumab therapy on the progression of erosions, especially in the feet, preventing the development of physical impairments and therefore improving the long-term clinical outcome in these patients. When, according to current standards, treatment in newly diagnosed RA patients is started with csDMARDs in a treat-to-target approach, aiming at remission [3, 4], subclinical inflammation is often seen in those who achieve the treatment target. MRI studies in RA patients clinically in remission on csDMARD therapy demonstrated that approximately one-third of them still showed radiographic progression, of whom 96% had signs of synovitis [5, 6]. In the present study, results (although not statistically significant) suggest a superior joint sparing effect of tocilizumab compared with that of MTX, even when corrected for disease activity, thus correcting for the faster and better clinical effect of tocilizumab, which is in line with the findings described above and with results of previous studies on other biologicals [9–12]. IL-6 is positively correlated with synovitis [16, 18, 19], but also causes deterioration of bone and cartilage via several other mechanisms. First, a higher level of IL-6 increases the expression of VEGF, which as a result of angiogenesis leads to pannus formation and eventually causes cartilage damage and bone erosions [20]. Second, IL-6 stimulates osteoclastogenesis by interaction with RANK-B ligand (RANKL) [17], and also by mechanisms independent of RANK [21], leading to bone destruction. Third, IL-6 simulates secretion of MMPs by activating synoviocytes and chondrocytes, which causes cartilage degradation [17]. Thus, via these mechanisms, the superior joint sparing effect of tocilizumab compared with that of MTX, even if corrected for disease activity, can be explained. In the present study, when evaluating the two subscores of the SHS separately, we found no significant differences in the progression of JSN, which we confirmed in the hand with a computerized JSW quantification method, and only the changes in erosions over 2 years were significantly different between the tocilizumab arms and the MTX arm. These results are largely in accordance with the findings of the FUNCTION study in which also more progression of erosions and JSN was observed in patients with early RA treated with MTX compared with those treated with tocilizumab plus MTX and tocilizumab. In the tocilizumab plus MTX group, erosions were signficantly less in that study at week 52 with JSN only numerically lower when compared with the MTX arm; these differences were not statistically significant after 2 years [22, 23]. As both the U-Act-Early and the FUNCTION study were not powered to analyse radiographic changes, which frequently are minimal especially in early RA patients treated to target, P-values should be interpreted with caution as lack of significance could well be due to insufficient statistical power (i.e. type II error). Nevertheless, also in established RA more joint sparing effect of tocilizumab has been demonstrated compared with that of MTX therapy supporting the effectiveness of this therapy in various patient populations [24]. As in our study erosions contributed most to the progression of joint damage measured by the SHS, we further analysed progression of erosive damage in the limbs separately and found significant differences only in the feet when comparing the tocilizumab arms with the MTX arm. Although this is the first tocilizumab trial evaluating limb specific structural joint changes, it has been shown previously that in RA patients treated with csDMARD, the long-term progression of erosions is higher in feet compared with hands or wrists [25]. An explanation of more rapid progression of erosions in the lower limb might be the different scoring ranges of the hands (range 0–5) and feet (range 0–10), that is, changes in individual joints of the feet have a greater impact on progression rates. However, the maximum theoretical score of feet is less than that of hands and wrists. Another explanation could be that joints in the feet are exposed to high weight-bearing pressures, while standing and walking, and are therefore more prone to develop structural changes when inflammation is present. In conclusion, this study demonstrates that initiating a treat-to-target tocilizumab therapy, with or without MTX, in DMARD-naïve patients with early RA is more effective in preventing progression of erosions compared with a step-up MTX strategy. Especially in the feet we found less erosion in the tocilizumab-based strategies showing that by rapidly reducing inflammation, the functional ability can be preserved in newly diagnosed patients. Acknowledgements In particular we thank Maal van Everdingen for scoring the radiographs. Furthermore, we gratefully thank all patients for their participation, and also those who were involved in conducting this study and all other personnel for their contribution, especially the rheumatologists, study-nurses and laboratory employees of the participating institutions: Amphia Ziekenhuis Breda, Breda, Netherlands; Antonius Ziekenhuis Sneek, Sneek, Netherlands; Atrium Medisch Centrum, Heerlen, Netherlands; Gemini Ziekenhuis, Den Helder, Netherlands; Gelre Ziekenhuis, Apeldoorn, Netherlands; Groene Hart Ziekenhuis, Gouda, Netherlands; Kennemer Gasthuis, Haarlem, Netherlands; Leids Universitair Medisch Centrum, Leiden, Netherlands; Meander Medisch Centrum, Amersfoort, Netherlands; Medisch Centrum Alkmaar, Alkmaar, Netherlands; Medisch Centrum Leeuwarden, Leeuwarden, Netherlands; Medisch Centrum Zuiderzee, Lelystad, Netherlands; Reade, Amsterdam, Netherlands; Rivas Zorggroep, Gorinchem, Netherlands; St. Antonius Ziekenhuis Nieuwegein, Nieuwegein, Netherlands; St Maartenskliniek Woerden, Woerden, Netherlands; St Maartenskliniek Nijmegen, Nijmegen, Netherlands; Spaarne Gasthuis, Hoofddorp, Netherlands; Tergooi Ziekenhuis, Hilversum, Netherlands; Universitair Medisch Centrum Groningen, Groningen, Netherlands; Universitair Medisch Centrum Utrecht, Utrecht, Netherlands. The department of the authors who included patients (J.W.G.J. and J.W.J.B.) in the UAct-Early trial received reimbursements from Roche Nederland BV. The U-Act-Early trial was funded by Roche Nederland BV. Funding: No specific funding was received from any bodies in the public, commercial or not-for-profit sectors to carry out the work described in this manuscript. Disclosure statement: A.P.-S. is an employee of F. Hoffmann-La Roche Ltd. M.E.A.B. is an employee of Roche Nederland BV. J.W.J.B. reported grants and fees from Roche, AbbVie, Bristol-Myers Squibb, Merck Sharp & Dohme, Pfizer, and UCB. F.P.J.G.L has received unrestricted grant support from Roche. J.M.v.L. has received honoraria from Arthrogen, BMS, Eli Lilly, Janssen, MSD, Pfizer, Roche, and research grants from Astra Zeneca, MSD, and Roche-Genentech. All other authors have declared no conflicts of interest. References 1 Carpenter L, Nikiphorou E, Sharpe R et al.   Have radiographic progression rates in early rheumatoid arthritis changed? A systematic review and meta-analysis of long-term cohorts. Rheumatology  2016; 55: 1053– 65. Google Scholar CrossRef Search ADS   2 Balanescu A, Wiland P. Maximizing early treatment with biologics in patients with rheumatoid arthritis: the ultimate breakthrough in joints preservation. Rheumatol Int  2013; 33: 1379– 86. Google Scholar CrossRef Search ADS PubMed  3 Smolen JS, Landewe R, Bijlsma J et al.   EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2016 update. Ann Rheum Dis  2017; 76: 960– 77. Google Scholar CrossRef Search ADS PubMed  4 Singh JA, Saag KG, Bridges SLJr et al.   2015 American College of Rheumatology Guideline for the Treatment of Rheumatoid Arthritis. Arthritis Rheumatol  2016; 68: 1– 26. Google Scholar PubMed  5 Aletaha D, Smolen JS. Joint damage in rheumatoid arthritis progresses in remission according to the Disease Activity Score in 28 joints and is driven by residual swollen joints. Arthritis Rheum  2011; 63: 3702– 11. Google Scholar CrossRef Search ADS PubMed  6 Gandjbakhch F, Conaghan PG, Ejbjerg B et al.   Synovitis and osteitis are very frequent in rheumatoid arthritis clinical remission: results from an MRI study of 294 patients in clinical remission or low disease activity state. J Rheumatol  2011; 38: 2039– 44. Google Scholar CrossRef Search ADS PubMed  7 van der Heijde D. How to read radiographs according to the Sharp/van der Heijde method. J Rheumatol  2000; 27: 261– 3. Google Scholar PubMed  8 Bijlsma JW, Welsing PM, Woodworth TG et al.   Early rheumatoid arthritis treated with tocilizumab, methotrexate, or their combination (U-Act-Early): a multicentre, randomised, double-blind, double-dummy, strategy trial. Lancet  2016; 388: 343– 55. Google Scholar CrossRef Search ADS PubMed  9 Klarenbeek NB, Guler-Yuksel M, van der Heijde DM et al.   Clinical synovitis in a particular joint is associated with progression of erosions and joint space narrowing in that same joint, but not in patients initially treated with infliximab. Ann Rheum Dis  2010; 69: 2107– 13. Google Scholar CrossRef Search ADS PubMed  10 Landewe R, van der Heijde D, Klareskog L, van Vollenhoven R, Fatenejad S. Disconnect between inflammation and joint destruction after treatment with etanercept plus methotrexate: results from the trial of etanercept and methotrexate with radiographic and patient outcomes. Arthritis Rheum  2006; 54: 3119– 25. Google Scholar CrossRef Search ADS PubMed  11 Smolen JS, Han C, Bala M et al.   Evidence of radiographic benefit of treatment with infliximab plus methotrexate in rheumatoid arthritis patients who had no clinical improvement: a detailed subanalysis of data from the anti-tumor necrosis factor trial in rheumatoid arthritis with concomitant therapy study. Arthritis Rheum  2005; 52: 1020– 30. Google Scholar CrossRef Search ADS PubMed  12 Saleem B, Brown AK, Keen H et al.   Disease remission state in patients treated with the combination of tumor necrosis factor blockade and methotrexate or with disease-modifying antirheumatic drugs: A clinical and imaging comparative study. Arthritis Rheum  2009; 60: 1915– 22. Google Scholar CrossRef Search ADS PubMed  13 Huo Y, Vincken KL, van der Heijde D et al.   Automatic quantification of radiographic finger joint space width of patients with early rheumatoid arthritis. IEEE Trans Biomed Eng  2016; 63: 2177– 86. Google Scholar CrossRef Search ADS PubMed  14 Huo Y, Vincken KI, van der Heijde D et al.   Automatic quantification of radiographic wrist joint space width of patients with rheumatoid arthritis. IEEE Trans Biomed Eng  2017; Advance Access published 26 January 2017, doi: 10.1109/TBME.2017.2659223. 15 Huo Y, Veldhuizen RD, van der Heijde D et al.   Automated joint space width quantification of hand and wrist joints: a proof of concept study. Clin Exp Rheumatol  2016; 34(Suppl. 101): S34– S9. 16 Dayer JM, Choy E. Therapeutic targets in rheumatoid arthritis: the interleukin-6 receptor. Rheumatology  2010; 49: 15– 24. Google Scholar CrossRef Search ADS PubMed  17 Choy E. Understanding the dynamics: pathways involved in the pathogenesis of rheumatoid arthritis. Rheumatology  2012; 51(Suppl 5): v3– 11. Google Scholar CrossRef Search ADS PubMed  18 Madhok R, Crilly A, Watson J, Capell HA. Serum interleukin 6 levels in rheumatoid arthritis: correlations with clinical and laboratory indices of disease activity. Ann Rheum Dis  1993; 52: 232– 4. Google Scholar CrossRef Search ADS PubMed  19 Sack U, Kinne RW, Marx T et al.   Interleukin-6 in synovial fluid is closely associated with chronic synovitis in rheumatoid arthritis. Rheumatol Int  1993; 13: 45– 51. Google Scholar CrossRef Search ADS PubMed  20 Ballara S, Taylor PC, Reusch P et al.   Raised serum vascular endothelial growth factor levels are associated with destructive change in inflammatory arthritis. Arthritis Rheum  2001; 44: 2055– 64. Google Scholar CrossRef Search ADS PubMed  21 O'Brien W, Fissel BM, Maeda Y et al.   RANK-independent osteoclast formation and bone erosion in inflammatory arthritis. Arthritis Rheumatol  2016; 68: 2889– 900. Google Scholar CrossRef Search ADS PubMed  22 Burmester GR, Rigby WF, van Vollenhoven RF et al.   Tocilizumab in early progressive rheumatoid arthritis: FUNCTION, a randomised controlled trial. Ann Rheum Dis  2016; 75: 1081– 91. Google Scholar CrossRef Search ADS PubMed  23 Burmester GR, Rigby WF, van Vollenhoven RF et al.   Tocilizumab combination therapy or monotherapy or methotrexate monotherapy in methotrexate-naive patients with early rheumatoid arthritis: 2-year clinical and radiographic results from the randomised, placebo-controlled FUNCTION trial. Ann Rheum Dis  2017; 76: 1279– 84. Google Scholar CrossRef Search ADS PubMed  24 Smolen JS, Avila JC, Aletaha D. Tocilizumab inhibits progression of joint damage in rheumatoid arthritis irrespective of its anti-inflammatory effects: disassociation of the link between inflammation and destruction. Ann Rheum Dis  2012; 71: 687– 93. Google Scholar CrossRef Search ADS PubMed  25 Hulsmans HM, Jacobs JW, van der Heijde DM et al.   The course of radiologic damage during the first six years of rheumatoid arthritis. Arthritis Rheum  2000; 43: 1927– 40. Google Scholar CrossRef Search ADS PubMed  © The Author 2017. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Rheumatology Oxford University Press

Radiographic joint damage in early rheumatoid arthritis patients: comparing tocilizumab- and methotrexate-based treat-to-target strategies

Loading next page...
 
/lp/ou_press/radiographic-joint-damage-in-early-rheumatoid-arthritis-patients-H3h95Ugekr
Publisher
Oxford University Press
Copyright
© The Author 2017. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oup.com
ISSN
1462-0324
eISSN
1462-0332
D.O.I.
10.1093/rheumatology/kex386
Publisher site
See Article on Publisher Site

Abstract

Abstract Objective To evaluate the progression of erosions and joint space narrowing (JSN) in feet and hands in the U-Act-Early trial. Methods In this trial, 317 newly diagnosed DMARD-naïve RA patients initiated randomly tocilizumab, or step-up MTX or a combination of the two. Radiographs were scored at baseline and after 52 and 104 weeks using the Sharp–van der Heijde erosion and JSN score. Between the strategy arms, changes from baseline and the proportions of patients without radiographic progression (change from baseline ≤0) were compared. Results Mean changes from baseline in erosion and JSN scores for the whole study population were after 52 weeks 0.59 and 0.18 and after 104 weeks 0.70 and 0.50, respectively. For JSN, at both time points no differences in progression were found between strategies (P ⩾ 0.09). For erosions, the progression was significantly lower at week 104 in both tocilizumab arms when compared with the MTX arm ((p≤0.023). Less progression of erosions in the feet was found after 104 weeks in both tocilizumab arms (P ⩽ 0.046); this was not significant for the hands (P ⩾ 0.11). The proportion of patients without progression in erosions was higher in the tocilizumab arms at week 52 (tocilizumab plus MTX: 87%, P = 0.038; tocilizumab: 81%, P = 0.29) and 104 (tocilizumab plus MTX: 85%, P = 0.001; tocilizumab: 77%, P = 0.028), compared with the MTX arm (74 and 60%, respectively). Conclusion In DMARD-naïve early RA patients, initiating a tocilizumab-based treat-to-target strategy inhibits the progression of erosions, especially in the feet, more compared with initiation of a step-up MTX strategy. Trial registration ClinicalTrials.gov, https://clinicaltrials.gov, NCT01034137. rheumatoid arthritis, tocilizumab, methotrexate, radiographic joint damage, Sharp/van der Heijde score Rheumatology key messages Fewer erosions were found in the tocilizumab-based strategies when compared with the methotrexate-based strategy. Especially in the feet there was less progression of erosions in RA patients initiating tocilizumab. No differences were found between strategies in joint space narrowing of the hands and feet. Introduction In RA, structural joint damage as a consequence of synovitis is commonly seen [1]. Typically bones are affected, resulting in erosions and loss of cartilage, leading to joint space narrowing (JSN), which eventually causes joint deformities and functional impairments if joint inflammation is not targeted adequately. With the development of conventional synthetic (csDMARDs) and biologic DMARDs (bDMARDs) and the implementation of treat-to-target approaches, aiming for clinical remission, a remarkable gain is made in preventing joint damage and therefore in improving the long-term outcome of RA patients [2]. As smaller joints are more frequently affected in early RA, radiographs of both hands and feet are the current standard to assess structural changes, and radiography is therefore considered to be a pivotal tool for assessing therapeutic efficacy. International recommendations/guidelines for the management of RA recommend that a csDMARD, preferably MTX, should be (part of) the first treatment strategy and that, if the treatment target is not achieved, adding a bDMARD may be considered [3, 4]. To date, MTX remains the anchor drug in treating newly diagnosed RA because of its beneficial efficacy–safety and cost-effectiveness ratios [3]. Nevertheless, in about 30% of the patients achieving the treatment target on this drug, progression of radiographic changes is still observed as a result of subclinical inflammation [5, 6]. Starting therapy with a bDMARD, with or without MTX, may improve the clinical outcome by more effectively reducing disease activity. Tapering and finally discontinuing the biological once remission is achieved might minimize health care costs and the number of patients who are over-treated, as some would have achieved the treatment target also with MTX therapy alone. In the 2-year, multicentre, phase 3, double-blind, placebo-controlled U-Act-Early strategy trial (ClinicalTrials.gov identifier NCT01034137), DMARD-naïve patients with early RA were randomized to start treatment with tocilizumab, step-up MTX or a combination of the two and were treated to target until sustained remission (defined as DAS28 <2.6 with four or fewer swollen joints for ⩾24 weeks) was achieved. Clinical efficacy and safety data have been reported previously in which significantly less joint damage, as assessed by the Sharp–van der Heijde score (SHS) [7], was found in both tocilizumab arms when compared with the MTX arm [8]. This article reports in more detail the effect of the treatment strategies on the progression of erosions and JSN separately, according to the SHS method, in the feet and hands. In addition, we evaluated if there was an association between progression of joint damage and the three therapeutic strategies and whether this association was in part independent from suppression of disease activity, as reported in literature for other biologicals [9–12]. Methods Study design and population Patients with newly diagnosed RA visiting one of the 21 participating rheumatology departments in the Netherlands were invited to participate in the U-Act-Early strategy trial. To be eligible, patients had to be >18 years, be DMARD-naïve, meet the 1987 ACR or the 2010 ACR/EULAR classification criteria for RA, have active disease (DAS28 ⩾2.6), and have been diagnosed with RA within the previous 12 months before inclusion. Patients were randomized (1:1:1) in permuted blocks of six, stratified by centre and baseline DAS28 (i.e. DAS28 <5.1 or ⩾5.1) to initiate tocilizumab plus MTX, or tocilizumab plus placebo-MTX, or MTX plus placebo-tocilizumab therapy and were treated to target until sustained remission was achieved. Placebo/tocilizumab (8 mg/kg) was given every 4 weeks intravenously up to a maximum of 800 mg per dose. MTX (orally) was started at 10 mg/week and increased stepwise with 5 mg every 4 weeks to 30 mg/week until remission or the maximum tolerable dose was achieved; the same scheme was used for placebo-MTX. If remission was not achieved with the initial therapy, HCQ was added as a second step in the initial treatment strategies. Hereafter, if the target still was not reached, the initial strategies ended and the subsequent treatment strategy was started. Patients randomized to the tocilizumab or MTX arm then discontinued the placebo and switched to tocilizumab plus MTX combination therapy; those who initially started with the combination switched to the standard of care (e.g. MTX combined with a TNF inhibitor). If the treatment target, sustained remission, was achieved, medication was tapered stepwise and finally stopped. First placebo/MTX was tapered with 5 mg every 4 weeks down to 10 mg/week and then discontinued; thereafter the dose of placebo/tocilizumab was decreased to 4 mg/kg and stopped after 3 months. Sustained drug-free remission (sDFR) was achieved when patients were drug-free for ⩾12 weeks while in sustained remission (one visit with low disease activity (DAS28 ⩽3.2) was allowed). A detailed description of the study design, including treatment strategies, has been previously reported [8]. The medical ethics research committee of the University Medical Centre Utrecht approved the study for all participating hospitals. All patients signed informed consent before entering the study. Sharp–van der Heijde score The SHS was used to evaluate progression of radiographic joint damage. Radiographs of the hands (including wrists) and feet were taken at baseline and after 52 and 104 weeks. Sixteen and six areas were considered for assessing erosions and JSN in both hands and feet, respectively. The maximum erosion score of hands and wrists is 160 and of the feet it is 120 (maximum total erosion score 280). The maximum JSN score of hands and wrists is 120 and of the feet it is 48 (maximum total JSN score 168). The sum of the erosion and JSN scores is the total SHS (maximum 448). All radiographs were scored centrally in chronological order by a blinded (experienced and trained) professional reader. Automated joint space width quantification To assess the JSN of the hands with a more sensitive method, we used an automated method for measuring joint space width (JSW) as previously described by Huo et al. [13, 14]. Briefly, this computerized joint space quantification method measures in millimetres the JSW of PIP joints, MCP joints and joints of the wrist, consisting of the trapezium-trapezoides-navicular, carpo-navicular-lunate and the radiocarpal joints, bilaterally. After the joint margins were automatically delineated, the average Euclidean distance between the upper and lower margin was calculated within the middle 60% of the joint span and the JSW was determined for each joint separately. The average JSW scores of all the joints (i.e. PIP, MCP and wrist joints) of both extremities were used to calculate the total JSW hand score. The automated joint space quantification method has been proven effective in measuring radiographic changes in the hands within DMARD-naïve patients with early RA [15]; JSW quantification of the foot joints with this method has to date not yet been developed. Statistical analyses Analyses were performed based on the intention-to-treat population: those who were randomized and received at least one dose of medication. Radiographic joint scores are described as mean (s.d.) and median (interquartile range, IQR). For between-group comparisons at follow-up, we calculated changes from baseline in total SHS, including erosion and JSN scores and both subscores separately in the hands and feet, and analysed these using a two-sided stratified non-parametric Mann–Whitney (Van Elteren) test, controlling for the two randomization stratification factors, baseline DAS28 (i.e. DAS28 <5.1 or ⩾5.1) and centre. The same analyses were performed for JSN scores of the hand as measured by the computerized JSW quantification method. In addition, the proportions of patients without radiographic progression (change from baseline ⩽0) were analysed using a Pearson chi-square test. To evaluate if the progression in radiographic joint damage within the strategy arms over time was associated with disease activity (as expected) and whether an effect on progression of joint damage of the treatment strategies remained after correction for disease activity, we performed longitudinal regression analysis (mixed model analysis using a random intercept). The dependent variable in this analysis was the delta (Δ) SHS over time (i.e. change from baseline to week 52 and change from week 52 to 104) and baseline DAS28, baseline SHS and treatment strategy were used as time-independent covariates and time (week after baseline) and the average disease activity (mean DAS28) during a progression period as time-varying covariates. These longitudinal regression analyses were also performed for the JSN and erosion scores separately. Furthermore, to evaluate the effect of tapering and discontinuing therapy as defined in the design of this study, we compared the change in SHS (as well as for erosion and JSN) in longitudinal regression analyses between patients achieving sDFR vs those achieving the treatment target, sustained remission (i.e. not able to discontinue medication). Missing radiograph scores were imputed using data from the withdrawal visit with a maximum time window of 180 days. All analyses were performed based on the entire treatment strategy, thus including the addition of HCQ and switching from placebo to active therapy (in the tocilizumab and MTX arms). Patients switching to the standard of care according to protocol were no longer followed within the study. A P < 0.05 (two tailed) was considered statistically significant; R version 3.4.0 was used for data analyses. Results Patient population Of the recruited patients, 317 were eligible to participate in the U-Act-Early study (tocilizumab plus MTX arm, 106; tocilizumab arm, 103; MTX arm, 108); only minor differences in the baseline characteristics between the three strategy arms were noted [8]. The majority of the study population was female (67%), was seropositive for RF (72%) or anti-CCP (70%), had on average high disease activity [mean (s.d.) DAS28 5.2 (1.1) at baseline with a median (IQR) symptom duration of 26 (16–43) days], and had none or minor radiographic joint damage [median (IQR) SHS 0 (0–1)]. In total, radiographs were obtained of 97 (92%), 92 (89%) and 91 (84%) patients at week 52 and of 76 (72%), 75 (73%) and 75 (69%) patients at week 104 in the tocilizumab plus MTX, tocilizumab and MTX arms, respectively. Patient disposition After 52 weeks of treatment, 95% of the patients in the tocilizumab plus MTX- and 92% in the tocilizumab-based treatment strategies were still on the initial treatment regimen; in the MTX arm this proportion was 61% (Fig. 1). The proportions receiving or having received HCQ in addition, as part of the initial treatment strategy, during the first 52 weeks of the study were 30% in the tocilizumab plus MTX arm, 21% in the tocilizumab arm and 62% in the MTX arm. During the second year, the proportions of patients within the treatment strategies who received or had received HCQ further increased only marginally (tocilizumab plus MTX, 37%; tocilizumab, 22%; MTX, 66%). After 104 weeks, the large majority of the patients in the tocilizumab-based strategies still remained on the initial treatment regimen (tocilizumab plus MTX, 92%; tocilizumab, 87%), in contrast to the MTX strategy arm (54%). Fig. 1 View largeDownload slide Proportion (%) of patients remaining on the initial regimen during the entire study TCZ: tocilizumab. Fig. 1 View largeDownload slide Proportion (%) of patients remaining on the initial regimen during the entire study TCZ: tocilizumab. Radiographic joint damage The cumulative frequencies of changes from baseline to weeks 52 and 104 in the SHS are shown in Fig. 2. In the previously published results of U-Act-Early it was reported that changes from baseline in radiographic joint damage, as evaluated by the SHS, were significantly lower in the tocilizumab plus MTX arm compared with the MTX arm after 52 weeks (P = 0.016; Table 1) [8]. After 104 weeks, less progression in joint damage was noted in both tocilizumab arms (tocilizumab plus MTX, P = 0.021; tocilizumab, P = 0.038) when compared with the MTX arm. In the present study, when evaluating the changes of the JSN and erosion scores separately, no significant differences were found in the JSN scores at week 52 (P ⩾ 0.09) or 104 (P ⩾ 0.25) between the three treatment arms. For erosions, the changes from baseline to week 52 did not statistically differ between the treatment arms (P ⩾ 0.20), but after 104 weeks, significantly lower erosion progression scores were found in the tocilizumab plus MTX (P = 0.016) and tocilizumab arm (P = 0.023) when compared with the MTX arm. Figure 3 shows proportions of patients without radiographic progression (change from baseline ⩽0). In those assigned to initiate tocilizumab plus MTX, the proportions of patients without progression in SHS were significantly higher at weeks 52 (84% vs 67%, P = 0.009) and 104 (74% vs 52%, P = 0.006), compared with those assigned to the MTX strategy; for the tocilizumab strategy, the comparisons with the MTX strategy were not statistically significant (77%, P = 0.13 and 65%, P = 0.10, respectively). The proportion of patients without progression in erosions was higher in the tocilizumab-based strategies at week 52 (tocilizumab plus MTX: 87%, P = 0.038; tocilizumab: 81%, P = 0.29) and at week 104 (tocilizumab plus MTX: 85%, P < 0.001; tocilizumab: 77%, P = 0.028), compared with the MTX strategy (74% and 60%, respectively). No significant differences were found between the strategies in the proportion of patients without progression in JSN (P ⩾ 0.22). When analysing the changes from baseline in JSN and erosion scores of both the hands and feet separately, no significant differences were found between the strategies in both extremities in JSN scores during the trial period (P ⩾ 0.12; Table 2). Also, no significant differences were noted in the erosion score of hands during follow-up (P ⩾ 0.11). After 104 weeks, significantly less progression of erosions was found in the feet in both tocilizumab arms (tocilizumab plus MTX, P = 0.046; tocilizumab, P = 0.022) compared with in the MTX arm; this was not significant at week 52 (P ⩾ 0.36). Table 1 Change from baseline in radiographic joint damagea Joint damage  Tocilizumab plus MTX (n = 106)  Tocilizumab (n = 103)  MTX (n = 108)  Comparative tests; P-value  Total SHSb      Week 52        TCZ + MTX vs MTX; P = 0.016          Mean (s.d.)  0.50 (1.50)  0.79 (3.24)  0.96 (2.87)  TCZ vs MTX; P = NS          Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  TCZ + MTX vs TCZ; P = NS      Week 104        TCZ + MTX vs MTX; P = 0.021          Mean (s.d.)  1.18 (3.92)  1.45 (4.27)  1.53 (2.42)  TCZ vs MTX; P = 0.038          Median (IQR)  0.00 (0.00–1.00)  0.00 (0.00–2.00)  0.00 (0.00–2.56)  TCZ + MTX vs TCZ; P = NS  JSN score      Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.20          Mean (s.d.)  0.10 (0.47)  0.24 (1.19)  0.21 (0.73)            Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)        Week 104        TCZ + MTX vs TCZ vs MTX; P ≥ 0.25          Mean (s.d.)  0.18 (0.72)  0.77 (2.41)  0.53 (1.58)            Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  Erosion score      Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.20          Mean (s.d.)  0.40 (1.35)  0.57 (2.38)/  0.79 (3.00)            Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–1.00)      Week 104        TCZ + MTX vs MTX; P = 0.016          Mean (s.d.)  0.41 (1.25)  0.67 (2.58)/  1.01 (1.68)  TCZ vs MTX; P = 0.023          Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–1.00)  TCZ+MTX vs TCZ; P = NS  Joint damage  Tocilizumab plus MTX (n = 106)  Tocilizumab (n = 103)  MTX (n = 108)  Comparative tests; P-value  Total SHSb      Week 52        TCZ + MTX vs MTX; P = 0.016          Mean (s.d.)  0.50 (1.50)  0.79 (3.24)  0.96 (2.87)  TCZ vs MTX; P = NS          Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  TCZ + MTX vs TCZ; P = NS      Week 104        TCZ + MTX vs MTX; P = 0.021          Mean (s.d.)  1.18 (3.92)  1.45 (4.27)  1.53 (2.42)  TCZ vs MTX; P = 0.038          Median (IQR)  0.00 (0.00–1.00)  0.00 (0.00–2.00)  0.00 (0.00–2.56)  TCZ + MTX vs TCZ; P = NS  JSN score      Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.20          Mean (s.d.)  0.10 (0.47)  0.24 (1.19)  0.21 (0.73)            Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)        Week 104        TCZ + MTX vs TCZ vs MTX; P ≥ 0.25          Mean (s.d.)  0.18 (0.72)  0.77 (2.41)  0.53 (1.58)            Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  Erosion score      Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.20          Mean (s.d.)  0.40 (1.35)  0.57 (2.38)/  0.79 (3.00)            Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–1.00)      Week 104        TCZ + MTX vs MTX; P = 0.016          Mean (s.d.)  0.41 (1.25)  0.67 (2.58)/  1.01 (1.68)  TCZ vs MTX; P = 0.023          Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–1.00)  TCZ+MTX vs TCZ; P = NS  Joint damage assessed according to the Sharp–van der Heijde method. aData not normally distributed. bPreviously published data. JSN: joint space narrowing; NS: not significant; SHS: Sharp–van der Heijde score; TCZ: tocilizumab. Table 2 Change from baseline in erosion and joint space narrowinga Joint damage in hands/feet  Tocilizumab plus MTX (n = 106)  Tocilizumab (n = 103)  MTX (n = 108)  Comparative tests; P-value  JSN score   Hands          Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.16              Mean (s.d.)  0.04 (0.25)  0.17 (1.05)  0.08 (0.32)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)          Week 104        TCZ + MTX vs TCZ vs MTX; P ≥ 0.68              Mean (s.d.)  0.11 (0.48)  0.41 (1.74)  0.20 (0.81)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)   Feet          Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.12              Mean (s.d.)  0.06 (0.29)  0.07 (0.50)  0.16 (0.60)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)          Week 104        TCZ + MTX vs TCZ vs MTX; P ≥ 0.12              Mean (s.d.)  0.07 (0.35)  0.35 (1.33)  0.36 (1.24)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  Erosion score   Hands          Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.18              Mean (s.d.)  0.15 (0.96)  0.44 (2.25)  0.29 (1.34)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)          Week 104        TCZ + MTX vs TCZ vs MTX; P ≥ 0.11              Mean (s.d.)  0.19 (1.02)  0.45 (2.51)  0.47 (1.82)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)   Feet          Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.36              Mean (s.d.)  0.24 (0.89)  0.13 (0.48)  0.49 (1.78)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)          Week 104        TCZ + MTX vs MTX; P = 0.046              Mean (s.d.)  0.22 (0.63)  0.23 (0.67)  0.71 (1.38)  TCZ vs MTX; P = 0.022              Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–1.00)  TCZ + MTX vs TCZ; P = NS  Joint damage in hands/feet  Tocilizumab plus MTX (n = 106)  Tocilizumab (n = 103)  MTX (n = 108)  Comparative tests; P-value  JSN score   Hands          Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.16              Mean (s.d.)  0.04 (0.25)  0.17 (1.05)  0.08 (0.32)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)          Week 104        TCZ + MTX vs TCZ vs MTX; P ≥ 0.68              Mean (s.d.)  0.11 (0.48)  0.41 (1.74)  0.20 (0.81)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)   Feet          Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.12              Mean (s.d.)  0.06 (0.29)  0.07 (0.50)  0.16 (0.60)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)          Week 104        TCZ + MTX vs TCZ vs MTX; P ≥ 0.12              Mean (s.d.)  0.07 (0.35)  0.35 (1.33)  0.36 (1.24)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  Erosion score   Hands          Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.18              Mean (s.d.)  0.15 (0.96)  0.44 (2.25)  0.29 (1.34)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)          Week 104        TCZ + MTX vs TCZ vs MTX; P ≥ 0.11              Mean (s.d.)  0.19 (1.02)  0.45 (2.51)  0.47 (1.82)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)   Feet          Week 52        TCZ + MTX vs TCZ vs MTX; P ≥ 0.36              Mean (s.d.)  0.24 (0.89)  0.13 (0.48)  0.49 (1.78)                Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–0.00)          Week 104        TCZ + MTX vs MTX; P = 0.046              Mean (s.d.)  0.22 (0.63)  0.23 (0.67)  0.71 (1.38)  TCZ vs MTX; P = 0.022              Median (IQR)  0.00 (0.00–0.00)  0.00 (0.00–0.00)  0.00 (0.00–1.00)  TCZ + MTX vs TCZ; P = NS  Joint damage measured according to the Sharp–van der Heijde method. aData not normally distributed. JSN: joint space narrowing; NS: not significant; TCZ: tocilizumab. Fig. 2 View largeDownload slide Cumulative probability (%) plots of change from baseline in Sharp–van der Heijde score (A) Change from baseline to week 52; (B) change from baseline to week 104. Asterisks indicate scores beyond the range of the y-axes. SHS: Sharp–van der Heijde score; TCZ: tocilizumab. Fig. 2 View largeDownload slide Cumulative probability (%) plots of change from baseline in Sharp–van der Heijde score (A) Change from baseline to week 52; (B) change from baseline to week 104. Asterisks indicate scores beyond the range of the y-axes. SHS: Sharp–van der Heijde score; TCZ: tocilizumab. Fig. 3 View largeDownload slide Proportion (%) of patients without radiographic progression Change from baseline to weeks 52 and 104. *P < 0.05 when compared with the MTX arm. All scores according to the Sharp–van der Heijde method; Erosion: erosion score; JSN: joint space narrowing score; TCZ: tocilizumab. Fig. 3 View largeDownload slide Proportion (%) of patients without radiographic progression Change from baseline to weeks 52 and 104. *P < 0.05 when compared with the MTX arm. All scores according to the Sharp–van der Heijde method; Erosion: erosion score; JSN: joint space narrowing score; TCZ: tocilizumab. Joint damage in sustained drug-free remission It was previously reported that a significantly higher proportion of patients who initiated tocilizumab plus MTX (37/106, 35%) or tocilizumab (28/103, 27%) therapy achieved sDFR when compared with those initiating MTX (12/108, 11%) therapy (tocilizumab plus MTX vs MTX: P < 0.001; tocilizumab vs MTX: P = 0.004) [8]. When performing longitudinal analyses to evaluate the radiographic progression in those achieving sDFR vs those achieving sustained remission (i.e. not able to discontinue medication), significantly lower SHS (mean −0.75, 95% CI: −1.38, −1.11; P = 0.022) was found in patients achieving sDFR indicating less progression. These differences were also significant for JSN (mean −0.38, 95% CI: −0.67, 0.08; P = 0.012 vs sustained remission) but not for erosion (mean −0.44, 95% CI: −1.05, 0.17; P = 0.16 vs sustained remission) scores. No treatment-dependent differences of this effect were observed (P ⩾ 0.08). Association of radiographic progression with disease activity In the longitudinal analysis we found less progression in joint damage over time in the tocilizumab strategies (tocilizumab plus MTX: mean SHS −0.71, 95% CI: −0.33, −1.36; P = 0.030; tocilizumab: mean SHS −0.53, 95% CI: −1.18, 0.12; P = 0.12) compared with MTX strategy, corrected for baseline DAS28 and SHS and time (week of visit). After correcting also for DAS28 score over time, the effect of the tocilizumab strategies on progression of joint damage decreased slightly (tocilizumab plus MTX: mean SHS −0.55, 95% CI: −1.22, 0.11; P = 0.10 vs MTX; tocilizumab: mean SHS −0.39, 95% CI: −1.05, 0.28; P = 0.26 vs MTX) but was no longer statistically significant. As expected, we found that higher disease activity over time was significantly related to higher progression of joint damage (mean DAS28 0.29, 95% CI: 0.05, 0.52; P = 0.015). As tocilizumab is known to affect acute-phase proteins [16, 17], we repeated the analysis using the clinical disease activity index (CDAI) as disease activity measure, which is a composite score not containing inflammatory markers, and found similar results (tocilizumab plus MTX: mean CDAI −0.61, 95% CI: −0.01, 0.63; P = 0.09 vs MTX; tocilizumab: mean CDAI −0.43, 95% CI: −1.14, 0.28; P = 0.23 vs MTX) confirming the (non-statistically significant) effect of tocilizumab on joint damage. As significant differences were found between the strategy arms during follow-up in erosions but not in JSN, these analyses were also applied to both subscores separately. For erosions, the progression over time, corrected for disease activity, was lower (not significant) in the tocilizumab strategies (tocilizumab plus MTX: mean −0.67, 95% CI: −1.36, 0.02; P = 0.059 vs MTX; tocilizumab: mean −0.50, 95% CI: −1.18, 0.18; P = 0.15 vs MTX); for JSN, the disease activity corrected mean (95% CI) progression over time was in the tocilizumab plus MTX arm −0.27 (−0.61, 0.08) and in the tocilizumab arm 0.01 (−0.33, 0.35), both not statistically significant (P ⩾ 0.13) when compared with the MTX arm. Joint space quantification The change from baseline in JSW, as measured by joint space quantification, after weeks 52 and 104 in the PIP, MCP, wrist and hand joints is shown in Table 3. No significant differences were found between the treatment strategies in the PIP (P ⩾ 0.21), MCP (P ⩾ 0.18), wrist (P ⩾ 0.53) and hand (P ⩾ 0.21) joints during follow-up, confirming the JSN results in the hands of the SHS. Table 3 Change from baseline in millimetres in joint space width of the hand Joint space quantification  Tocilizumab plus MTX (n = 106)  Tocilizumab (n = 103)  MTX (n = 108)  Comparative tests; P-values  Hand      Week 52  −0.019 (0.21)  0.022 (0.24)  −0.014 (0.22)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.22      Week 104  −0.063 (0.29)  −0.027 (0.22)  −0.056 (0.27)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.21  Wrist      Week 52  −0.008 (0.17)  0.019 (0.20)  0.002 (0.19)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.56      Week 104  −0.045 (0.18)  −0.027 (0.18)  −0.029 (0.24)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.53  MCP joints      Week 52  −0.005 (0.10)  −0.008 (0.07)  −0.013 (0.06)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.20      Week 104  −0.015 (0.14)  −0.010 (0.07)  −0.016 (0.07)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.18  PIP joints      Week 52  0.000 (0.05)  0.009 (0.05)  0.000 (0.06)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.23      Week 104  0.000 (0.08)  0.005 (0.07)  −0.003 (0.07)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.21  Joint space quantification  Tocilizumab plus MTX (n = 106)  Tocilizumab (n = 103)  MTX (n = 108)  Comparative tests; P-values  Hand      Week 52  −0.019 (0.21)  0.022 (0.24)  −0.014 (0.22)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.22      Week 104  −0.063 (0.29)  −0.027 (0.22)  −0.056 (0.27)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.21  Wrist      Week 52  −0.008 (0.17)  0.019 (0.20)  0.002 (0.19)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.56      Week 104  −0.045 (0.18)  −0.027 (0.18)  −0.029 (0.24)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.53  MCP joints      Week 52  −0.005 (0.10)  −0.008 (0.07)  −0.013 (0.06)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.20      Week 104  −0.015 (0.14)  −0.010 (0.07)  −0.016 (0.07)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.18  PIP joints      Week 52  0.000 (0.05)  0.009 (0.05)  0.000 (0.06)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.23      Week 104  0.000 (0.08)  0.005 (0.07)  −0.003 (0.07)  TCZ + MTX vs TCZ vs MTX; P ≥ 0.21  Values given as mean (s.d.). Joint width measured with automated joint space quantification. TCZ: tocilizumab. Discussion To our knowledge, U-Act-Early is the first randomized controlled trial reporting on the 2-year progression of radiographic joint damage in DMARD-naïve patients with very early RA treated to target with a tocilizumab-based strategy. Previously published results show that initiation of tocilizumab therapy, with or without MTX, resulted in significantly less joint damage, as measured by the SHS, when compared with initiating step-up MTX therapy, which is the current standard for starting DMARD treatment in RA patients [8]. As an extension to the previous results, this study demonstrates an inhibiting effect of treat-to-target tocilizumab therapy on the progression of erosions, especially in the feet, preventing the development of physical impairments and therefore improving the long-term clinical outcome in these patients. When, according to current standards, treatment in newly diagnosed RA patients is started with csDMARDs in a treat-to-target approach, aiming at remission [3, 4], subclinical inflammation is often seen in those who achieve the treatment target. MRI studies in RA patients clinically in remission on csDMARD therapy demonstrated that approximately one-third of them still showed radiographic progression, of whom 96% had signs of synovitis [5, 6]. In the present study, results (although not statistically significant) suggest a superior joint sparing effect of tocilizumab compared with that of MTX, even when corrected for disease activity, thus correcting for the faster and better clinical effect of tocilizumab, which is in line with the findings described above and with results of previous studies on other biologicals [9–12]. IL-6 is positively correlated with synovitis [16, 18, 19], but also causes deterioration of bone and cartilage via several other mechanisms. First, a higher level of IL-6 increases the expression of VEGF, which as a result of angiogenesis leads to pannus formation and eventually causes cartilage damage and bone erosions [20]. Second, IL-6 stimulates osteoclastogenesis by interaction with RANK-B ligand (RANKL) [17], and also by mechanisms independent of RANK [21], leading to bone destruction. Third, IL-6 simulates secretion of MMPs by activating synoviocytes and chondrocytes, which causes cartilage degradation [17]. Thus, via these mechanisms, the superior joint sparing effect of tocilizumab compared with that of MTX, even if corrected for disease activity, can be explained. In the present study, when evaluating the two subscores of the SHS separately, we found no significant differences in the progression of JSN, which we confirmed in the hand with a computerized JSW quantification method, and only the changes in erosions over 2 years were significantly different between the tocilizumab arms and the MTX arm. These results are largely in accordance with the findings of the FUNCTION study in which also more progression of erosions and JSN was observed in patients with early RA treated with MTX compared with those treated with tocilizumab plus MTX and tocilizumab. In the tocilizumab plus MTX group, erosions were signficantly less in that study at week 52 with JSN only numerically lower when compared with the MTX arm; these differences were not statistically significant after 2 years [22, 23]. As both the U-Act-Early and the FUNCTION study were not powered to analyse radiographic changes, which frequently are minimal especially in early RA patients treated to target, P-values should be interpreted with caution as lack of significance could well be due to insufficient statistical power (i.e. type II error). Nevertheless, also in established RA more joint sparing effect of tocilizumab has been demonstrated compared with that of MTX therapy supporting the effectiveness of this therapy in various patient populations [24]. As in our study erosions contributed most to the progression of joint damage measured by the SHS, we further analysed progression of erosive damage in the limbs separately and found significant differences only in the feet when comparing the tocilizumab arms with the MTX arm. Although this is the first tocilizumab trial evaluating limb specific structural joint changes, it has been shown previously that in RA patients treated with csDMARD, the long-term progression of erosions is higher in feet compared with hands or wrists [25]. An explanation of more rapid progression of erosions in the lower limb might be the different scoring ranges of the hands (range 0–5) and feet (range 0–10), that is, changes in individual joints of the feet have a greater impact on progression rates. However, the maximum theoretical score of feet is less than that of hands and wrists. Another explanation could be that joints in the feet are exposed to high weight-bearing pressures, while standing and walking, and are therefore more prone to develop structural changes when inflammation is present. In conclusion, this study demonstrates that initiating a treat-to-target tocilizumab therapy, with or without MTX, in DMARD-naïve patients with early RA is more effective in preventing progression of erosions compared with a step-up MTX strategy. Especially in the feet we found less erosion in the tocilizumab-based strategies showing that by rapidly reducing inflammation, the functional ability can be preserved in newly diagnosed patients. Acknowledgements In particular we thank Maal van Everdingen for scoring the radiographs. Furthermore, we gratefully thank all patients for their participation, and also those who were involved in conducting this study and all other personnel for their contribution, especially the rheumatologists, study-nurses and laboratory employees of the participating institutions: Amphia Ziekenhuis Breda, Breda, Netherlands; Antonius Ziekenhuis Sneek, Sneek, Netherlands; Atrium Medisch Centrum, Heerlen, Netherlands; Gemini Ziekenhuis, Den Helder, Netherlands; Gelre Ziekenhuis, Apeldoorn, Netherlands; Groene Hart Ziekenhuis, Gouda, Netherlands; Kennemer Gasthuis, Haarlem, Netherlands; Leids Universitair Medisch Centrum, Leiden, Netherlands; Meander Medisch Centrum, Amersfoort, Netherlands; Medisch Centrum Alkmaar, Alkmaar, Netherlands; Medisch Centrum Leeuwarden, Leeuwarden, Netherlands; Medisch Centrum Zuiderzee, Lelystad, Netherlands; Reade, Amsterdam, Netherlands; Rivas Zorggroep, Gorinchem, Netherlands; St. Antonius Ziekenhuis Nieuwegein, Nieuwegein, Netherlands; St Maartenskliniek Woerden, Woerden, Netherlands; St Maartenskliniek Nijmegen, Nijmegen, Netherlands; Spaarne Gasthuis, Hoofddorp, Netherlands; Tergooi Ziekenhuis, Hilversum, Netherlands; Universitair Medisch Centrum Groningen, Groningen, Netherlands; Universitair Medisch Centrum Utrecht, Utrecht, Netherlands. The department of the authors who included patients (J.W.G.J. and J.W.J.B.) in the UAct-Early trial received reimbursements from Roche Nederland BV. The U-Act-Early trial was funded by Roche Nederland BV. Funding: No specific funding was received from any bodies in the public, commercial or not-for-profit sectors to carry out the work described in this manuscript. Disclosure statement: A.P.-S. is an employee of F. Hoffmann-La Roche Ltd. M.E.A.B. is an employee of Roche Nederland BV. J.W.J.B. reported grants and fees from Roche, AbbVie, Bristol-Myers Squibb, Merck Sharp & Dohme, Pfizer, and UCB. F.P.J.G.L has received unrestricted grant support from Roche. J.M.v.L. has received honoraria from Arthrogen, BMS, Eli Lilly, Janssen, MSD, Pfizer, Roche, and research grants from Astra Zeneca, MSD, and Roche-Genentech. All other authors have declared no conflicts of interest. References 1 Carpenter L, Nikiphorou E, Sharpe R et al.   Have radiographic progression rates in early rheumatoid arthritis changed? A systematic review and meta-analysis of long-term cohorts. Rheumatology  2016; 55: 1053– 65. Google Scholar CrossRef Search ADS   2 Balanescu A, Wiland P. Maximizing early treatment with biologics in patients with rheumatoid arthritis: the ultimate breakthrough in joints preservation. Rheumatol Int  2013; 33: 1379– 86. Google Scholar CrossRef Search ADS PubMed  3 Smolen JS, Landewe R, Bijlsma J et al.   EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2016 update. Ann Rheum Dis  2017; 76: 960– 77. Google Scholar CrossRef Search ADS PubMed  4 Singh JA, Saag KG, Bridges SLJr et al.   2015 American College of Rheumatology Guideline for the Treatment of Rheumatoid Arthritis. Arthritis Rheumatol  2016; 68: 1– 26. Google Scholar PubMed  5 Aletaha D, Smolen JS. Joint damage in rheumatoid arthritis progresses in remission according to the Disease Activity Score in 28 joints and is driven by residual swollen joints. Arthritis Rheum  2011; 63: 3702– 11. Google Scholar CrossRef Search ADS PubMed  6 Gandjbakhch F, Conaghan PG, Ejbjerg B et al.   Synovitis and osteitis are very frequent in rheumatoid arthritis clinical remission: results from an MRI study of 294 patients in clinical remission or low disease activity state. J Rheumatol  2011; 38: 2039– 44. Google Scholar CrossRef Search ADS PubMed  7 van der Heijde D. How to read radiographs according to the Sharp/van der Heijde method. J Rheumatol  2000; 27: 261– 3. Google Scholar PubMed  8 Bijlsma JW, Welsing PM, Woodworth TG et al.   Early rheumatoid arthritis treated with tocilizumab, methotrexate, or their combination (U-Act-Early): a multicentre, randomised, double-blind, double-dummy, strategy trial. Lancet  2016; 388: 343– 55. Google Scholar CrossRef Search ADS PubMed  9 Klarenbeek NB, Guler-Yuksel M, van der Heijde DM et al.   Clinical synovitis in a particular joint is associated with progression of erosions and joint space narrowing in that same joint, but not in patients initially treated with infliximab. Ann Rheum Dis  2010; 69: 2107– 13. Google Scholar CrossRef Search ADS PubMed  10 Landewe R, van der Heijde D, Klareskog L, van Vollenhoven R, Fatenejad S. Disconnect between inflammation and joint destruction after treatment with etanercept plus methotrexate: results from the trial of etanercept and methotrexate with radiographic and patient outcomes. Arthritis Rheum  2006; 54: 3119– 25. Google Scholar CrossRef Search ADS PubMed  11 Smolen JS, Han C, Bala M et al.   Evidence of radiographic benefit of treatment with infliximab plus methotrexate in rheumatoid arthritis patients who had no clinical improvement: a detailed subanalysis of data from the anti-tumor necrosis factor trial in rheumatoid arthritis with concomitant therapy study. Arthritis Rheum  2005; 52: 1020– 30. Google Scholar CrossRef Search ADS PubMed  12 Saleem B, Brown AK, Keen H et al.   Disease remission state in patients treated with the combination of tumor necrosis factor blockade and methotrexate or with disease-modifying antirheumatic drugs: A clinical and imaging comparative study. Arthritis Rheum  2009; 60: 1915– 22. Google Scholar CrossRef Search ADS PubMed  13 Huo Y, Vincken KL, van der Heijde D et al.   Automatic quantification of radiographic finger joint space width of patients with early rheumatoid arthritis. IEEE Trans Biomed Eng  2016; 63: 2177– 86. Google Scholar CrossRef Search ADS PubMed  14 Huo Y, Vincken KI, van der Heijde D et al.   Automatic quantification of radiographic wrist joint space width of patients with rheumatoid arthritis. IEEE Trans Biomed Eng  2017; Advance Access published 26 January 2017, doi: 10.1109/TBME.2017.2659223. 15 Huo Y, Veldhuizen RD, van der Heijde D et al.   Automated joint space width quantification of hand and wrist joints: a proof of concept study. Clin Exp Rheumatol  2016; 34(Suppl. 101): S34– S9. 16 Dayer JM, Choy E. Therapeutic targets in rheumatoid arthritis: the interleukin-6 receptor. Rheumatology  2010; 49: 15– 24. Google Scholar CrossRef Search ADS PubMed  17 Choy E. Understanding the dynamics: pathways involved in the pathogenesis of rheumatoid arthritis. Rheumatology  2012; 51(Suppl 5): v3– 11. Google Scholar CrossRef Search ADS PubMed  18 Madhok R, Crilly A, Watson J, Capell HA. Serum interleukin 6 levels in rheumatoid arthritis: correlations with clinical and laboratory indices of disease activity. Ann Rheum Dis  1993; 52: 232– 4. Google Scholar CrossRef Search ADS PubMed  19 Sack U, Kinne RW, Marx T et al.   Interleukin-6 in synovial fluid is closely associated with chronic synovitis in rheumatoid arthritis. Rheumatol Int  1993; 13: 45– 51. Google Scholar CrossRef Search ADS PubMed  20 Ballara S, Taylor PC, Reusch P et al.   Raised serum vascular endothelial growth factor levels are associated with destructive change in inflammatory arthritis. Arthritis Rheum  2001; 44: 2055– 64. Google Scholar CrossRef Search ADS PubMed  21 O'Brien W, Fissel BM, Maeda Y et al.   RANK-independent osteoclast formation and bone erosion in inflammatory arthritis. Arthritis Rheumatol  2016; 68: 2889– 900. Google Scholar CrossRef Search ADS PubMed  22 Burmester GR, Rigby WF, van Vollenhoven RF et al.   Tocilizumab in early progressive rheumatoid arthritis: FUNCTION, a randomised controlled trial. Ann Rheum Dis  2016; 75: 1081– 91. Google Scholar CrossRef Search ADS PubMed  23 Burmester GR, Rigby WF, van Vollenhoven RF et al.   Tocilizumab combination therapy or monotherapy or methotrexate monotherapy in methotrexate-naive patients with early rheumatoid arthritis: 2-year clinical and radiographic results from the randomised, placebo-controlled FUNCTION trial. Ann Rheum Dis  2017; 76: 1279– 84. Google Scholar CrossRef Search ADS PubMed  24 Smolen JS, Avila JC, Aletaha D. Tocilizumab inhibits progression of joint damage in rheumatoid arthritis irrespective of its anti-inflammatory effects: disassociation of the link between inflammation and destruction. Ann Rheum Dis  2012; 71: 687– 93. Google Scholar CrossRef Search ADS PubMed  25 Hulsmans HM, Jacobs JW, van der Heijde DM et al.   The course of radiologic damage during the first six years of rheumatoid arthritis. Arthritis Rheum  2000; 43: 1927– 40. Google Scholar CrossRef Search ADS PubMed  © The Author 2017. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oup.com

Journal

RheumatologyOxford University Press

Published: Feb 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off